{ "schemaVersion": "1.1.0", "@context": "https://schema.org", "@type": "DataCatalog", "@id": "https://course.dragonai.tech/datasets/index.json#catalog", "name": "DragonAI public research datasets", "description": "Machine-readable index of DragonAI public research datasets, stable download URLs, licenses, and related distributions.", "url": "https://course.dragonai.tech/datasets/index.json", "publisher": { "@type": "Organization", "@id": "https://dragonai.tech/#organization", "name": "DragonAI", "url": "https://dragonai.tech/" }, "dataset": [ { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-15-r2.json", "identifier": "ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-15-r2", "name": "RAG 与微调引用来源：当前百度—Bing 精确 URL 与主机对照 r2", "description": "以三条冻结无品牌查询对照最新百度与 Bing 中国匿名自然结果：百度 25 个位次、22 个不同 URL，Bing 30 个位次、28 个不同 URL；跨查询精确 URL 交集为 1。DragonAI 目标页只在 Bing 的一条查询位于第 1，在百度为 0，跨引擎目标命中为 0；这是搜索可见性，不是 AI 引用、品牌提及或跨引擎 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-15-r2.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "RAG 与微调引用来源：当前百度—Bing 精确 URL 与主机对照 r2", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-15-r2.json" }, { "@type": "DataDownload", "name": "当前百度—Bing 精确 URL 对照 r2（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-15-r2.csv" }, { "@type": "DataDownload", "name": "当前百度—Bing 同查询精确主机对照 r2（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-15-r2-hosts.csv" }, { "@type": "DataDownload", "name": "当前百度—Bing 精确 URL 对照 r2（Markdown）", "encodingFormat": "text/markdown", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-15-r2.md" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-14.json", "identifier": "ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-14", "name": "RAG and fine-tuning citation sources: current Baidu–Bing exact-URL overlap", "description": "A deterministic same-query comparison of 25 Baidu and 30 Bing China anonymous organic result positions for three frozen unbranded queries, covering 22 and 26 distinct URLs. The cross-query union is 47 URLs and one exact URL overlaps within the same query; the target and brand domain are absent on both surfaces. This point-in-time snapshot is not visibility or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-14.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "RAG and fine-tuning citation sources: current Baidu–Bing exact-URL overlap", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-14.json" }, { "@type": "DataDownload", "name": "Current Baidu–Bing exact-URL overlap CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-14.csv" }, { "@type": "DataDownload", "name": "Current Baidu–Bing same-query exact-host overlap CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-14-hosts.csv" }, { "@type": "DataDownload", "name": "Current Baidu–Bing exact-URL overlap Markdown", "encodingFormat": "text/markdown", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-14.md" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-forty-six-observation-host-priority-2026-08-14.json", "identifier": "ai-search-rag-baidu-forty-six-observation-host-priority-2026-08-14", "name": "RAG and fine-tuning citation sources: 46-observation Baidu exact-host priority matrix", "description": "Exact-host aggregation for three frozen unbranded Baidu queries across 46 comparable observations, exposing result occurrences, top-three slot occupancy, top-three observation coverage, position ranges, and latest URLs. The matrix prioritizes evidence review and distribution verification without treating frequency as source quality, ranking causality, visibility, or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-forty-six-observation-host-priority-2026-08-14.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "RAG and fine-tuning citation sources: 46-observation Baidu exact-host priority matrix", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-forty-six-observation-host-priority-2026-08-14.json" }, { "@type": "DataDownload", "name": "46-observation Baidu exact-host priority matrix CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-forty-six-observation-host-priority-2026-08-14.csv" }, { "@type": "DataDownload", "name": "46-observation Baidu exact-host priority matrix Markdown", "encodingFormat": "text/markdown", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-forty-six-observation-host-priority-2026-08-14.md" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-forty-six-observation-query-evidence-2026-08-14.json", "identifier": "ai-search-rag-baidu-forty-six-observation-query-evidence-2026-08-14", "name": "RAG and fine-tuning citation sources: 46-observation Baidu query evidence cards", "description": "Direct-answer evidence cards for three frozen unbranded Baidu queries across 46 comparable observations and 31 production releases, with URL position sequences and latest top-three results. DragonAI target and brand-domain hits remain zero; this is not GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-forty-six-observation-query-evidence-2026-08-14.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "RAG and fine-tuning citation sources: 46-observation Baidu query evidence cards", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-forty-six-observation-query-evidence-2026-08-14.json" }, { "@type": "DataDownload", "name": "46-observation Baidu query evidence cards CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-forty-six-observation-query-evidence-2026-08-14.csv" }, { "@type": "DataDownload", "name": "46-observation Baidu query evidence cards Markdown", "encodingFormat": "text/markdown", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-forty-six-observation-query-evidence-2026-08-14.md" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-research-authority-graph-v1.json", "identifier": "ai-search-rag-research-authority-graph-v1", "name": "RAG and fine-tuning citation research authority graph", "description": "A deterministic graph of the canonical DragonAI research target, 19 query-aligned contextual internal references, and one verified independent external repository reference. The external reference is not classified as earned editorial because the required human receipt is incomplete; every edge remains explicitly false for search visibility, AI citation, ranking improvement, and GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-research-authority-graph-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "RAG and fine-tuning citation research authority graph", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-research-authority-graph-v1.json" }, { "@type": "DataDownload", "name": "Research authority graph edge list CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-research-authority-graph-v1.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-release-observation-panel-2026-08-14.json", "identifier": "ai-search-rag-baidu-release-observation-panel-2026-08-14", "name": "Baidu strict unbranded release-aligned observation panel", "description": "Release-level aggregation of 46 comparable anonymous Baidu observations across 31 production releases, 138 strict-query snapshots and 1,137 organic positions. It separates release count from observation count and preserves zero target and brand-domain observations.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-release-observation-panel-2026-08-14.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Baidu strict unbranded release-aligned observation panel", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-release-observation-panel-2026-08-14.json" }, { "@type": "DataDownload", "name": "Baidu release-aligned observation panel CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-release-observation-panel-2026-08-14.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-overlap-2026-08-11.json", "identifier": "ai-search-rag-current-baidu-ai-overlap-2026-08-11", "name": "Current Baidu and AI citation source overlap for RAG and fine-tuning", "description": "A deterministic 70-row full outer join of the current 25-position, 22-URL Baidu snapshot and the current nine-run Qwen, Kimi and Gemini citation batch. Three normalized hosts and two exact URLs overlap; DragonAI is absent from both surfaces, so this is not shared-index evidence, target visibility or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-overlap-2026-08-11.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Current Baidu and AI citation source overlap for RAG and fine-tuning", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-overlap-2026-08-11.json" }, { "@type": "DataDownload", "name": "Current Baidu and AI citation hostname overlap CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-overlap-2026-08-11.csv" }, { "@type": "DataDownload", "name": "Current Baidu and AI citation exact URL full outer join CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-overlap-2026-08-11-urls.csv" }, { "@type": "DataDownload", "name": "Current Baidu and AI citation URL overlap summary figure", "encodingFormat": "image/svg+xml", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-overlap-2026-08-11.svg" }, { "@type": "DataDownload", "name": "Reproducible current Baidu and AI citation overlap deposit supplement", "encodingFormat": "application/zip", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-overlap-supplement-2026-08-11.zip" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-strict-snapshot-2026-08-11.json", "identifier": "ai-search-rag-baidu-current-strict-snapshot-2026-08-11", "name": "Baidu current strict unbranded RAG and fine-tuning snapshot", "description": "Current anonymous read-only results for the three frozen unbranded Baidu queries at 2026-08-14 15:04 UTC: 25 organic positions across 22 distinct URLs. The machine-readable source-group distribution is CSDN 4, Baijiahao 5, Tencent Cloud 4, Baidu AI Cloud 3, Zhihu 2, and other sources 7; DragonAI target-page and brand-domain appearances remain zero. Relative to the 12:53 UTC same-release observation, 24 query-level URLs were retained without rank movement and the BAAI Hub page entered the provider-source query at position 10. The grouping and adjacent change are descriptive, not source-quality rankings, ranking causality, AI citation evidence, or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-strict-snapshot-2026-08-11.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Baidu current strict unbranded RAG and fine-tuning snapshot", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-strict-snapshot-2026-08-11.json" }, { "@type": "DataDownload", "name": "Current strict Baidu snapshot CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-strict-snapshot-2026-08-11.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-14-150436.json", "identifier": "ai-search-rag-baidu-serp-entry-liveness-2026-08-14-150436", "name": "Baidu strict provider-source SERP entry liveness at 2026-08-14 15:04 UTC", "description": "Read-only liveness evidence for the one URL newly entering the fixed unbranded provider-source query at position 10: the BAAI Hub page returned live canonical HTML. The previous nine URLs retained their positions, while DragonAI target-page and brand-domain appearances remained zero across the strict suite. Entry and liveness do not establish source quality, ranking causality, target visibility, AI citation, or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-14-150436.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Baidu strict provider-source SERP entry liveness at 2026-08-14 15:04 UTC", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-14-150436.json" }, { "@type": "DataDownload", "name": "Baidu strict provider-source SERP entry liveness CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-14-150436.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-13t175038z.json", "identifier": "ai-search-rag-baidu-serp-entry-liveness-2026-08-13t175038z", "name": "Baidu strict SERP entry liveness at 2026-08-13 17:50 UTC", "description": "Read-only liveness evidence for the three URLs that entered the current adjacent strict Baidu snapshot: Red Hat and a Baijiahao safety-verification page returned HTML, while the CSDN request ended in a connection error. DragonAI target-page and brand-domain appearances remained zero. Entry and liveness do not establish source quality, ranking causality, target visibility, AI citation, or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-13t175038z.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Baidu strict SERP entry liveness at 2026-08-13 17:50 UTC", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-13t175038z.json" }, { "@type": "DataDownload", "name": "Baidu strict SERP entry liveness CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-13t175038z.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-13t170117z.json", "identifier": "ai-search-rag-baidu-serp-entry-liveness-2026-08-13t170117z", "name": "Baidu strict SERP entry liveness at 2026-08-13 17:01 UTC", "description": "Read-only liveness evidence for the one URL that entered the current adjacent strict Baidu snapshot: the BAAI Hub page returned live HTML, while DragonAI target-page and brand-domain appearances remained zero. Entry and liveness do not establish source quality, ranking causality, target visibility, AI citation, or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-13t170117z.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Baidu strict SERP entry liveness at 2026-08-13 17:01 UTC", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-13t170117z.json" }, { "@type": "DataDownload", "name": "Baidu strict SERP entry liveness CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-13t170117z.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-current-batch-2026-08-11.json", "identifier": "ai-search-citation-benchmark-current-batch-2026-08-11", "name": "千问、Kimi、Gemini 固定问题当前 9 次引用复测（2026-08-11）", "description": "公开同一固定无品牌问题的三平台各 3 次逐 URL 结果、零引用运行、模型版本与严格趋势边界；DragonAI 目标引用和品牌提及均为零。可下载的脱敏原始响应保留回答文本、搜索或 grounding 元数据、引用表面与用量字段，仅对提供商请求/响应标识符作带 SHA-256 承诺的脱敏；第三方内容不在 DragonAI 的 CC BY 4.0 授权范围内。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-current-batch-2026-08-11.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问、Kimi、Gemini 固定问题当前 9 次引用复测（2026-08-11）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-current-batch-2026-08-11.json" }, { "@type": "DataDownload", "name": "当前 9 次复测逐引用 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-current-batch-2026-08-11.csv" }, { "@type": "DataDownload", "name": "当前 9 次复测脱敏原始响应与复现代码 ZIP（版本 1）", "encodingFormat": "application/zip", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-benchmark-redacted-raw-evidence-2026-08-11-v1.zip" }, { "@type": "DataDownload", "name": "当前百度与三平台引用审计存档候选 ZIP（版本 1）", "encodingFormat": "application/zip", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-audit-deposit-2026-08-11-v1.zip" }, { "@type": "DataDownload", "name": "当前研究证据上下文交付包 ZIP（版本 9）", "encodingFormat": "application/zip", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-context-deposit-2026-08-14-v9.zip" }, { "@type": "DataDownload", "name": "当前研究证据上下文交付包 ZIP（版本 8）", "encodingFormat": "application/zip", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-context-deposit-2026-08-14-v8.zip" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-current-batch-2026-08-14.json", "identifier": "ai-search-citation-benchmark-current-batch-2026-08-14", "name": "千问、Kimi、Gemini 固定问题当前 9 次引用复测（2026-08-14）", "description": "同一固定无品牌问题的三平台各 3 次逐 URL 观察：共 91 次具体引用 URL 出现、50 个不同 URL、37 个不同主机；Kimi 三次均未展示具体 URL，DragonAI 目标引用和品牌提及均为 0/9。该数据只描述点时引用表面，不代表来源质量、索引、排名因果或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-current-batch-2026-08-14.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-search-citation-monitoring-method#article", "url": "https://course.dragonai.tech/guides/ai-search-citation-monitoring-method" }, { "@type": "Article", "@id": "https://course.dragonai.tech/guides/how-ai-search-chooses-citation-sources#article", "url": "https://course.dragonai.tech/guides/how-ai-search-chooses-citation-sources" }, { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问、Kimi、Gemini 固定问题当前 9 次引用复测（2026-08-14）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-current-batch-2026-08-14.json" }, { "@type": "DataDownload", "name": "2026-08-14 当前 9 次复测逐引用 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-current-batch-2026-08-14.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-live-churn-2026-08-02.json", "identifier": "ai-search-rag-baidu-selection-query-live-churn-2026-08-02", "name": "Baidu live churn for the RAG and fine-tuning source-selection query", "description": "A bounded, read-only comparison between the 26th strict observation and one later anonymous Baidu result page for the fixed unbranded source-selection query. Six of eight URLs persisted, two entered, two exited, three persistent URLs changed rank, and the top three stayed unchanged. This partial single-query capture is not a 27th strict observation, AI citation evidence, or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-live-churn-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Baidu live churn for the RAG and fine-tuning source-selection query", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-live-churn-2026-08-02.json" }, { "@type": "DataDownload", "name": "Baidu source-selection live churn CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-live-churn-2026-08-02.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-six-observation-query-evidence-2026-08-02.json", "identifier": "ai-search-rag-baidu-twenty-six-observation-query-evidence-2026-08-02", "name": "RAG and fine-tuning citation sources: 26-observation Baidu query evidence cards", "description": "Direct-answer evidence cards for three frozen unbranded Baidu queries across 26 comparable observations, with URL position sequences, persistent sources and latest top-three results. DragonAI target and brand-domain hits remain zero; this is not GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-six-observation-query-evidence-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/how-ai-search-chooses-citation-sources#article", "url": "https://course.dragonai.tech/guides/how-ai-search-chooses-citation-sources" }, { "@type": "Article", "@id": "https://course.dragonai.tech/guides/how-to-get-site-into-ai-search#article", "url": "https://course.dragonai.tech/guides/how-to-get-site-into-ai-search" }, { "@type": "Article", "@id": "https://course.dragonai.tech/guides/what-is-generative-engine-optimization#article", "url": "https://course.dragonai.tech/guides/what-is-generative-engine-optimization" }, { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "RAG and fine-tuning citation sources: 26-observation Baidu query evidence cards", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-six-observation-query-evidence-2026-08-02.json" }, { "@type": "DataDownload", "name": "26-observation Baidu query evidence cards CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-six-observation-query-evidence-2026-08-02.csv" }, { "@type": "DataDownload", "name": "26-observation Baidu query evidence cards Markdown", "encodingFormat": "text/markdown", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-six-observation-query-evidence-2026-08-02.md" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r8.json", "identifier": "ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r8", "name": "Baidu strict unbranded comparable-observation ledger (26 observations, revision 8)", "description": "Immutable cumulative ledger for 26 strict Baidu observations across 15 production releases, 78 query snapshots and 649 organic result positions, with source-report paths and hashes. Target and brand-domain hits remain zero.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r8.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Baidu strict unbranded comparable-observation ledger (26 observations, revision 8)", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r8.json" }, { "@type": "DataDownload", "name": "Comparable-observation ledger revision 8 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r8.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-12-r9.json", "identifier": "ai-search-rag-baidu-comparable-observation-ledger-2026-08-12-r9", "name": "Baidu strict unbranded comparable-observation ledger (33 observations, revision 9)", "description": "Immutable SHA-256-bound ledger for 33 comparable anonymous Baidu observations across 22 production releases, 99 query snapshots and 819 organic result positions. Target and brand-domain hits remain zero.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-12-r9.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Baidu strict unbranded comparable-observation ledger (33 observations, revision 9)", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-12-r9.json" }, { "@type": "DataDownload", "name": "Comparable-observation ledger revision 9 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-12-r9.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-14-r10.json", "identifier": "ai-search-rag-baidu-comparable-observation-ledger-2026-08-14-r10", "name": "Baidu strict unbranded comparable-observation ledger (46 observations, revision 10)", "description": "Immutable SHA-256-bound ledger for 46 comparable anonymous Baidu observations across 31 production releases, 138 query snapshots and 1,137 organic result positions. Target and brand-domain hits remain zero.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-14-r10.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Baidu strict unbranded comparable-observation ledger (46 observations, revision 10)", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-14-r10.json" }, { "@type": "DataDownload", "name": "Comparable-observation ledger revision 10 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-14-r10.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-six-observation-persistence-2026-08-02.json", "identifier": "ai-search-rag-baidu-serp-twenty-six-observation-persistence-2026-08-02", "name": "Baidu strict unbranded query-URL persistence across 26 observations", "description": "Query-URL persistence across 26 observations, 15 releases, 78 snapshots and 649 result positions. Eight of 51 query-URL pairs span all observations; none keeps an exact position throughout. Target and brand-domain hits remain zero.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-six-observation-persistence-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Baidu strict unbranded query-URL persistence across 26 observations", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-six-observation-persistence-2026-08-02.json" }, { "@type": "DataDownload", "name": "26-observation query-URL persistence CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-six-observation-persistence-2026-08-02.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-six-observation-2026-08-02.json", "identifier": "ai-search-rag-baidu-query-host-persistence-twenty-six-observation-2026-08-02", "name": "Baidu strict unbranded query-host persistence across 26 observations", "description": "Query-host persistence across 26 observations. Ten of 28 query-host pairs span the full window, one keeps its exact position, and five stay in the top three. DragonAI target and brand domains remain absent.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-six-observation-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Baidu strict unbranded query-host persistence across 26 observations", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-six-observation-2026-08-02.json" }, { "@type": "DataDownload", "name": "26-observation query-host persistence CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-six-observation-2026-08-02.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-six-observation-2026-08-02.json", "identifier": "ai-search-rag-baidu-persistent-top3-source-cards-twenty-six-observation-2026-08-02", "name": "Baidu persistent top-three source cards across 26 strict observations", "description": "Joins five query-host pairs that stayed in the top three throughout 26 observations to four hosts and 12 concrete result URLs. Target and brand-domain hits remain zero; persistence is not a quality or GEO-success claim.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-six-observation-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Baidu persistent top-three source cards across 26 strict observations", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-six-observation-2026-08-02.json" }, { "@type": "DataDownload", "name": "26-observation persistent top-three source cards CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-six-observation-2026-08-02.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-sixth-observation-churn-2026-08-02.json", "identifier": "ai-search-rag-baidu-serp-twenty-sixth-observation-churn-2026-08-02", "name": "Baidu strict-query adjacent churn from observation 25 to 26", "description": "Adjacent frozen SERP delta for three strict unbranded queries. Two queries are unchanged; the second retains five URLs, adds two, removes two and moves two. DragonAI target and brand-domain hits remain zero.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-sixth-observation-churn-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Baidu strict-query adjacent churn from observation 25 to 26", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-sixth-observation-churn-2026-08-02.json" }, { "@type": "DataDownload", "name": "Observation 25 to 26 strict Baidu churn CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-sixth-observation-churn-2026-08-02.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-five-observation-query-evidence-2026-08-02.json", "identifier": "ai-search-rag-baidu-twenty-five-observation-query-evidence-2026-08-02", "name": "RAG 与微调引用来源：25 次百度观察逐查询证据卡", "description": "把三条严格无品牌百度查询的 25 次冻结观察压缩为可摘录答案卡，同时公开逐 URL 位置序列、全窗口持续 URL、持续前三主机和最新前三结果。目标页与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-five-observation-query-evidence-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "RAG 与微调引用来源：25 次百度观察逐查询证据卡", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-five-observation-query-evidence-2026-08-02.json" }, { "@type": "DataDownload", "name": "25 次百度观察逐查询证据卡 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-five-observation-query-evidence-2026-08-02.csv" }, { "@type": "DataDownload", "name": "25 次百度观察逐查询证据卡 Markdown", "encodingFormat": "text/markdown", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-five-observation-query-evidence-2026-08-02.md" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r7.json", "identifier": "ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r7", "name": "百度严格无品牌搜索可比观察累计账本（25 次，修订 7）", "description": "25 次严格百度观察的累计账本：14 个生产发布、75 个查询快照、624 个自然结果位置；每次观察绑定源报告路径与 SHA-256。目标页与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r7.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度严格无品牌搜索可比观察累计账本（25 次，修订 7）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r7.json" }, { "@type": "DataDownload", "name": "百度严格无品牌搜索可比观察累计账本修订 7 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r7.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-five-observation-persistence-2026-08-02.json", "identifier": "ai-search-rag-baidu-serp-twenty-five-observation-persistence-2026-08-02", "name": "百度 RAG 与微调引用来源严格无品牌 25 次查询—URL 持续性（2026-08-02）", "description": "跨 14 个生产版本汇总 25 次严格无品牌百度观察、75 个查询快照和 624 个自然结果位置；50 个查询—URL 组合中有 8 个覆盖全部 25 次，没有 URL 在全窗口保持精确位置不变。DragonAI 目标页与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-five-observation-persistence-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌 25 次查询—URL 持续性（2026-08-02）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-five-observation-persistence-2026-08-02.json" }, { "@type": "DataDownload", "name": "百度严格无品牌 25 次查询—URL 持续性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-five-observation-persistence-2026-08-02.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-five-observation-2026-08-02.json", "identifier": "ai-search-rag-baidu-query-host-persistence-twenty-five-observation-2026-08-02", "name": "百度 RAG 与微调引用来源严格无品牌 25 次查询×主机持久性（2026-08-02）", "description": "将 25 次严格无品牌百度观察按查询与精确主机聚合。28 个查询×主机组合中 10 个覆盖全部 25 次，只有 1 个在全窗口保持精确位置不变，5 个始终位于前 3；DragonAI 目标主机与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-five-observation-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌 25 次查询×主机持久性（2026-08-02）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-five-observation-2026-08-02.json" }, { "@type": "DataDownload", "name": "百度严格无品牌 25 次查询×主机持久性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-five-observation-2026-08-02.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-five-observation-2026-08-02.json", "identifier": "ai-search-rag-baidu-persistent-top3-source-cards-twenty-five-observation-2026-08-02", "name": "百度 RAG 与微调引用来源严格无品牌 25 次观察持续前三来源卡（2026-08-02）", "description": "将 25 次严格无品牌百度观察中每次都进入前三位的 5 个查询—主机组合，与 12 个具体结果 URL、最新标题及逐文档覆盖联表。共覆盖 4 个不同主机；DragonAI 目标主机与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-five-observation-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌 25 次观察持续前三来源卡（2026-08-02）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-five-observation-2026-08-02.json" }, { "@type": "DataDownload", "name": "百度严格无品牌 25 次观察持续前三来源卡 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-five-observation-2026-08-02.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-fifth-observation-churn-2026-08-02.json", "identifier": "ai-search-rag-baidu-serp-twenty-fifth-observation-churn-2026-08-02", "name": "百度 RAG 与微调引用来源第 24 至 25 次严格观察相邻波动", "description": "固定三条严格无品牌百度查询的第 24 至 25 次相邻观察：第二、第三条查询的 15 个自然结果位次完全不变；第一条查询保留 9 个 URL，新增与移除各 1 个，并有 1 个保留 URL 前移一位。新增与移除记录是同一篇 CSDN 文章的不同规范主机路径；DragonAI 目标页和品牌域仍均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-fifth-observation-churn-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源第 24 至 25 次严格观察相邻波动", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-fifth-observation-churn-2026-08-02.json" }, { "@type": "DataDownload", "name": "第 24 至 25 次严格百度观察相邻波动 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-fifth-observation-churn-2026-08-02.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-four-observation-persistence-2026-08-02.json", "identifier": "ai-search-rag-baidu-serp-twenty-four-observation-persistence-2026-08-02", "name": "百度 RAG 与微调引用来源严格无品牌 24 次查询—URL 持续性（2026-08-02）", "description": "跨 13 个生产版本汇总 24 次严格无品牌百度观察、72 个查询快照和 599 个自然结果位置；49 个查询—URL 组合中有 9 个覆盖全部 24 次，只有 1 个在全窗口保持精确位置不变。DragonAI 目标页与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-four-observation-persistence-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌 24 次查询—URL 持续性（2026-08-02）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-four-observation-persistence-2026-08-02.json" }, { "@type": "DataDownload", "name": "百度严格无品牌 24 次查询—URL 持续性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-four-observation-persistence-2026-08-02.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-four-observation-2026-08-02.json", "identifier": "ai-search-rag-baidu-query-host-persistence-twenty-four-observation-2026-08-02", "name": "百度 RAG 与微调引用来源严格无品牌 24 次查询×主机持久性（2026-08-02）", "description": "将 24 次严格无品牌百度观察按查询与精确主机聚合。28 个查询×主机组合中 11 个覆盖全部 24 次，只有 2 个在全窗口保持精确位置不变，6 个始终位于前 3；DragonAI 目标主机与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-four-observation-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌 24 次查询×主机持久性（2026-08-02）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-four-observation-2026-08-02.json" }, { "@type": "DataDownload", "name": "百度严格无品牌 24 次查询×主机持久性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-four-observation-2026-08-02.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-four-observation-2026-08-02.json", "identifier": "ai-search-rag-baidu-persistent-top3-source-cards-twenty-four-observation-2026-08-02", "name": "百度 RAG 与微调引用来源严格无品牌 24 次观察持续前三来源卡（2026-08-02）", "description": "将 24 次严格无品牌百度观察中每次都进入前三位的 6 个查询—主机组合，与 13 个具体结果 URL、最新标题及逐文档覆盖联表。三个固定问法各有两个持续前三主机；DragonAI 目标主机与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-four-observation-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌 24 次观察持续前三来源卡（2026-08-02）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-four-observation-2026-08-02.json" }, { "@type": "DataDownload", "name": "百度严格无品牌 24 次观察持续前三来源卡 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-four-observation-2026-08-02.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-four-observation-query-evidence-2026-08-02.json", "identifier": "ai-search-rag-baidu-twenty-four-observation-query-evidence-2026-08-02", "name": "百度 RAG 与微调引用来源 24 次观察逐查询证据卡", "description": "三条严格无品牌百度查询的 24 次冻结观察证据卡，公开逐 URL 位置序列、全窗口持续 URL、持续前三主机与最新前三结果；目标页和品牌域均为零命中，不构成搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-four-observation-query-evidence-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源 24 次观察逐查询证据卡", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-four-observation-query-evidence-2026-08-02.json" }, { "@type": "DataDownload", "name": "24 次百度观察逐查询证据卡 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-four-observation-query-evidence-2026-08-02.csv" }, { "@type": "DataDownload", "name": "24 次百度观察逐查询证据卡 Markdown", "encodingFormat": "text/markdown", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-four-observation-query-evidence-2026-08-02.md" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r6.json", "identifier": "ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r6", "name": "百度严格无品牌搜索可比观察累计账本（24 次，修订 6）", "description": "24 次严格百度观察的累计账本：13 个生产发布、72 个查询快照、599 个自然结果位置；每次观察绑定源报告路径与 SHA-256。目标页与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r6.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度严格无品牌搜索可比观察累计账本（24 次，修订 6）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r6.json" }, { "@type": "DataDownload", "name": "百度严格无品牌搜索可比观察累计账本修订 6 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r6.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-fourth-observation-churn-2026-08-02.json", "identifier": "ai-search-rag-baidu-serp-twenty-fourth-observation-churn-2026-08-02", "name": "百度 RAG 与微调引用来源第 23 至 24 次严格观察相邻波动", "description": "固定三条严格无品牌百度查询的第 23 至 24 次相邻观察：前两条查询的 17 个自然结果位次完全不变；选型查询保留 7 个 URL，新增与移出各 1 个，5 个保留 URL 前移一位。DragonAI 目标页和品牌域仍均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-fourth-observation-churn-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源第 23 至 24 次严格观察相邻波动", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-fourth-observation-churn-2026-08-02.json" }, { "@type": "DataDownload", "name": "第 23 至 24 次严格百度观察相邻波动 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-fourth-observation-churn-2026-08-02.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-three-observation-query-answer-cards-2026-08-02.json", "identifier": "ai-search-rag-baidu-twenty-three-observation-query-answer-cards-2026-08-02", "name": "百度 RAG 与微调引用来源 23 次观察查询答案卡（2026-08-02）", "description": "三条严格无品牌百度查询的 23 次冻结观察与持续前三来源逐查询答案卡；另保留原始固定问题未进入百度套件的边界。目标页与品牌域仍为零命中，不构成搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-three-observation-query-answer-cards-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源 23 次观察查询答案卡（2026-08-02）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-three-observation-query-answer-cards-2026-08-02.json" }, { "@type": "DataDownload", "name": "23 次百度观察查询答案卡 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-three-observation-query-answer-cards-2026-08-02.csv" }, { "@type": "DataDownload", "name": "23 次百度观察查询答案卡 Markdown", "encodingFormat": "text/markdown", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-three-observation-query-answer-cards-2026-08-02.md" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-three-observation-2026-08-02.json", "identifier": "ai-search-rag-baidu-query-host-persistence-twenty-three-observation-2026-08-02", "name": "百度 RAG 与微调引用来源严格无品牌 23 次查询×主机持久性（2026-08-02）", "description": "将 23 次严格无品牌百度观测按查询与规范化主机聚合。28 个查询×主机组合中 11 个覆盖全部 23 次，只有 2 个在全窗口保持精确位置不变，6 个始终位于前 3；DragonAI 目标主机与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-three-observation-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌 23 次查询×主机持久性（2026-08-02）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-three-observation-2026-08-02.json" }, { "@type": "DataDownload", "name": "百度严格无品牌 23 次查询×主机持久性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-three-observation-2026-08-02.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-three-observation-persistence-2026-08-02.json", "identifier": "ai-search-rag-baidu-serp-twenty-three-observation-persistence-2026-08-02", "name": "百度 RAG 与微调引用来源严格无品牌 23 次查询—URL 持续性（2026-08-02）", "description": "跨 12 个生产版本汇总 23 次严格无品牌百度观察、69 个查询快照和 574 个自然结果位置；48 个查询—URL 组合中有 10 个覆盖全部 23 次，只有 1 个在全窗口保持精确位置不变。DragonAI 目标页与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-three-observation-persistence-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌 23 次查询—URL 持续性（2026-08-02）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-three-observation-persistence-2026-08-02.json" }, { "@type": "DataDownload", "name": "百度严格无品牌 23 次查询—URL 持续性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-three-observation-persistence-2026-08-02.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-three-observation-2026-08-02.json", "identifier": "ai-search-rag-baidu-persistent-top3-source-cards-twenty-three-observation-2026-08-02", "name": "百度 RAG 与微调引用来源严格无品牌 23 次观察持续前三来源卡（2026-08-02）", "description": "将 23 次严格无品牌百度观察中每次都进入前三位的 6 个查询—主机组合，与 12 个具体结果 URL、最新标题及逐文档覆盖确定性联表。三个固定问法各有两个持续前三主机；DragonAI 目标主机与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-three-observation-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌 23 次观察持续前三来源卡（2026-08-02）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-three-observation-2026-08-02.json" }, { "@type": "DataDownload", "name": "百度严格无品牌 23 次观察持续前三来源卡 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-three-observation-2026-08-02.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-2026-08-02.json", "identifier": "ai-search-rag-baidu-persistent-top3-source-cards-2026-08-02", "name": "百度 RAG 与微调引用来源严格无品牌 22 次观察持续前三来源卡（2026-08-02）", "description": "将 22 次严格无品牌百度观察中每次都进入前三位的 6 个查询—主机组合，与 12 个具体结果 URL、最新标题及逐文档覆盖确定性联表。三个固定问法各有两个持续前三主机；DragonAI 目标主机与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌 22 次观察持续前三来源卡（2026-08-02）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-2026-08-02.json" }, { "@type": "DataDownload", "name": "百度严格无品牌 22 次观察持续前三来源卡 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-2026-08-02.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-two-observation-2026-08-02.json", "identifier": "ai-search-rag-baidu-query-host-persistence-twenty-two-observation-2026-08-02", "name": "百度 RAG 与微调引用来源严格无品牌 22 次查询×主机持久性（2026-08-02）", "description": "将 22 次严格无品牌百度观测按查询与规范化主机聚合。28 个查询×主机组合中 11 个覆盖全部 22 次，只有 2 个在全窗口保持精确位置不变，6 个始终位于前 3；DragonAI 目标主机与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-two-observation-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌 22 次查询×主机持久性（2026-08-02）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-two-observation-2026-08-02.json" }, { "@type": "DataDownload", "name": "百度严格无品牌 22 次查询×主机持久性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-two-observation-2026-08-02.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-two-observation-persistence-2026-08-01.json", "identifier": "ai-search-rag-baidu-serp-twenty-two-observation-persistence-2026-08-01", "name": "百度 RAG 与微调引用来源严格无品牌 22 次查询—URL 持续性（2026-08-01）", "description": "跨 11 个生产版本汇总 22 次严格无品牌百度观察、66 个查询快照和 549 个自然结果位置；48 个查询—URL 组合中有 10 个覆盖全部 22 次，只有 1 个在全窗口保持精确位置不变。DragonAI 目标页与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-two-observation-persistence-2026-08-01.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌 22 次查询—URL 持续性（2026-08-01）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-two-observation-persistence-2026-08-01.json" }, { "@type": "DataDownload", "name": "百度严格无品牌 22 次查询—URL 持续性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-two-observation-persistence-2026-08-01.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r5.json", "identifier": "ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r5", "name": "百度严格无品牌搜索可比观察累计账本（23 次，修订 5）", "description": "23 次严格百度观察的累计账本：12 个生产版本、69 个查询快照、574 个自然结果位置；目标页与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r5.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度严格无品牌搜索可比观察累计账本（23 次，修订 5）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r5.json" }, { "@type": "DataDownload", "name": "百度严格无品牌搜索可比观察累计账本修订 5 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r5.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r4.json", "identifier": "ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r4", "name": "百度严格无品牌搜索可比观察累计账本（22 次，修订 4）", "description": "22 次严格百度观察的累计账本：11 个生产版本、66 个查询快照、549 个自然结果位置；目标页与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r4.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度严格无品牌搜索可比观察累计账本（22 次，修订 4）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r4.json" }, { "@type": "DataDownload", "name": "百度严格无品牌搜索可比观察累计账本 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r4.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r3.json", "identifier": "ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r3", "name": "百度严格无品牌搜索可比观察累计账本（21 次，修订 3）", "description": "21 次严格百度观察的累计账本：10 个生产版本、63 个查询快照、524 个自然结果位置；目标页与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r3.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度严格无品牌搜索可比观察累计账本（21 次，修订 3）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r3.json" }, { "@type": "DataDownload", "name": "百度严格无品牌搜索可比观察累计账本修订 3 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r3.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-02-055317.json", "identifier": "ai-search-rag-baidu-serp-entry-liveness-2026-08-02-055317", "name": "百度第 23 次严格观察新增 URL 存活证据", "description": "第 23 次严格百度观察的相邻结果变动与只读存活证据：1 个 URL 新增、1 个移出；新增 URL 返回 HTTP 200，但最终页面为百度安全验证 HTML。该证据不构成搜索可见性、来源质量、正文可读性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-02-055317.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度第 23 次严格观察新增 URL 存活证据", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-02-055317.json" }, { "@type": "DataDownload", "name": "百度第 23 次严格观察新增 URL 存活证据 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-02-055317.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-01-222307.json", "identifier": "ai-search-rag-baidu-serp-entry-liveness-2026-08-01-222307", "name": "百度第 22 次严格观察新增 URL 存活证据", "description": "第 22 次严格百度观察的相邻结果变动与只读存活证据：2 个新增、2 个移出，新增 URL 中 1 个可读取 HTML、1 个返回 HTTP 404；不构成搜索可见性、来源质量或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-01-222307.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度第 22 次严格观察新增 URL 存活证据", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-01-222307.json" }, { "@type": "DataDownload", "name": "百度第 22 次严格观察新增 URL 存活证据 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-01-222307.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-first-observation-entry-liveness-2026-08-01.json", "identifier": "ai-search-rag-baidu-twenty-first-observation-entry-liveness-2026-08-01", "name": "百度第 21 次严格观察新增 URL 存活证据", "description": "第 21 次严格百度观察的结果集变动与只读存活证据：2 个新增、2 个移出，新增 URL 返回 HTTP 521 与 403；不构成可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-first-observation-entry-liveness-2026-08-01.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度第 21 次严格观察新增 URL 存活证据", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-first-observation-entry-liveness-2026-08-01.json" }, { "@type": "DataDownload", "name": "百度第 21 次严格观察新增 URL 存活证据 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-first-observation-entry-liveness-2026-08-01.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r2.json", "identifier": "ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r2", "name": "百度严格无品牌搜索可比观察累计账本（20 次，修订 2）", "description": "绑定 20 次同口径百度严格无品牌观察、9 次生产发布、60 个查询快照和 499 个自然结果位次；每行保留源报告路径与 SHA-256，并保留目标页和品牌域零命中。该账本未调用模型或执行站外发布，不构成搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r2.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度严格无品牌搜索可比观察累计账本（20 次，修订 2）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r2.json" }, { "@type": "DataDownload", "name": "百度严格无品牌搜索可比观察累计账本 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r2.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r1.json", "identifier": "ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r1", "name": "百度严格无品牌搜索可比观察累计账本（19 次，修订 1）", "description": "绑定 19 次同口径百度严格无品牌观察、8 次生产发布、57 个查询快照和 474 个自然结果位次；每行保留源报告路径与 SHA-256，并保留目标页和品牌域零命中。该账本未调用模型或执行站外发布，不构成搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度严格无品牌搜索可比观察累计账本（19 次，修订 1）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r1.json" }, { "@type": "DataDownload", "name": "百度严格无品牌搜索可比观察累计账本 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r1.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-product-evaluation-suite-template-v1.json", "identifier": "ai-product-evaluation-suite-template-v1", "name": "AI 产品评测集 JSON 模板 v1", "description": "用于定义产品场景、测试样例、风险切片、指标、阈值、人工复核和生产监测字段的空白模板，并附三条虚构示例。", "url": "https://course.dragonai.tech/datasets/ai-product-evaluation-suite-template-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/templates/ai-product-evaluation-set#article", "url": "https://course.dragonai.tech/templates/ai-product-evaluation-set" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 产品评测集 JSON 模板 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-product-evaluation-suite-template-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-product-evaluation-suite-schema-v1.json", "identifier": "ai-product-evaluation-suite-schema-v1", "name": "AI 产品评测套件 JSON Schema v1", "description": "用于自动校验产品级 AI 评测套件必需字段、枚举值、指标定义、切片规则、测试样例与发布决策结构的 JSON Schema Draft 2020-12 文件。", "url": "https://course.dragonai.tech/datasets/ai-product-evaluation-suite-schema-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/templates/ai-product-evaluation-set#article", "url": "https://course.dragonai.tech/templates/ai-product-evaluation-set" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 产品评测套件 JSON Schema v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-product-evaluation-suite-schema-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/dragonai-content-review-policy-v1.json", "identifier": "dragonai-content-review-policy-v1", "name": "DragonAI 内容审核政策 JSON v1", "description": "公开 DragonAI 知识库五角色审核门槛、权重、硬失败规则、人工批准边界、bundle-v2 绑定范围和一条已发布内容示例回执的机器可读文件。", "url": "https://course.dragonai.tech/datasets/dragonai-content-review-policy-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/brand/editorial-methodology#article", "url": "https://course.dragonai.tech/brand/editorial-methodology" } ], "distribution": [ { "@type": "DataDownload", "name": "DragonAI 内容审核政策 JSON v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/dragonai-content-review-policy-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-product-manager-readiness-matrix-v1.json", "identifier": "ai-product-manager-readiness-matrix-v1", "name": "AI 产品经理八维能力与产物矩阵 v1", "description": "把问题定义、产品策略、用户交互、数据与上下文、原型系统、评测实验、协同采用、治理运营八类能力映射为可观察产物和验证动作。", "url": "https://course.dragonai.tech/datasets/ai-product-manager-readiness-matrix-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-product-manager-skills#article", "url": "https://course.dragonai.tech/guides/ai-product-manager-skills" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 产品经理八维能力与产物矩阵 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-product-manager-readiness-matrix-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-product-manager-eight-week-execution-roadmap-v1.json", "identifier": "ai-product-manager-eight-week-execution-roadmap-v1", "name": "AI 产品经理八周执行路线图 v1", "description": "把同一个 AI 产品练习按问题、策略、交互、数据、原型、评测、发布和治理拆成八周，逐周记录输入、最低产物、验收门禁、失败回退和阶段交接证据。", "url": "https://course.dragonai.tech/datasets/ai-product-manager-eight-week-execution-roadmap-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-product-manager-skills#article", "url": "https://course.dragonai.tech/guides/ai-product-manager-skills" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 产品经理八周执行路线图 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-product-manager-eight-week-execution-roadmap-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-product-manager-portfolio-evidence-template-v1.json", "identifier": "ai-product-manager-portfolio-evidence-template-v1", "name": "AI 产品经理作品集六张证据卡模板 v1", "description": "为每个作品集项目分别记录来源类型、证据环境、真实用户、生产部署、数据条件，以及问题决策、个人贡献、用户交互、系统数据、评测发布和失败复盘六张证据卡。", "url": "https://course.dragonai.tech/datasets/ai-product-manager-portfolio-evidence-template-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-product-manager-portfolio#article", "url": "https://course.dragonai.tech/guides/ai-product-manager-portfolio" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 产品经理作品集六张证据卡模板 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-product-manager-portfolio-evidence-template-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-product-decision-defense-cards-v1.json", "identifier": "ai-product-decision-defense-cards-v1", "name": "AI 产品决策十二道答辩卡 v1", "description": "用于口头答辩和项目复盘的十二道决策卡，逐题记录结论、依据、取舍、失败、下一步、证据版本和复核问题，不生成总分或能力等级。", "url": "https://course.dragonai.tech/datasets/ai-product-decision-defense-cards-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-product-manager-interview-questions#article", "url": "https://course.dragonai.tech/guides/ai-product-manager-interview-questions" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 产品决策十二道答辩卡 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-product-decision-defense-cards-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-product-decision-ownership-map-v1.json", "identifier": "ai-product-decision-ownership-map-v1", "name": "AI 产品七类决策协作图 v1", "description": "用七类决策记录产品判断、协作输入、专业责任、最小产物、失败追问和复审条件，不生成能力等级或职业结果预测。", "url": "https://course.dragonai.tech/datasets/ai-product-decision-ownership-map-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/what-does-ai-product-manager-do#article", "url": "https://course.dragonai.tech/guides/what-does-ai-product-manager-do" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 产品七类决策协作图 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-product-decision-ownership-map-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-pm-course-due-diligence-checklist-v1.json", "identifier": "ai-pm-course-due-diligence-checklist-v1", "name": "AI 产品经理课程报名尽调清单 v1", "description": "用于逐项核验目标匹配、课程产物、公开样例、反馈机制、师资证据、学习条件、结果主张、费用、合同退款、变更退出、个人信息与宣传申诉的空白清单。", "url": "https://course.dragonai.tech/datasets/ai-pm-course-due-diligence-checklist-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/how-to-choose-ai-product-manager-course#article", "url": "https://course.dragonai.tech/guides/how-to-choose-ai-product-manager-course" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 产品经理课程报名尽调清单 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-pm-course-due-diligence-checklist-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-pm-course-syllabus-audit-matrix-v1.json", "identifier": "ai-pm-course-syllabus-audit-matrix-v1", "name": "AI 产品经理课程大纲核验矩阵 v1", "description": "从八类核验模块中按目标确定必需项，映射为可观察学习目标、练习、产物和验收动作，并用缺失、仅提及、有练习、有证据四种状态逐项记录课程覆盖。", "url": "https://course.dragonai.tech/datasets/ai-pm-course-syllabus-audit-matrix-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-product-manager-course-syllabus#article", "url": "https://course.dragonai.tech/guides/ai-product-manager-course-syllabus" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 产品经理课程大纲核验矩阵 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-pm-course-syllabus-audit-matrix-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-product-prd-field-checklist-v1.json", "identifier": "ai-product-prd-field-checklist-v1", "name": "AI 产品 PRD 适用字段与验收清单 v1", "description": "把 AI 产品 PRD 拆成决策背景、用户任务、行为合同、数据上下文、评测门禁、失败接管、Agent 权限、治理、运行约束和发布运营十个可核验部分。", "url": "https://course.dragonai.tech/datasets/ai-product-prd-field-checklist-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-product-prd#article", "url": "https://course.dragonai.tech/guides/ai-product-prd" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 产品 PRD 适用字段与验收清单 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-product-prd-field-checklist-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/rag-product-decision-checklist-v1.json", "identifier": "rag-product-decision-checklist-v1", "name": "RAG 产品决策深度与七项核验清单 v1", "description": "用解释、规格化、评测诊断、生产运营四级深度，以及选型、语料治理、摄取索引、检索、生成交互、评测、运行七项决策核验产品经理的 RAG 实践能力。", "url": "https://course.dragonai.tech/datasets/rag-product-decision-checklist-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/rag-for-product-managers#article", "url": "https://course.dragonai.tech/guides/rag-for-product-managers" } ], "distribution": [ { "@type": "DataDownload", "name": "RAG 产品决策深度与七项核验清单 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/rag-product-decision-checklist-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/rag-fine-tuning-decision-matrix-v1.json", "identifier": "rag-fine-tuning-decision-matrix-v1", "name": "RAG、微调与混合方案决策矩阵 v1", "description": "用问题类型、更新频率、来源追溯、数据条件、评测、成本延迟、运维和组合必要性八项字段比较基础方案、RAG、微调与混合方案。", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-decision-matrix-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/comparisons/rag-vs-fine-tuning#article", "url": "https://course.dragonai.tech/comparisons/rag-vs-fine-tuning" } ], "distribution": [ { "@type": "DataDownload", "name": "RAG、微调与混合方案决策矩阵 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/rag-fine-tuning-decision-matrix-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/rag-fine-tuning-quick-selector-v1.json", "identifier": "rag-fine-tuning-quick-selector-v1", "name": "RAG 与微调八情景快速选型表 v1", "description": "把八种常见选型情景映射为继续基线、验证 RAG、验证微调、消融后验证组合或先补数据五类动作，并逐项给出必须验证的证据与停止条件。", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-quick-selector-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "RAG 与微调八情景快速选型表 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/rag-fine-tuning-quick-selector-v1.json" }, { "@type": "DataDownload", "name": "RAG 与微调八情景快速选型表（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/rag-fine-tuning-quick-selector-v1.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/rag-fine-tuning-reliable-source-evidence-map-v1.json", "identifier": "rag-fine-tuning-reliable-source-evidence-map-v1", "name": "RAG 与微调可靠来源—决策—验证映射 v1", "description": "把百度智能云、腾讯云文档及四份国际一手技术与治理来源逐项映射为可支持的选型判断、应执行的验证动作和不可推出的结论，绑定固定 RAG 与微调选型问题。", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-reliable-source-evidence-map-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "RAG 与微调可靠来源—决策—验证映射 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/rag-fine-tuning-reliable-source-evidence-map-v1.json" }, { "@type": "DataDownload", "name": "RAG 与微调可靠来源—决策—验证映射（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/rag-fine-tuning-reliable-source-evidence-map-v1.csv" }, { "@type": "DataDownload", "name": "RAG 与微调来源声明范围验证器（Python）", "encodingFormat": "text/x-python", "contentUrl": "https://course.dragonai.tech/datasets/rag-fine-tuning-source-claim-validator-v1.py" }, { "@type": "DataDownload", "name": "RAG 与微调来源声明范围验证器夹具（JSON）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/rag-fine-tuning-source-claim-validator-v1-fixtures.json" }, { "@type": "DataDownload", "name": "RAG 与微调来源声明范围验证器说明（Markdown）", "encodingFormat": "text/markdown", "contentUrl": "https://course.dragonai.tech/datasets/rag-fine-tuning-source-claim-validator-v1-guide.md" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-reference-v1.json", "identifier": "rag-fine-tuning-selection-reference-v1", "name": "RAG、微调与组合方案可执行选型参考 v1", "description": "零依赖 Python 决策合同：严格校验 12 个布尔事实与两条跨字段不变量；两个候选通过同一门槛后，在同一评测集、负载、成本周期和评分量表下对五项实测值做无权重、容差感知的 Pareto 比较。12 个选择用例、4 个比较用例与 5 个拒绝用例共同验证失败关闭。", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-reference-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "RAG、微调与组合方案可执行选型参考 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-reference-v1.json" }, { "@type": "DataDownload", "name": "RAG、微调与组合方案可运行 Python 参考实现", "encodingFormat": "text/x-python", "contentUrl": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-reference-v1.py" }, { "@type": "DataDownload", "name": "RAG、微调与组合方案机器回归夹具", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-reference-v1-fixtures.json" }, { "@type": "DataDownload", "name": "RAG、微调与组合方案可执行选型说明", "encodingFormat": "text/markdown", "contentUrl": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-reference-v1-guide.md" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-13.json", "identifier": "ai-search-rag-baidu-selection-query-current-delta-2026-08-13", "name": "百度无品牌 RAG 微调选型查询当前变化 v1", "description": "匿名只读复核固定无品牌选型查询，逐位次记录 8 个当前结果，并与 2026-08-12 基线比较 4 个留存、4 个进入、4 个退出和完全变化的前三主机。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-13.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度无品牌 RAG 微调选型查询当前变化 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-13.json" }, { "@type": "DataDownload", "name": "百度无品牌 RAG 微调选型查询当前变化（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-13.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-13-r2.json", "identifier": "ai-search-rag-baidu-selection-query-current-delta-2026-08-13-r2", "name": "百度无品牌 RAG 微调选型查询当前变化 r2", "description": "匿名只读复核固定无品牌选型查询：相对 2026-08-13 01:11 UTC 基线，当前 8 条自然结果保留 4、进入 4、退出 4，2 条留存 URL 换位，前三主机为腾讯云/百度智能云/百家号；DragonAI 目标页、品牌域与生成式摘要文本均为 0，且本轮未捕获可核对引用链接。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-13-r2.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/how-ai-search-chooses-citation-sources#article", "url": "https://course.dragonai.tech/guides/how-ai-search-chooses-citation-sources" }, { "@type": "Article", "@id": "https://course.dragonai.tech/guides/how-to-get-site-into-ai-search#article", "url": "https://course.dragonai.tech/guides/how-to-get-site-into-ai-search" }, { "@type": "Article", "@id": "https://course.dragonai.tech/guides/what-is-generative-engine-optimization#article", "url": "https://course.dragonai.tech/guides/what-is-generative-engine-optimization" }, { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度无品牌 RAG 微调选型查询当前变化 r2", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-13-r2.json" }, { "@type": "DataDownload", "name": "百度无品牌 RAG 微调选型查询当前变化 r2（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-13-r2.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-13-r3.json", "identifier": "ai-search-rag-baidu-selection-query-current-delta-2026-08-13-r3", "name": "百度无品牌 RAG 微调选型查询同发布变化 r3", "description": "2026-08-13 22:00 UTC 匿名只读复核固定选型查询：相对 21:22 UTC 同发布观察，8 条自然结果保留 4、进入 4、退出 4，3 条留存 URL 换位；前三主机为腾讯云、百家号和智源社区，第 4 位新进入标题直接命中 RAG 和微调选型的知乎页面。DragonAI 目标页和品牌域均为 0。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-13-r3.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度无品牌 RAG 微调选型查询同发布变化 r3", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-13-r3.json" }, { "@type": "DataDownload", "name": "百度无品牌 RAG 微调选型查询同发布变化 r3（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-13-r3.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-14-r4.json", "identifier": "ai-search-rag-baidu-selection-query-current-delta-2026-08-14-r4", "name": "百度无品牌 RAG 微调选型查询同发布变化 r4", "description": "2026-08-14 00:53 UTC 匿名只读复核固定选型查询：相对 2026-08-13 22:00 UTC 同发布观察，8 条自然结果保留 7、进入 1、退出 1，前六位 URL 与位次不变；第 7 位进入什么值得买的 RAG、微调与长上下文决策框架，原第 7 位百家号结果退出，Red Hat 保持第 8 位。DragonAI 目标页和品牌域均为 0。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-14-r4.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度无品牌 RAG 微调选型查询同发布变化 r4", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-14-r4.json" }, { "@type": "DataDownload", "name": "百度无品牌 RAG 微调选型查询同发布变化 r4（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-14-r4.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-14-r5.json", "identifier": "ai-search-rag-baidu-selection-query-current-delta-2026-08-14-r5", "name": "百度无品牌 RAG 微调选型查询同发布变化 r5", "description": "2026-08-14 12:53 UTC 匿名只读复核固定选型查询：相对同日 00:53 UTC 的同发布观察，8 条自然结果保留 7、进入 1、退出 1，其中 1 条留存结果下移；前六位 URL 与位次不变，第 7 位进入新的 CSDN 企业级 RAG 技术方案，什么值得买从第 7 位移至第 8 位，Red Hat 退出。DragonAI 目标页和品牌域均为 0。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-14-r5.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度无品牌 RAG 微调选型查询同发布变化 r5", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-14-r5.json" }, { "@type": "DataDownload", "name": "百度无品牌 RAG 微调选型查询同发布变化 r5（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-14-r5.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-current-document-features-2026-08-14-r4.json", "identifier": "ai-search-rag-baidu-selection-current-document-features-2026-08-14-r4", "name": "百度当前 RAG 微调选型结果文档交付特征 r4", "description": "对 2026-08-14 00:53 UTC 固定无品牌选型查询的 8 个当前自然结果执行一次只读页面获取：3 个返回可用 HTML 文档、3 个为 HTML 壳、2 个为 HTTP 或请求错误；可用文档中 3 个有摘要、2 个有 canonical、0 个暴露预登记扩展名的下载链接。失败不回填，结构共现不构成排名原因、来源质量或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-current-document-features-2026-08-14-r4.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度当前 RAG 微调选型结果文档交付特征 r4", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-current-document-features-2026-08-14-r4.json" }, { "@type": "DataDownload", "name": "百度当前 RAG 微调选型结果文档交付特征 r4（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-current-document-features-2026-08-14-r4.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-target-delivery-comparison-2026-08-14-r4.json", "identifier": "ai-search-rag-baidu-selection-target-delivery-comparison-2026-08-14-r4", "name": "DragonAI 目标页与百度当前选型结果文档交付特征对照 r4", "description": "使用同一只读 HTML 特征提取器，对当前生产目标页与 2026-08-14 00:53 UTC 固定无品牌选型查询八个自然结果中 3 个成功返回的 HTML 文档进行 16 项交付特征对照。目标页有 3 项高于全部可用竞争文档；5 个失败或 HTML 壳响应不插补。结构差异不构成排名原因、来源质量、目标可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-target-delivery-comparison-2026-08-14-r4.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "DragonAI 目标页与百度当前选型结果文档交付特征对照 r4", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-target-delivery-comparison-2026-08-14-r4.json" }, { "@type": "DataDownload", "name": "DragonAI 目标页与百度当前选型结果文档交付特征对照 r4（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-target-delivery-comparison-2026-08-14-r4.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-current-document-features-2026-08-13-r3.json", "identifier": "ai-search-rag-baidu-selection-current-document-features-2026-08-13-r3", "name": "百度当前 RAG 微调选型结果文档交付特征 r3", "description": "对 2026-08-13 22:00 UTC 固定无品牌选型查询的 8 个自然结果执行一次只读页面获取：3 个返回可用 HTML 文档、3 个为 HTML 壳、2 个返回 HTTP 错误；可用文档中 3 个有摘要、2 个有 canonical、0 个暴露预登记扩展名的下载链接。失败不回填，结构共现不构成排名原因、来源质量或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-current-document-features-2026-08-13-r3.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度当前 RAG 微调选型结果文档交付特征 r3", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-current-document-features-2026-08-13-r3.json" }, { "@type": "DataDownload", "name": "百度当前 RAG 微调选型结果文档交付特征 r3（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-current-document-features-2026-08-13-r3.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-target-delivery-comparison-2026-08-13-r3.json", "identifier": "ai-search-rag-baidu-selection-target-delivery-comparison-2026-08-13-r3", "name": "DragonAI 目标页与百度当前选型结果文档交付特征对照 r3", "description": "使用同一只读 HTML 特征提取器，对当前生产目标页与固定无品牌选型查询八个自然结果中 3 个成功返回的 HTML 文档进行 16 项交付特征对照。目标页有 3 项高于全部可用竞争文档；5 个失败或 HTML 壳响应不插补。结构差异不构成排名原因、来源质量、目标可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-target-delivery-comparison-2026-08-13-r3.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "DragonAI 目标页与百度当前选型结果文档交付特征对照 r3", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-target-delivery-comparison-2026-08-13-r3.json" }, { "@type": "DataDownload", "name": "DragonAI 目标页与百度当前选型结果文档交付特征对照 r3（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-target-delivery-comparison-2026-08-13-r3.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/rag-citation-traceability-reference-v1.json", "identifier": "rag-citation-traceability-reference-v1", "name": "RAG 引用可追溯最小参考实现 v1", "description": "零依赖 Python 参考实现：把逐条陈述绑定到稳定片段 ID 与来源 URL，并在无匹配证据时失败关闭；附可直接运行的自测试与实现边界。", "url": "https://course.dragonai.tech/datasets/rag-citation-traceability-reference-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "RAG 引用可追溯最小参考实现 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/rag-citation-traceability-reference-v1.json" }, { "@type": "DataDownload", "name": "RAG 引用可追溯可运行 Python 参考实现", "encodingFormat": "text/x-python", "contentUrl": "https://course.dragonai.tech/datasets/rag-citation-traceability-reference-v1.py" }, { "@type": "DataDownload", "name": "RAG 引用可追溯参考实现说明", "encodingFormat": "text/markdown", "contentUrl": "https://course.dragonai.tech/datasets/rag-citation-traceability-reference-v1-guide.md" }, { "@type": "DataDownload", "name": "RAG 引用可追溯机器可读测试夹具", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/rag-citation-traceability-reference-v1-fixtures.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-finetuning-citation-sample-2026-07.json", "identifier": "ai-search-rag-finetuning-citation-sample-2026-07", "name": "千问、Kimi、Gemini 的 RAG 与微调选型引用样本（2026-07）", "description": "同一中文无品牌问题在千问、Kimi、Gemini 各运行三次后的引用计数、主机名分布、平台交集、目标引用与方法限制。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-finetuning-citation-sample-2026-07.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问、Kimi、Gemini 的 RAG 与微调选型引用样本（2026-07）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-finetuning-citation-sample-2026-07.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-snapshot-2026-07-22.json", "identifier": "ai-search-citation-benchmark-snapshot-2026-07-22", "name": "千问、Kimi、Gemini 固定引用问题双窗口基准（2026-07-21—22）", "description": "同一固定中文无品牌元问题在正式发布前基线 B 与发布后时间盒 C 的18条逐运行记录、具体引用 URL、零目标结果和模型漂移边界。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-snapshot-2026-07-22.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问、Kimi、Gemini 固定引用问题双窗口基准（2026-07-21—22）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-snapshot-2026-07-22.json" }, { "@type": "DataDownload", "name": "正式窗口汇总 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-formal-window-summary-2026-07-22.csv" }, { "@type": "DataDownload", "name": "逐运行引用 URL CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-run-level-urls-2026-07-22.csv" }, { "@type": "DataDownload", "name": "按窗口与平台汇总的引用主机频次和运行覆盖 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-host-frequency-2026-07-22.csv" }, { "@type": "DataDownload", "name": "平台间与跨窗口引用主机集合重叠 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-host-overlap-2026-07-22.csv" }, { "@type": "DataDownload", "name": "逐 URL 复现与跨窗口持续性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-url-persistence-2026-07-22.csv" }, { "@type": "DataDownload", "name": "逐声明证据定位索引（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-claim-evidence-index-2026-07-22.csv" }, { "@type": "DataDownload", "name": "指标、分母、公式与跨窗口可比性定义（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-metric-definitions-2026-07-22.csv" }, { "@type": "DataDownload", "name": "逐声明证据溯源图（JSON-LD）", "encodingFormat": "application/ld+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-claim-evidence-graph-2026-07-22.json" }, { "@type": "DataDownload", "name": "三平台主要引用主机证据（JSON）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-leading-hosts-2026-07-22.json" }, { "@type": "DataDownload", "name": "四条无品牌检索问法 × 三平台主要引用主机十六行证据表（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-leading-hosts-2026-07-22.csv" }, { "@type": "DataDownload", "name": "固定无品牌查询的直接答案、复现协议与证据卡（JSON）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-exact-query-answer-2026-07-22.json" }, { "@type": "DataDownload", "name": "固定无品牌查询直接答案 JSON Schema", "encodingFormat": "application/schema+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-exact-query-answer-schema-2026-07-22.json" }, { "@type": "DataDownload", "name": "十九条可引用事实、证据片段与规则片段扁平表（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-citation-ready-claims-2026-07-22.csv" }, { "@type": "DataDownload", "name": "完整研究存档包（含 README、CITATION.cff、校验清单与仓库元数据）", "encodingFormat": "application/zip", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-archive-2026-07-22.zip" }, { "@type": "DataDownload", "name": "BagIt 1.0 数字保存包（含 Payload-Oxum 与 SHA-256 payload/tag manifests）", "encodingFormat": "application/zip", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-bagit-2026-07-22.zip" }, { "@type": "DataDownload", "name": "JATS 1.3 machine-readable benchmark article", "encodingFormat": "application/xml", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-report-2026-07-22.jats.xml" }, { "@type": "DataDownload", "name": "OAI Dublin Core repository-ingestion metadata", "encodingFormat": "application/xml", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-oai-dublin-core-2026-07-22.xml" }, { "@type": "DataDownload", "name": "推荐引用元数据（CITATION.cff）", "encodingFormat": "text/yaml", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-citation-2026-07-22.cff" }, { "@type": "DataDownload", "name": "CSL-JSON 引用元数据（UTF-8）", "encodingFormat": "application/vnd.citationstyles.csl+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-citation-2026-07-22.csl.json" }, { "@type": "DataDownload", "name": "RO-Crate 1.3 研究对象元数据（JSON-LD）", "encodingFormat": "application/ld+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-ro-crate-metadata-2026-07-22.json" }, { "@type": "DataDownload", "name": "BibTeX 引用元数据（UTF-8）", "encodingFormat": "application/x-bibtex", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-citation-2026-07-22.bib" }, { "@type": "DataDownload", "name": "RIS 引用元数据（UTF-8）", "encodingFormat": "application/x-research-info-systems", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-citation-2026-07-22.ris" }, { "@type": "DataDownload", "name": "研究存档清单（MANIFEST.json）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-manifest-2026-07-22.json" }, { "@type": "DataDownload", "name": "文件校验和（SHA256SUMS）", "encodingFormat": "text/plain", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-sha256sums-2026-07-22.txt" }, { "@type": "DataDownload", "name": "研究存档复核说明（README）", "encodingFormat": "text/markdown", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-readme-2026-07-22.md" }, { "@type": "DataDownload", "name": "中文固定问题答案、证据地图与复核说明（README.zh-CN）", "encodingFormat": "text/markdown", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-readme-zh-cn-2026-07-22.md" }, { "@type": "DataDownload", "name": "五十四个表格投影的数据字典（CODEBOOK）", "encodingFormat": "text/markdown", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-codebook-2026-07-22.md" }, { "@type": "DataDownload", "name": "固定问题引用基准的完整方法与证据边界（METHODS）", "encodingFormat": "text/markdown", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-methods-2026-07-22.md" }, { "@type": "DataDownload", "name": "权威基准快照 JSON Schema（Draft 2020-12）", "encodingFormat": "application/schema+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-snapshot-schema-v1.1.0.json" }, { "@type": "DataDownload", "name": "CC BY 4.0 许可、署名与第三方权利边界说明", "encodingFormat": "text/plain", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-license-2026-07-22.txt" }, { "@type": "DataDownload", "name": "仓库提交元数据（Zenodo JSON）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-zenodo-metadata-2026-07-22.json" }, { "@type": "DataDownload", "name": "DataCite 4.7 仓库元数据模板（JSON API）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-datacite-metadata-2026-07-22.json" }, { "@type": "DataDownload", "name": "RFC 9264 研究对象 Linkset（JSON）", "encodingFormat": "application/linkset+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-linkset-2026-07-22.json" }, { "@type": "DataDownload", "name": "MLCommons Croissant 1.0 数据集元数据（JSON-LD）", "encodingFormat": "application/ld+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-croissant-metadata-2026-07-22.json" }, { "@type": "DataDownload", "name": "Frictionless Tabular Data Package 元数据（JSON）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-datapackage-2026-07-22.json" }, { "@type": "DataDownload", "name": "W3C DCAT 3 dataset catalog record (JSON-LD)", "encodingFormat": "application/ld+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-dcat-metadata-2026-07-22.json" }, { "@type": "DataDownload", "name": "双窗口汇总、主机频次、集合重叠与 URL 持续性离线复算脚本", "encodingFormat": "text/x-python", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-reproduce-summary-2026-07-22.py" }, { "@type": "DataDownload", "name": "提供商窗口核心指标摘要表（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-provider-window-summary-2026-07-22.csv" }, { "@type": "DataDownload", "name": "B v2 与 C 历史口径不可直接比较摘要图", "encodingFormat": "image/svg+xml", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-summary-2026-07-22.svg" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-2026-07-28.json", "identifier": "ai-search-citation-benchmark-observation-2026-07-28", "name": "千问、Kimi、Gemini 固定问题非因果补充观察（2026-07-28）", "description": "同一固定中文无品牌问题的九次真实提供商输出补充观察，逐运行保留具体引用 URL、零目标结果、竞争主机、模型漂移、归因可靠性与非因果边界；不是新的正式 B/C 窗口。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-2026-07-28.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问、Kimi、Gemini 固定问题非因果补充观察（2026-07-28）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-2026-07-28.json" }, { "@type": "DataDownload", "name": "九次补充观察逐运行引用 URL（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-2026-07-28.csv" }, { "@type": "DataDownload", "name": "九次补充观察竞争主机频次、运行覆盖与来源类型审计（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-host-frequency-2026-07-28.csv" }, { "@type": "DataDownload", "name": "3 providers x 5 source types: citation occurrences, hosts and citation-surface coverage (CSV)", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-source-type-summary-2026-07-28.csv" }, { "@type": "DataDownload", "name": "46-host x 3-provider citation-occurrence and run-coverage matrix with explicit zero cells (CSV)", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-provider-host-matrix-2026-07-28.csv" }, { "@type": "DataDownload", "name": "9-run x 46-host citation incidence matrix with explicit zero cells (CSV)", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-run-host-incidence-2026-07-28.csv" }, { "@type": "DataDownload", "name": "Within-provider same-run competitor-host co-citation edge list (CSV)", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-host-cocitation-edges-2026-07-28.csv" }, { "@type": "DataDownload", "name": "138-row provider by competitor-host co-citation node profile with explicit zero cells (CSV)", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-host-cocitation-node-profile-2026-07-28.csv" }, { "@type": "DataDownload", "name": "Three-provider co-citation graph structure summary with density, components and explicit zero denominators (CSV)", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-host-cocitation-graph-summary-2026-07-28.csv" }, { "@type": "DataDownload", "name": "Nine-run real-provider source-mix and co-citation evidence figure (SVG)", "encodingFormat": "image/svg+xml", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-summary-2026-07-28.svg" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-five-batch-run-level-2026-07-30.json", "identifier": "ai-search-citation-benchmark-five-batch-run-level-2026-07-30", "name": "千问、Kimi、Gemini 五批次 45 次运行逐引用 URL 数据", "description": "把同一固定中文无品牌问题的五个真实提供商批次统一为 45 条运行记录与 552 次具体引用 URL 出现；JSON 保留逐运行引用数组，566 行 CSV 为 14 次零引用运行各保留一条显式空 URL 行，便于完整重算分母。DragonAI 目标引用与品牌提及均为零。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-five-batch-run-level-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问、Kimi、Gemini 五批次 45 次运行逐引用 URL 数据", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-five-batch-run-level-2026-07-30.json" }, { "@type": "DataDownload", "name": "五批次逐运行引用 URL 与显式零运行（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-five-batch-run-level-2026-07-30.csv" }, { "@type": "DataDownload", "name": "五批次 AI 引用与百度六轮严格查询当前证据摘要图（SVG）", "encodingFormat": "image/svg+xml", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-current-evidence-summary-2026-07-30.svg" }, { "@type": "DataDownload", "name": "五批次 AI 引用与百度九次严格观察证据摘要图（SVG）", "encodingFormat": "image/svg+xml", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-nine-observation-evidence-summary-2026-07-30.svg" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-five-batch-url-profile-2026-07-30.json", "identifier": "ai-search-citation-benchmark-five-batch-url-profile-2026-07-30", "name": "千问、Kimi、Gemini 五批次具体引用 URL 跨平台画像", "description": "把五批次 45 次运行的 552 次具体引用 URL 出现压缩为 194 个精确 URL 画像，逐行保留平台、批次、运行与出现次数。193 个 URL 仅由一家平台展示，只有 arxiv.org/abs/2312.10997 同时由 Kimi 与 Gemini 展示；11 个 URL 覆盖全部五批次，DragonAI 目标 URL 仍为零。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-five-batch-url-profile-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问、Kimi、Gemini 五批次具体引用 URL 跨平台画像", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-five-batch-url-profile-2026-07-30.json" }, { "@type": "DataDownload", "name": "五批次具体引用 URL 跨平台覆盖画像（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-five-batch-url-profile-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-url-delivery-shape-2026-07-30.json", "identifier": "ai-search-citation-benchmark-url-delivery-shape-2026-07-30", "name": "千问、Kimi、Gemini 五批次引用 URL 交付形态画像", "description": "将五批次 45 次运行中的 194 个精确引用 URL 确定性拆解为路径深度、认可文件后缀、查询参数、片段和 HTTPS 字段。182 个 URL 是无认可文件后缀的深层页面，156 个至少有两级路径；DragonAI 目标页形态与最高频类别一致，但 45 次运行中的目标引用仍为零。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-url-delivery-shape-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问、Kimi、Gemini 五批次引用 URL 交付形态画像", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-url-delivery-shape-2026-07-30.json" }, { "@type": "DataDownload", "name": "五批次逐引用 URL 交付形态画像（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-url-delivery-shape-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-citation-retrieval-evidence-cards-2026-07-30.json", "identifier": "ai-search-rag-citation-retrieval-evidence-cards-2026-07-30", "name": "千问、Kimi、Gemini RAG 与微调引用来源中文检索证据卡", "description": "把四条中文无品牌问法绑定到同一审核答案、八条可引用声明、逐项证据 URL、百度严格观察状态和显式非成功边界。JSON 适合检索系统导入，Markdown 适合人工复核；它不构成搜索可见性、外链证据或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-citation-retrieval-evidence-cards-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问、Kimi、Gemini RAG 与微调引用来源中文检索证据卡", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-citation-retrieval-evidence-cards-2026-07-30.json" }, { "@type": "DataDownload", "name": "四条中文无品牌问法的可引用证据卡（Markdown）", "encodingFormat": "text/markdown", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-citation-retrieval-evidence-cards-2026-07-30.md" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-provider-overlap-matrix-2026-07-30.json", "identifier": "ai-search-citation-benchmark-provider-overlap-matrix-2026-07-30", "name": "千问、Kimi、Gemini 五批次引用 URL 与主机跨平台重合矩阵", "description": "把五批次 45 次运行中 194 个精确引用 URL 和 114 个主机压缩为三个平台对；逐对公开精确 URL 与主机的交集、并集、Jaccard 和重合系数。只有 Kimi × Gemini 共享 1 个精确 URL；千问 × Kimi 仅在主机层共享 CSDN 与知乎专栏。目标引用仍为零。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-provider-overlap-matrix-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问、Kimi、Gemini 五批次引用 URL 与主机跨平台重合矩阵", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-provider-overlap-matrix-2026-07-30.json" }, { "@type": "DataDownload", "name": "三平台对精确 URL 与主机重合矩阵（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-provider-overlap-matrix-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-provider-concentration-2026-07-30.json", "identifier": "ai-search-citation-benchmark-provider-concentration-2026-07-30", "name": "千问、Kimi、Gemini 五批次引用 URL 与主机集中度", "description": "基于同一固定问题的五批次、45 次真实提供商运行和 552 次具体引用 URL 出现，逐平台发布 URL 与主机层的 Top-1、Top-5 占比、HHI 和有效来源数。主机 HHI 在本窗口中依次为千问 0.157644、Kimi 0.142400、Gemini 0.031173；这是冻结观察的描述性集中度，不代表来源质量、长期偏好、排序因果或 GEO 成功，目标引用与品牌提及仍为零。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-provider-concentration-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问、Kimi、Gemini 五批次引用 URL 与主机集中度", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-provider-concentration-2026-07-30.json" }, { "@type": "DataDownload", "name": "三平台引用 URL 与主机集中度（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-provider-concentration-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-provider-host-persistence-2026-07-30.json", "identifier": "ai-search-citation-benchmark-provider-host-persistence-2026-07-30", "name": "千问、Kimi、Gemini 五批次固定问题引用主机持续性矩阵", "description": "汇总五个同查询批次、45 次真实提供商运行与 552 次具体引用 URL 出现，将 3 家平台和 114 个观察主机展开为 342 行显式零矩阵；逐行保留五批次出现向量、运行覆盖、来源类型、零目标结果和非长期偏好边界。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-provider-host-persistence-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问、Kimi、Gemini 五批次固定问题引用主机持续性矩阵", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-provider-host-persistence-2026-07-30.json" }, { "@type": "DataDownload", "name": "五批次提供商 × 引用主机显式零矩阵（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-provider-host-persistence-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-2026-07-30.json", "identifier": "ai-search-rag-cross-surface-host-persistence-2026-07-30", "name": "五批 AI 引用域 × 百度六轮严格无品牌结果跨表面持久性交集", "description": "将五批次、45 次固定问题 AI 引用主机与六轮、18 个百度严格无品牌查询快照按规范化主机全外联结为 123 行显式零面板；6 个主机跨表面重合，只有 cloud.tencent.com 同时满足 AI 某提供商覆盖五批与百度至少一个 URL 覆盖六轮，DragonAI 两侧均未出现。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "五批 AI 引用域 × 百度六轮严格无品牌结果跨表面持久性交集", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-2026-07-30.json" }, { "@type": "DataDownload", "name": "123 行跨表面主机持久性显式零面板（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-source-type-summary-2026-07-30.json", "identifier": "ai-search-rag-cross-surface-source-type-summary-2026-07-30", "name": "五批 AI 引用域 × 百度六轮严格无品牌结果跨表面来源类型汇总", "description": "将 123 行跨表面主机持久性面板压缩为 6 个互斥来源类型；6 个跨表面重合主机中，厂商文档 3 个、专家社区 2 个、其他实践者或出版者 1 个，唯一双侧持续主机 cloud.tencent.com 属于厂商文档。百度侧 9 个没有既有 AI 来源类型的主机保留为未分类，不做推断式补标；DragonAI 两侧均未出现。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-source-type-summary-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "五批 AI 引用域 × 百度六轮严格无品牌结果跨表面来源类型汇总", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-source-type-summary-2026-07-30.json" }, { "@type": "DataDownload", "name": "6 行跨表面来源类型汇总（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-source-type-summary-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-baseline-attribution-v2.json", "identifier": "ai-search-citation-benchmark-baseline-attribution-v2", "name": "千问、Kimi、Gemini 固定问题发布前基线：引用归因规则 v2 离线重算", "description": "不重新调用模型，按答案可归因规则重算固定 B 基线的九次原始响应；逐运行绑定回答哈希、原始载荷哈希、引用 URL、旧规则差值和零目标引用结果。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-baseline-attribution-v2.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问、Kimi、Gemini 固定问题发布前基线：引用归因规则 v2 离线重算", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-baseline-attribution-v2.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-attribution-lineage-v1.json", "identifier": "ai-search-citation-benchmark-attribution-lineage-v1", "name": "千问、Kimi、Gemini 固定问题引用资产归因版本与可比性谱系 v1", "description": "逐项登记 26 个公开分析资产的归因规则版本、SHA-256、严格趋势端点资格和替代关系；仅 B 窗口 v2 重算基线可作为未来同口径严格端点，C 窗口与补充观察保持历史未版本化边界。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-attribution-lineage-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问、Kimi、Gemini 固定问题引用资产归因版本与可比性谱系 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-attribution-lineage-v1.json" }, { "@type": "DataDownload", "name": "归因版本与可比性谱系 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-attribution-lineage-v1.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-citation-source-type-taxonomy-v1.json", "identifier": "ai-search-citation-source-type-taxonomy-v1", "name": "千问、Kimi、Gemini 固定问题引用来源类型透明分类 v1", "description": "基于正式 B/C 双窗口引用主机频次表，按公开主机名后缀规则把 92 个平台窗口主机行、58 个唯一主机名和 230 次引用出现映射为五类来源，并明确回退分类、模型漂移与非质量排名边界。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-source-type-taxonomy-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问、Kimi、Gemini 固定问题引用来源类型透明分类 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-source-type-taxonomy-v1.json" }, { "@type": "DataDownload", "name": "平台窗口来源类型汇总 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-source-type-provider-window-v1.csv" }, { "@type": "DataDownload", "name": "逐主机来源类型审计 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-source-type-host-audit-v1.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-citation-url-source-type-persistence-v1.json", "identifier": "ai-search-citation-url-source-type-persistence-v1", "name": "Qwen、Kimi、Gemini 逐 URL 持续性与来源类型联结数据 v1", "description": "将固定 B/C 双窗口的 98 个“平台 × 具体引用 URL”持续性记录，与公开主机名后缀来源类型规则确定性联结；逐行保留平台、模型、精确 URL、来源类型、两窗运行覆盖、严格可比门禁和非质量排名边界。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-url-source-type-persistence-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Qwen、Kimi、Gemini 逐 URL 持续性与来源类型联结数据 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-url-source-type-persistence-v1.json" }, { "@type": "DataDownload", "name": "逐 URL 持续性与来源类型联结 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-url-source-type-persistence-v1.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-provider-citation-surface-crosswalk-v1.json", "identifier": "ai-search-provider-citation-surface-crosswalk-v1", "name": "千问、Kimi、Gemini 官方来源返回表面与基准适配器对照 v1", "description": "逐行对照三家第一方文档公开的来源返回表面、DragonAI 固定问题基准的适配器与 URL 提取路径，并明确产品帮助页、API 字段、索引、候选集和实际引用之间的推断边界。", "url": "https://course.dragonai.tech/datasets/ai-search-provider-citation-surface-crosswalk-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问、Kimi、Gemini 官方来源返回表面与基准适配器对照 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-provider-citation-surface-crosswalk-v1.json" }, { "@type": "DataDownload", "name": "三平台官方来源返回表面与基准适配器对照 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-provider-citation-surface-crosswalk-v1.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-unbranded-baseline-2026-07-29.json", "identifier": "ai-search-rag-baidu-unbranded-baseline-2026-07-29", "name": "百度 RAG 与微调引用来源严格无品牌搜索基线（2026-07-29）", "description": "公开 3 条严格无品牌查询和 1 条 site 诊断查询在百度匿名自然搜索结果中的逐位次记录、规范化目标匹配结果与结果集哈希；保留目标页未出现的零结果。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-unbranded-baseline-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌搜索基线（2026-07-29）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-unbranded-baseline-2026-07-29.json" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格搜索基线 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-unbranded-baseline-2026-07-29.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-host-query-matrix-2026-07-29.json", "identifier": "ai-search-rag-baidu-host-query-matrix-2026-07-29", "name": "百度严格无品牌查询 × 观察主机出现矩阵（2026-07-29）", "description": "从既有百度严格无品牌搜索基线确定性派生 13 个观察主机在 3 条查询中的出现、缺席、次数与位次；site: 诊断被排除在全部分母之外，矩阵不构成来源质量排名、权威证据、推荐或 GEO 成果。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-host-query-matrix-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度严格无品牌查询 × 观察主机出现矩阵（2026-07-29）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-host-query-matrix-2026-07-29.json" }, { "@type": "DataDownload", "name": "百度严格无品牌查询 × 观察主机出现矩阵 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-host-query-matrix-2026-07-29.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-source-surface-matrix-2026-07-29.json", "identifier": "ai-search-rag-baidu-source-surface-matrix-2026-07-29", "name": "百度严格无品牌查询 × 来源表面类型矩阵（2026-07-29）", "description": "从同一百度严格无品牌基线按预登记主机名后缀确定性派生 3 条查询 × 5 类来源表面及总体汇总共 20 行；25 个结果分为社区/专栏 8、厂商/云技术站 8、发布平台/门户 6、机构研究枢纽 2、其他观察主机 1。分类不确认发布者身份、编辑控制、质量、权威、推荐、因果排名因素或 GEO 成果。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-source-surface-matrix-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度严格无品牌查询 × 来源表面类型矩阵（2026-07-29）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-source-surface-matrix-2026-07-29.json" }, { "@type": "DataDownload", "name": "百度严格无品牌查询 × 来源表面类型矩阵 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-source-surface-matrix-2026-07-29.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-title-term-matrix-2026-07-29.json", "identifier": "ai-search-rag-baidu-title-term-matrix-2026-07-29", "name": "百度严格无品牌自然结果标题词覆盖矩阵（2026-07-29）", "description": "从同一百度严格无品牌基线确定性派生 25 个自然结果标题 × 7 个固定问题概念的显式零矩阵；RAG 命中 20 个标题、微调 11 个、引用 2 个、来源 0 个。字面命中不构成排名原因、来源质量、索引、正向搜索可见性或 GEO 成果。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-title-term-matrix-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度严格无品牌自然结果标题词覆盖矩阵（2026-07-29）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-title-term-matrix-2026-07-29.json" }, { "@type": "DataDownload", "name": "百度严格无品牌自然结果标题词覆盖矩阵 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-title-term-matrix-2026-07-29.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-evidence-join-2026-07-29.json", "identifier": "ai-search-rag-baidu-serp-evidence-join-2026-07-29", "name": "百度严格无品牌自然结果逐行联合证据表（2026-07-29）", "description": "按 queryId + position + url 唯一键，把同一冻结百度基线的 3 条严格查询、25 个自然位次、标题、URL、规范化主机、来源表面分类和 7 个标题概念状态合并为逐结果证据表；保留来源哈希与显式非成功边界。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-evidence-join-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度严格无品牌自然结果逐行联合证据表（2026-07-29）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-evidence-join-2026-07-29.json" }, { "@type": "DataDownload", "name": "百度严格无品牌自然结果逐行联合证据表 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-evidence-join-2026-07-29.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-churn-2026-07-29.json", "identifier": "ai-search-rag-baidu-serp-churn-2026-07-29", "name": "百度 RAG 与微调引用来源严格无品牌结果集波动（2026-07-29）", "description": "比较同一生产发布、同一固定严格无品牌查询套件在 2026-07-29 的两次只读观察：两次各有 25 个自然结果位次，22 个 URL 被保留、3 个新增、3 个移除，其中 2 个保留 URL 位次移动；DragonAI 目标页两次均未出现，结果集波动不构成目标排名变化或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-churn-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌结果集波动（2026-07-29）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-churn-2026-07-29.json" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌结果集波动 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-churn-2026-07-29.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-adjacent-churn-2026-07-30.json", "identifier": "ai-search-rag-baidu-serp-adjacent-churn-2026-07-30", "name": "百度 RAG 与微调引用来源严格无品牌第三至第四时点相邻区间波动（2026-07-30）", "description": "比较同一生产发布、同一固定严格无品牌查询套件的第三与第四次只读观察：两次各有 25 个自然结果位次，25 个 URL 全部留存、无新增或移除；第三条查询有 2 个 URL 在第 7、8 位互换，另外两条查询完全不变。DragonAI 目标页与 site 诊断仍未出现，竞争结果换位不构成目标排名变化、正向搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-adjacent-churn-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌第三至第四时点相邻区间波动（2026-07-30）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-adjacent-churn-2026-07-30.json" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌第三至第四时点相邻区间波动 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-adjacent-churn-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-fourth-to-fifth-churn-2026-07-30.json", "identifier": "ai-search-rag-baidu-serp-fourth-to-fifth-churn-2026-07-30", "name": "百度 RAG 与微调引用来源严格无品牌第四至第五时点相邻区间波动（2026-07-30）", "description": "比较同一生产发布、同一固定严格无品牌查询套件的第四与第五次只读观察：前两条查询结果集不变；第三条查询保留 6 个 URL、新增 2 个、移除 2 个，3 个保留 URL 位次后移。DragonAI 目标页与 site 诊断仍未出现，竞争结果更替不构成目标排名变化、正向搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-fourth-to-fifth-churn-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌第四至第五时点相邻区间波动（2026-07-30）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-fourth-to-fifth-churn-2026-07-30.json" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌第四至第五时点相邻区间波动 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-fourth-to-fifth-churn-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-four-observation-persistence-2026-07-30.json", "identifier": "ai-search-rag-baidu-serp-four-observation-persistence-2026-07-30", "name": "百度 RAG 与微调引用来源严格无品牌四时点 URL 持续性（2026-07-30）", "description": "合并同一生产发布与固定严格无品牌查询套件的四次百度自然结果观察：12 个查询快照共 100 个自然位次、29 个查询—URL 组合；22 个组合覆盖 4 / 4 次观察，其中 20 个四次位次完全相同，3 个组合在已出现窗口中存在位次变化。DragonAI 目标页四次均未出现，竞争 URL 持续或换位不构成排名因果、正向目标可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-four-observation-persistence-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌四时点 URL 持续性（2026-07-30）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-four-observation-persistence-2026-07-30.json" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌四时点 URL 持续性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-four-observation-persistence-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-fifth-to-sixth-churn-2026-07-30.json", "identifier": "ai-search-rag-baidu-serp-fifth-to-sixth-churn-2026-07-30", "name": "百度 RAG 与微调引用来源严格无品牌第五至第六时点相邻区间波动（2026-07-30）", "description": "比较同一生产发布、同一固定严格无品牌查询套件的第五与第六次只读观察：两次各有 25 个自然结果位次，共保留 23 个 URL、新增 2 个、移除 2 个；第一条查询另有 1 个保留 URL 从第 3 位移至第 2 位，第二条查询无保留 URL 换位，第三条查询完全不变。DragonAI 目标页与 site 诊断仍未出现，竞争结果更替不构成目标排名变化、正向搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-fifth-to-sixth-churn-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌第五至第六时点相邻区间波动（2026-07-30）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-fifth-to-sixth-churn-2026-07-30.json" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌第五至第六时点相邻区间波动 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-fifth-to-sixth-churn-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-cross-release-churn-2026-07-30.json", "identifier": "ai-search-rag-baidu-serp-cross-release-churn-2026-07-30", "name": "百度 RAG 与微调引用来源严格无品牌跨发布相邻结果变动（2026-07-30）", "description": "比较固定三条严格无品牌查询在相邻两次匿名百度自然结果观察中的变化：两次各有 25 个自然结果位次，共保留 19 个 URL、新增 6 个、移除 6 个，7 个保留 URL 位次变化。两次观察跨越两个生产发布，DragonAI 目标页均未出现；该数据不支持发布效果、排名因素、正向目标可见性或 GEO 成功归因。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-cross-release-churn-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌跨发布相邻结果变动（2026-07-30）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-cross-release-churn-2026-07-30.json" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌跨发布相邻结果变动 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-cross-release-churn-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-five-observation-persistence-2026-07-30.json", "identifier": "ai-search-rag-baidu-serp-five-observation-persistence-2026-07-30", "name": "百度 RAG 与微调引用来源严格无品牌五时点 URL 持续性（2026-07-30）", "description": "合并同一生产发布与固定严格无品牌查询套件的五次百度自然结果观察：15 个查询快照共 125 个自然位次、31 个查询—URL 组合；20 个组合覆盖 5 / 5 次观察，其中 15 个五次位次完全相同，6 个组合在已出现窗口中存在位次变化。DragonAI 目标页五次均未出现，竞争 URL 持续、进入、退出或换位不构成排名因果、正向目标可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-five-observation-persistence-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌五时点 URL 持续性（2026-07-30）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-five-observation-persistence-2026-07-30.json" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌五时点 URL 持续性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-five-observation-persistence-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-six-observation-persistence-2026-07-30.json", "identifier": "ai-search-rag-baidu-serp-six-observation-persistence-2026-07-30", "name": "百度 RAG 与微调引用来源严格无品牌六时点 URL 持续性（2026-07-30）", "description": "合并同一生产发布与固定严格无品牌查询套件的六次百度自然结果观察：18 个查询快照共 150 个自然位次、33 个查询—URL 组合；18 个组合覆盖 6 / 6 次观察，其中 12 个六次位次完全相同，7 个组合在已出现窗口中存在位次变化。DragonAI 目标页六次均未出现，竞争 URL 持续、进入、退出或换位不构成排名因果、正向目标可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-six-observation-persistence-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌六时点 URL 持续性（2026-07-30）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-six-observation-persistence-2026-07-30.json" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌六时点 URL 持续性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-six-observation-persistence-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-eight-observation-persistence-2026-07-30.json", "identifier": "ai-search-rag-baidu-serp-eight-observation-persistence-2026-07-30", "name": "百度 RAG 与微调引用来源严格无品牌八时点 URL 持续性（2026-07-30）", "description": "汇总三个生产发布上的八次匿名百度自然结果观察：24 个严格查询快照共 200 个自然结果位次，形成 36 个查询—URL 组合；16 个组合覆盖 8/8 次观察，其中 7 个八次位次完全相同，12 个组合在已出现窗口中发生位次变化。DragonAI 目标页八次均未出现；竞争 URL 持续、进入、退出或换位不构成排名因果、正向目标可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-eight-observation-persistence-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌八时点 URL 持续性（2026-07-30）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-eight-observation-persistence-2026-07-30.json" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌八时点 URL 持续性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-eight-observation-persistence-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-nine-observation-persistence-2026-07-30.json", "identifier": "ai-search-rag-baidu-serp-nine-observation-persistence-2026-07-30", "name": "百度 RAG 与微调引用来源严格无品牌九时点 URL 持续性（2026-07-30）", "description": "汇总三个生产发布上的九次匿名百度自然结果观察：27 个严格查询快照共 225 个自然结果位次，形成 37 个查询—URL 组合；15 个组合覆盖 9/9 次观察，其中 7 个九次位次完全相同，12 个组合在已出现窗口中发生位次变化。DragonAI 目标页九次均未出现；竞争 URL 持续、进入、退出或换位不构成排名因果、正向目标可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-nine-observation-persistence-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌九时点 URL 持续性（2026-07-30）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-nine-observation-persistence-2026-07-30.json" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌九时点 URL 持续性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-nine-observation-persistence-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-ten-observation-persistence-2026-07-31.json", "identifier": "ai-search-rag-baidu-serp-ten-observation-persistence-2026-07-31", "name": "百度 RAG 与微调引用来源严格无品牌十时点 URL 持续性（2026-07-31）", "description": "汇总三个生产发布上的十次匿名百度自然结果观察：30 个严格查询快照共 250 个自然结果位次，形成 39 个查询—URL 组合；14 个组合覆盖 10/10 次观察，其中 5 个十次位次完全相同，13 个组合在已出现窗口中发生位次变化。DragonAI 目标页十次均未出现；竞争 URL 持续、进入、退出或换位不构成排名因果、正向目标可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-ten-observation-persistence-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌十时点 URL 持续性（2026-07-31）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-ten-observation-persistence-2026-07-31.json" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌十时点 URL 持续性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-ten-observation-persistence-2026-07-31.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-ninth-to-tenth-churn-2026-07-31.json", "identifier": "ai-search-rag-baidu-serp-ninth-to-tenth-churn-2026-07-31", "name": "百度 RAG 与微调引用来源严格无品牌第九至第十时点相邻区间波动（2026-07-31）", "description": "比较同一生产发布、同一匿名百度自然搜索表面、同一三条严格无品牌查询的第九与第十次只读观察。25 个自然结果位次中保留 22 个，新增与移除各 3 个；只有 1 个保留 URL 发生位次变化。DragonAI 目标页在两次观察中均未出现，竞争结果波动不构成目标排名变化、正向搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-ninth-to-tenth-churn-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌第九至第十时点相邻区间波动（2026-07-31）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-ninth-to-tenth-churn-2026-07-31.json" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌第九至第十时点波动 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-ninth-to-tenth-churn-2026-07-31.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-eight-observation-2026-07-30.json", "identifier": "ai-search-rag-cross-surface-host-persistence-eight-observation-2026-07-30", "name": "千问、Kimi、Gemini 五批引用 × 百度八次严格无品牌结果跨表面主机持续性（2026-07-30）", "description": "将五批 45 次固定问题 AI 引用主机与三次生产发布上的八次、24 个百度严格无品牌查询快照按规范化主机全外联结。123 个联合主机中有 6 个跨表面重合，3 个至少在一家 AI 提供商覆盖全部五批，4 个至少有一个百度 URL 覆盖全部八次，只有 cloud.tencent.com 同时满足两项；DragonAI 在两侧均未出现。该交集不证明共享索引、来源质量、排名因果、品牌权威或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-eight-observation-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问、Kimi、Gemini 五批引用 × 百度八次严格无品牌结果跨表面主机持续性（2026-07-30）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-eight-observation-2026-07-30.json" }, { "@type": "DataDownload", "name": "五批 AI 引用 × 百度八次严格查询跨表面主机持续性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-eight-observation-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-eight-observation-answer-supplement-2026-07-30.json", "identifier": "ai-search-rag-eight-observation-answer-supplement-2026-07-30", "name": "RAG 与微调 AI 引用来源固定问题：八次百度观察机器答案补充（2026-07-30）", "description": "在不改写已版本化 v3.0 研究存档的前提下，把五批 45 次 AI 引用运行与跨三次生产发布的八次百度严格无品牌观察绑定为当前机器答案补充。四条无品牌问法、两条可引用声明和三份哈希绑定来源均显式保留零目标结果与非成功边界。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-eight-observation-answer-supplement-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "RAG 与微调 AI 引用来源固定问题：八次百度观察机器答案补充（2026-07-30）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-eight-observation-answer-supplement-2026-07-30.json" }, { "@type": "DataDownload", "name": "八次百度观察机器答案补充逐问法与声明 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-eight-observation-answer-supplement-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-top3-four-observation-evidence-2026-07-30.json", "identifier": "ai-search-rag-baidu-top3-four-observation-evidence-2026-07-30", "name": "百度 RAG 与微调引用来源四时点稳定前三位联结证据（2026-07-30）", "description": "把四次严格无品牌观察均保持同一前三位位置的 9 个查询—URL 记录，与标题词、来源表面、窗口 C AI 引用重合和只读文档交付特征逐 URL 联结。9 行均覆盖 4 / 4 次观察并保持同位；1 行精确 URL 也出现在 AI 引用中，5 行所在主机出现在 AI 引用中，3 个文档成功返回可分析 HTML。该联结不证明排名因果、来源质量、正向目标可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-top3-four-observation-evidence-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源四时点稳定前三位联结证据（2026-07-30）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-top3-four-observation-evidence-2026-07-30.json" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源四时点稳定前三位联结证据 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-top3-four-observation-evidence-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-observation-panel-2026-07-29.json", "identifier": "ai-search-rag-baidu-serp-observation-panel-2026-07-29", "name": "百度 RAG 与微调引用来源严格无品牌三时点观察面板（2026-07-29）", "description": "把同一匿名百度自然搜索表面、同一三条严格无品牌查询的三次只读观察按时点与相邻区间保存：共 75 个自然结果位次、29 个不同 URL；两段区间合计 45 个 URL 留存、5 个新增、5 个移除、2 个留存 URL 位次移动，DragonAI 目标页三次均未出现。竞争结果波动不构成目标排名变化或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-observation-panel-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌三时点观察面板（2026-07-29）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-observation-panel-2026-07-29.json" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌三时点观察面板 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-observation-panel-2026-07-29.csv" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌逐 URL 三时点持续性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-url-persistence-2026-07-29.csv" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌逐主机三时点持续性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-host-persistence-2026-07-29.csv" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌查询 × 主机三时点显式零矩阵 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-query-host-matrix-2026-07-29.csv" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌查询 × 主机相邻区间转移事件 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-query-host-transitions-2026-07-29.csv" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌查询 × 精确 URL 相邻区间转移事件 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-query-url-transitions-2026-07-29.csv" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌竞争主机转移汇总 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-host-transition-summary-2026-07-29.csv" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌主机倒数位次曝光 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-host-rank-exposure-2026-07-29.csv" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌逐 URL 倒数位次曝光 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-url-rank-exposure-2026-07-29.csv" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌来源表面倒数位次曝光 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-surface-rank-exposure-2026-07-29.csv" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌查询 × 来源表面倒数位次曝光矩阵 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-query-surface-rank-exposure-2026-07-29.csv" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌查询 × 主机倒数位次曝光矩阵 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-query-host-rank-exposure-2026-07-29.csv" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌查询 × URL 倒数位次曝光矩阵 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-query-url-rank-exposure-2026-07-29.csv" }, { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源严格无品牌查询快照主机集中度 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-query-snapshot-host-concentration-2026-07-29.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-ai-citation-host-overlap-2026-07-29.json", "identifier": "ai-search-rag-baidu-ai-citation-host-overlap-2026-07-29", "name": "百度严格无品牌自然结果主机 × 千问、Kimi、Gemini 引用主机交叉表（2026-07-29）", "description": "按小写、移除尾点与前导 www 的固定规则，把百度 3 条严格无品牌查询中的 14 个自然结果主机，与固定窗口 C 中千问、Kimi、Gemini 展示的 50 个引用主机做全外联结；60 个联合主机中只有 4 个跨表面重合，DragonAI 目标主机两侧均未出现。主机交集不构成索引共享、排名因果、来源质量、平台偏好或 GEO 成果。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-ai-citation-host-overlap-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度严格无品牌自然结果主机 × 千问、Kimi、Gemini 引用主机交叉表（2026-07-29）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-ai-citation-host-overlap-2026-07-29.json" }, { "@type": "DataDownload", "name": "百度严格无品牌自然结果主机 × 三平台引用主机交叉表 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-ai-citation-host-overlap-2026-07-29.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-ai-citation-url-overlap-2026-07-29.json", "identifier": "ai-search-rag-baidu-ai-citation-url-overlap-2026-07-29", "name": "百度严格无品牌自然结果 URL × 千问、Kimi、Gemini 引用 URL 交叉表（2026-07-29）", "description": "按小写协议与主机、端口、片段、尾斜杠规范化规则，把百度 3 条严格无品牌查询中的 25 个自然结果 URL，与固定窗口 C 中千问、Kimi、Gemini 展示的 74 个不同引用 URL 做全外联结；98 个联合 URL 中只有 1 个跨表面重合，DragonAI 目标 URL 两侧均未出现。URL 交集不构成共享索引、内容吸收、排名因果、来源质量、平台偏好或 GEO 成果。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-ai-citation-url-overlap-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度严格无品牌自然结果 URL × 千问、Kimi、Gemini 引用 URL 交叉表（2026-07-29）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-ai-citation-url-overlap-2026-07-29.json" }, { "@type": "DataDownload", "name": "百度严格无品牌自然结果 URL × 三平台引用 URL 交叉表 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-ai-citation-url-overlap-2026-07-29.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-competitive-result-evidence-2026-07-29.json", "identifier": "ai-search-rag-baidu-competitive-result-evidence-2026-07-29", "name": "百度严格无品牌当前竞争结果逐行证据表（2026-07-29）", "description": "把第二次生产观察中的 25 个当前百度严格无品牌自然结果，与结果集波动、来源表面、标题概念、AI 引用主机重合和精确 URL 重合合并成一张逐结果表。7 个当前结果的主机进入固定 AI 引用窗口，只有 1 个结果的精确 URL 重合；DragonAI 目标结果仍为 0，不能计作搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-competitive-result-evidence-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度严格无品牌当前竞争结果逐行证据表（2026-07-29）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-competitive-result-evidence-2026-07-29.json" }, { "@type": "DataDownload", "name": "百度严格无品牌当前竞争结果逐行证据表 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-competitive-result-evidence-2026-07-29.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-top3-document-features-2026-07-29.json", "identifier": "ai-search-rag-baidu-top3-document-features-2026-07-29", "name": "百度严格无品牌查询前三位竞争文档交付特征（2026-07-29）", "description": "对冻结百度严格查询最新观察中的 9 个前三位结果执行一次只读页面获取，完整保留 3 个可用文档特征、3 个 HTTP 或请求失败和 3 个仅返回 HTML 壳的行；记录标题摘要、canonical、标题层级、JSON-LD、链接和下载件计数。页面特征与前三位共现不构成排名因果、来源质量、推荐、目标可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-top3-document-features-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度严格无品牌查询前三位竞争文档交付特征（2026-07-29）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-top3-document-features-2026-07-29.json" }, { "@type": "DataDownload", "name": "百度严格无品牌查询前三位竞争文档交付特征 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-top3-document-features-2026-07-29.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-target-delivery-comparison-2026-07-29.json", "identifier": "ai-search-rag-baidu-target-delivery-comparison-2026-07-29", "name": "DragonAI 目标页与百度前三位可抓取文档交付特征对照（2026-07-29）", "description": "用同一只读 HTML 投影字段对照当前生产目标页与百度严格无品牌查询前三位中 3 个成功抓取的竞争文档，共 16 项元数据、结构、链接和下载特征。目标页在 8 项计数上高于全部 3 个可用竞争文档，但严格查询仍未观察到目标页；该对照不构成排名因果、来源质量、推荐或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-target-delivery-comparison-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "DragonAI 目标页与百度前三位可抓取文档交付特征对照（2026-07-29）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-target-delivery-comparison-2026-07-29.json" }, { "@type": "DataDownload", "name": "DragonAI 目标页与百度前三位可抓取文档交付特征对照 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-target-delivery-comparison-2026-07-29.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-citation-monitoring-protocol-v1.json", "identifier": "ai-search-citation-monitoring-protocol-v1", "name": "AI 搜索引用监测协议模板 v1", "description": "固定查询、平台与模型、联网设置、重复运行、引用归因、品牌提及和证据分层所需的机器可读字段与校验规则。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-monitoring-protocol-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-search-citation-monitoring-method#article", "url": "https://course.dragonai.tech/guides/ai-search-citation-monitoring-method" }, { "@type": "Article", "@id": "https://course.dragonai.tech/guides/how-ai-search-chooses-citation-sources#article", "url": "https://course.dragonai.tech/guides/how-ai-search-chooses-citation-sources" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 搜索引用监测协议模板 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-monitoring-protocol-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-citation-source-stability-metrics-v1.json", "identifier": "ai-search-citation-source-stability-metrics-v1", "name": "AI 搜索引用来源稳定性 4 项指标与复算样本 v1", "description": "公开引用展示运行覆盖率、重复 URL 占比、主机名 Jaccard 重叠和最高主机名集中度的公式、分子分母、比较要求与 9 次运行示例。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-source-stability-metrics-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-search-citation-source-stability-metrics#article", "url": "https://course.dragonai.tech/guides/ai-search-citation-source-stability-metrics" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 搜索引用来源稳定性 4 项指标与复算样本 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-source-stability-metrics-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-product-manager-project-case-study-example-v1.json", "identifier": "ai-product-manager-project-case-study-example-v1", "name": "AI 产品经理项目实战案例记录 v1", "description": "一个明确标注为虚构教学样例的客服知识库项目记录，覆盖问题卡、方案比较、行为合同、数据字段、评测门禁、失败复盘和受限发布决定。", "url": "https://course.dragonai.tech/datasets/ai-product-manager-project-case-study-example-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-product-manager-project-case-study#article", "url": "https://course.dragonai.tech/guides/ai-product-manager-project-case-study" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 产品经理项目实战案例记录 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-product-manager-project-case-study-example-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-agent-product-specification-card-v1.json", "identifier": "ai-agent-product-specification-card-v1", "name": "AI Agent 产品规格与上线门禁卡 v1", "description": "把智能体产品拆成必要性、任务边界、工具身份权限、人工控制、循环停止、行动轨迹评测和受限发布七部分，并提供空白复核记录。", "url": "https://course.dragonai.tech/datasets/ai-agent-product-specification-card-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-agent-product-manager#article", "url": "https://course.dragonai.tech/guides/ai-agent-product-manager" } ], "distribution": [ { "@type": "DataDownload", "name": "AI Agent 产品规格与上线门禁卡 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-agent-product-specification-card-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-product-requirement-evidence-chain-v1.json", "identifier": "ai-product-requirement-evidence-chain-v1", "name": "AI 产品需求证据链模板 v1", "description": "把原始需求证据连接到当前任务与基线、理想输出、方案比较、数据权限、失败切片和进入 PRD 的状态决定。", "url": "https://course.dragonai.tech/datasets/ai-product-requirement-evidence-chain-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-product-requirement-analysis#article", "url": "https://course.dragonai.tech/guides/ai-product-requirement-analysis" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 产品需求证据链模板 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-product-requirement-evidence-chain-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-product-model-selection-scorecard-v1.json", "identifier": "ai-product-model-selection-scorecard-v1", "name": "AI 产品大模型选型证据卡 v1", "description": "按任务合同、硬门槛、候选身份、实验控制、逐例质量、运行成本、供应治理和退出条件记录大模型选型证据。", "url": "https://course.dragonai.tech/datasets/ai-product-model-selection-scorecard-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-product-model-selection#article", "url": "https://course.dragonai.tech/guides/ai-product-model-selection" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 产品大模型选型证据卡 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-product-model-selection-scorecard-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-product-prototype-validation-record-v1.json", "identifier": "ai-product-prototype-validation-record-v1", "name": "AI 产品原型验证记录 v1", "description": "把验证问题、最低原型保真度、AI 行为合同、任务与失败切片、参与者协议、逐次运行和继续或停止决定连接起来。", "url": "https://course.dragonai.tech/datasets/ai-product-prototype-validation-record-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-product-prototype-validation#article", "url": "https://course.dragonai.tech/guides/ai-product-prototype-validation" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 产品原型验证记录 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-product-prototype-validation-record-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-product-canary-release-record-v1.json", "identifier": "ai-product-canary-release-record-v1", "name": "AI 产品灰度发布证据卡 v1", "description": "把发布单元、准入门槛、分群对照、AI 原生监控、逐阶段观测、放量或回滚决定和发布后责任连接起来。", "url": "https://course.dragonai.tech/datasets/ai-product-canary-release-record-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-product-canary-release#article", "url": "https://course.dragonai.tech/guides/ai-product-canary-release" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 产品灰度发布证据卡 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-product-canary-release-record-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-product-hallucination-governance-card-v1.json", "identifier": "ai-product-hallucination-governance-card-v1", "name": "AI 产品幻觉治理证据卡 v1", "description": "把风险边界、逐声明证据、评测切片、控制矩阵、运行事件、事故复盘和发布门禁连接起来。", "url": "https://course.dragonai.tech/datasets/ai-product-hallucination-governance-card-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-product-hallucination-governance#article", "url": "https://course.dragonai.tech/guides/ai-product-hallucination-governance" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 产品幻觉治理证据卡 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-product-hallucination-governance-card-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-product-cost-evidence-card-v1.json", "identifier": "ai-product-cost-evidence-card-v1", "name": "AI 产品成本与价值证据卡 v1", "description": "把成本范围、分摊规则、任务级用量、合格结果、全量单位成本、情景预测和价值决定连接起来。", "url": "https://course.dragonai.tech/datasets/ai-product-cost-evidence-card-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-product-cost-estimation#article", "url": "https://course.dragonai.tech/guides/ai-product-cost-estimation" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 产品成本与价值证据卡 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-product-cost-evidence-card-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/geo-monitoring-tool-due-diligence-checklist-v1.json", "identifier": "geo-monitoring-tool-due-diligence-checklist-v1", "name": "GEO 监测工具 12 项验收清单与 7 天试跑表 v1", "description": "用统一查询集核验 GEO 监测工具的产品表面、原生联网、逐次证据、失败保留、引用归因、分母政策、导出复算、版本变化和合规边界。", "url": "https://course.dragonai.tech/datasets/geo-monitoring-tool-due-diligence-checklist-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/how-to-choose-geo-monitoring-tool#article", "url": "https://course.dragonai.tech/guides/how-to-choose-geo-monitoring-tool" } ], "distribution": [ { "@type": "DataDownload", "name": "GEO 监测工具 12 项验收清单与 7 天试跑表 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/geo-monitoring-tool-due-diligence-checklist-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/geo-definition-implementation-evidence-map-v1.json", "identifier": "geo-definition-implementation-evidence-map-v1", "name": "GEO 定义、6 步实施与七层证据阶梯 v1", "description": "把生成式引擎优化拆成定义、六个实施阶段、七层证据、最小看板字段、非成功信号和禁止捷径，供内容与监测流程复用。", "url": "https://course.dragonai.tech/datasets/geo-definition-implementation-evidence-map-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/what-is-generative-engine-optimization#article", "url": "https://course.dragonai.tech/guides/what-is-generative-engine-optimization" } ], "distribution": [ { "@type": "DataDownload", "name": "GEO 定义、6 步实施与七层证据阶梯 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/geo-definition-implementation-evidence-map-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/geo-seo-overlap-decision-matrix-v1.json", "identifier": "geo-seo-overlap-decision-matrix-v1", "name": "GEO 与 SEO 8 维协同决策矩阵 v1", "description": "把 GEO 与 SEO 的共同基础、各自证据、错误推论、四级预算顺序、非成功信号和禁止手段拆成可复用机器字段。", "url": "https://course.dragonai.tech/datasets/geo-seo-overlap-decision-matrix-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/comparisons/geo-vs-seo#article", "url": "https://course.dragonai.tech/comparisons/geo-vs-seo" } ], "distribution": [ { "@type": "DataDownload", "name": "GEO 与 SEO 8 维协同决策矩阵 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/geo-seo-overlap-decision-matrix-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-publisher-readiness-checklist-v1.json", "identifier": "ai-search-publisher-readiness-checklist-v1", "name": "AI 搜索发布者 10 项就绪清单与 DragonAI 审计 v1", "description": "从公开访问、robots 用途分离、WAF、索引指令、canonical、服务端正文、发现入口、信息价值、真实爬虫到固定无品牌引用验证，逐层记录证据和不可推论边界。", "url": "https://course.dragonai.tech/datasets/ai-search-publisher-readiness-checklist-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/how-to-get-site-into-ai-search#article", "url": "https://course.dragonai.tech/guides/how-to-get-site-into-ai-search" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 搜索发布者 10 项就绪清单与 DragonAI 审计 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-publisher-readiness-checklist-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-product-manager-technical-glossary-v1.json", "identifier": "ai-product-manager-technical-glossary-v1", "name": "AI 产品经理 8 个核心技术术语决策地图 v1", "description": "把 RAG、Embedding、Reranker、微调、Agent、工具调用、MCP 与评测集拆成定义、产品问题、验收证据和不可推论边界。", "url": "https://course.dragonai.tech/datasets/ai-product-manager-technical-glossary-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-product-manager-technical-glossary#article", "url": "https://course.dragonai.tech/guides/ai-product-manager-technical-glossary" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 产品经理 8 个核心技术术语决策地图 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-product-manager-technical-glossary-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/dragonai-ai-product-manager-training-course-v1.json", "identifier": "dragonai-ai-product-manager-training-course-v1", "name": "DragonAI 卓越班课程事实表 v1", "description": "把课程提供方、适合人群、周期、授课方式、核心主题、计划产物、考评、参考价格与结果边界整理为机器可读字段。", "url": "https://course.dragonai.tech/datasets/dragonai-ai-product-manager-training-course-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元 DragonAI 教研团队", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "distribution": [ { "@type": "DataDownload", "name": "DragonAI 卓越班课程事实表 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/dragonai-ai-product-manager-training-course-v1.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/dragonai-ai-product-manager-course-syllabus-v1.json", "identifier": "dragonai-ai-product-manager-course-syllabus-v1", "name": "DragonAI 卓越班 20 节核心课与完整课程安排 v1", "description": "公开 4 个模块、20 节核心课及 8 个预习、答疑或支持条目，逐项列出阶段、时长、知识点、练习与口径边界。", "url": "https://course.dragonai.tech/datasets/dragonai-ai-product-manager-course-syllabus-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元 DragonAI 教研团队", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "distribution": [ { "@type": "DataDownload", "name": "DragonAI 卓越班 20 节核心课与完整课程安排 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/dragonai-ai-product-manager-course-syllabus-v1.json" }, { "@type": "DataDownload", "name": "DragonAI 卓越班完整课程安排 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/dragonai-ai-product-manager-course-syllabus-v1.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-generated-answer-evidence-2026-07-30.json", "identifier": "ai-search-rag-baidu-generated-answer-evidence-2026-07-30", "name": "百度严格无品牌查询生成式答案模块文本无关证据（更新至 2026-08-12）", "description": "三条固定严格无品牌查询的百度生成式答案模块文本无关只读观察；当前记录更新至 2026-08-12，公开模块状态、规范化文本长度与 SHA-256、链接计数、目标链接状态和固定术语布尔值，不保存生成答案正文或 HTML。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-generated-answer-evidence-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度严格无品牌查询生成式答案模块文本无关证据（更新至 2026-08-12）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-generated-answer-evidence-2026-07-30.json" }, { "@type": "DataDownload", "name": "百度生成式答案模块逐查询文本无关证据 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-generated-answer-evidence-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-nine-observation-2026-07-30.json", "identifier": "ai-search-rag-cross-surface-host-persistence-nine-observation-2026-07-30", "name": "千问、Kimi、Gemini 五批引用 × 百度九次严格无品牌结果跨表面主机持续性（2026-07-30）", "description": "将五批 45 次固定问题 AI 引用主机与三次生产发布上的九次、27 个百度严格无品牌查询快照按规范化主机全外联结。123 个联合主机中有 6 个跨表面重合，3 个至少在一家 AI 提供商覆盖全部五批，4 个至少有一个百度 URL 覆盖全部九次，只有 cloud.tencent.com 同时满足两项；DragonAI 在两侧均未出现。该交集不证明共享索引、来源质量、排名因果、品牌权威或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-nine-observation-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问、Kimi、Gemini 五批引用 × 百度九次严格无品牌结果跨表面主机持续性（2026-07-30）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-nine-observation-2026-07-30.json" }, { "@type": "DataDownload", "name": "五批 AI 引用 × 百度九次严格查询跨表面主机持续性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-nine-observation-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-url-persistence-nine-observation-2026-07-31.json", "identifier": "ai-search-rag-cross-surface-url-persistence-nine-observation-2026-07-31", "name": "千问、Kimi、Gemini 五批引用 URL × 百度九次严格无品牌结果逐 URL 持久性交叉表", "description": "把五批 45 次 AI 运行中的 194 个具体引用 URL，与百度三条严格无品牌查询九次观察中的 35 个不同自然结果 URL 按固定规范化规则全外联结。227 个联合 URL 中只有智源社区与知乎专栏各一个具体页面同时出现，但没有 URL 同时覆盖 AI 全五批与百度全九次；DragonAI 两侧仍为零。该交集不证明共享索引、来源质量、排名因果或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-url-persistence-nine-observation-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问、Kimi、Gemini 五批引用 URL × 百度九次严格无品牌结果逐 URL 持久性交叉表", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-url-persistence-nine-observation-2026-07-31.json" }, { "@type": "DataDownload", "name": "五批 AI 引用 URL × 百度九次严格结果逐 URL 持久性交叉表 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-url-persistence-nine-observation-2026-07-31.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-source-type-summary-nine-observation-2026-07-30.json", "identifier": "ai-search-rag-cross-surface-source-type-summary-nine-observation-2026-07-30", "name": "五批 AI 引用 × 百度九次严格结果跨表面来源类型汇总（2026-07-30）", "description": "将九次口径的 123 个跨表面联合主机压缩为 6 个互斥来源类型。6 个重合主机中厂商文档 3 个、专家社区 2 个、其他实践者或出版者 1 个；唯一双侧持续主机 cloud.tencent.com 属于厂商文档。九次百度观察共保留 225 个自然结果位次和 81 个前三位位次；DragonAI 两侧仍未出现，类型计数不构成来源质量、排名因果或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-source-type-summary-nine-observation-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "五批 AI 引用 × 百度九次严格结果跨表面来源类型汇总（2026-07-30）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-source-type-summary-nine-observation-2026-07-30.json" }, { "@type": "DataDownload", "name": "五批 AI 引用 × 百度九次严格结果跨表面来源类型 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-source-type-summary-nine-observation-2026-07-30.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-post-deploy-baseline-2026-07-31.json", "identifier": "ai-search-rag-baidu-post-deploy-baseline-2026-07-31", "name": "百度 RAG 与微调引用来源新生产发布后严格无品牌基线（2026-07-31）", "description": "绑定生产发布 20260731T130258Z-c8abe8ac79d6，公开三条严格无品牌查询与一条 site 诊断查询的匿名百度自然结果、逐位次 URL、结果集哈希及显式零目标结果。三条严格查询共保留 25 个自然结果位次，DragonAI 目标页与 site 诊断均未出现；该数据集不构成收录、引用、排名原因或 GEO 成功证据。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-post-deploy-baseline-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源新生产发布后严格无品牌基线（2026-07-31）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-post-deploy-baseline-2026-07-31.json" }, { "@type": "DataDownload", "name": "新生产发布后百度严格无品牌逐查询逐位次 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-post-deploy-baseline-2026-07-31.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-release-baseline-2026-08-01.json", "identifier": "ai-search-rag-baidu-current-release-baseline-2026-08-01", "name": "百度 RAG 与微调来源当前严格无品牌快照与相邻变动（2026-08-01）", "description": "绑定生产发布 20260801T154118Z-4215d0a，公开三条严格无品牌查询与一条 site 诊断查询的匿名百度自然结果、相邻结果变动，并用逐报告 SHA-256 汇总跨七次生产发布的 18 次可比观察、54 个严格查询快照与 449 个自然结果位次。当前三条严格查询共 25 个自然位次，DragonAI 目标页及品牌域均未出现；该数据集不证明发布效果、排名原因或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-release-baseline-2026-08-01.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调来源当前严格无品牌快照与相邻变动（2026-08-01）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-release-baseline-2026-08-01.json" }, { "@type": "DataDownload", "name": "当前生产发布百度严格无品牌逐查询逐位次 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-release-baseline-2026-08-01.csv" }, { "@type": "DataDownload", "name": "18 次可比百度观察逐报告证据账本 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-release-baseline-2026-08-01-cumulative-observations.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/baidu-serp-entry-liveness-2026-08-01t105842z.json", "identifier": "baidu-serp-entry-liveness-2026-08-01t105842z", "name": "百度严格无品牌结果新进入 URL 存活性诊断（2026-08-01）", "description": "对当前生产发布与前一可比观察之间新进入结果集的 2 个 URL 做匿名只读抓取，记录 HTTP 存活性、响应指纹和页面结构。该诊断不改写排名，也不构成 DragonAI 搜索可见性、来源质量或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/baidu-serp-entry-liveness-2026-08-01t105842z.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "distribution": [ { "@type": "DataDownload", "name": "百度严格无品牌结果新进入 URL 存活性诊断（2026-08-01）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/baidu-serp-entry-liveness-2026-08-01t105842z.json" }, { "@type": "DataDownload", "name": "百度新进入结果 URL 存活性诊断 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/baidu-serp-entry-liveness-2026-08-01t105842z.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-01t173506z.json", "identifier": "ai-search-rag-baidu-serp-entry-liveness-2026-08-01t173506z", "name": "百度固定无品牌结果新进入 URL 存活诊断（2026-08-01 17:35 UTC）", "description": "对当前生产发布与前一可比观察之间新增的 5 个百度严格无品牌结果 URL 做匿名只读存活诊断，并保留 5 个退出 URL；该数据不改写排名，也不构成 DragonAI 搜索可见性、来源质量、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-01t173506z.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度固定无品牌结果新进入 URL 存活诊断（2026-08-01 17:35 UTC）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-01t173506z.json" }, { "@type": "DataDownload", "name": "百度固定无品牌结果新进入 URL 存活诊断 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-01t173506z.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-01.json", "identifier": "ai-search-rag-current-release-discovery-gap-2026-08-01", "name": "当前生产发布 Bingbot 发现面至研究目标页抓取缺口（2026-08-01）", "description": "绑定生产发布 20260801T104826Z-1b8a06a386d9 的受控访问日志与只读可发现性诊断：Bingbot 已成功抓取首页，9 个发现面均直接引用唯一研究目标页，目标页对 Googlebot、Bingbot 与 Baiduspider 的交付、canonical、robots、结构化数据及条件请求均通过，但目标页合格抓取仍为 0。该数据集不证明收录、排名、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-01.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "distribution": [ { "@type": "DataDownload", "name": "当前生产发布 Bingbot 发现面至研究目标页抓取缺口（2026-08-01）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-01.json" }, { "@type": "DataDownload", "name": "当前生产发布发现面至目标页抓取缺口 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-01.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r4.json", "identifier": "ai-search-rag-current-release-discovery-gap-2026-08-14-r4", "name": "Bingbot research-sitemap-to-target crawl gap (current release, revision 4)", "description": "Privacy-minimized evidence for release 20260813T210114Z-75d814fc4b80: verified Bingbot fetched sitemap.xml, research-sitemap.xml and the insights hub, each of which references the exact research target, while the target still had zero qualifying Googlebot or Bingbot fetches. All monitored indexability checks pass. This is discovery evidence, not indexing, ranking, citation or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r4.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bingbot research-sitemap-to-target crawl gap (current release, revision 4)", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r4.json" }, { "@type": "DataDownload", "name": "Current-release research-sitemap-to-target gap revision 4 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r4.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r5.json", "identifier": "ai-search-rag-current-release-discovery-gap-2026-08-14-r5", "name": "Bingbot six-surface-to-target crawl gap (current release, revision 5)", "description": "Privacy-minimized evidence for release 20260813T210114Z-75d814fc4b80: verified Bingbot fetched six target-referencing discovery surfaces across XML, feed, hub and plain-text formats, while the exact research target still had zero qualifying Googlebot or Bingbot fetches. All monitored indexability checks pass. This is discovery evidence, not indexing, ranking, citation or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r5.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "distribution": [ { "@type": "DataDownload", "name": "Bingbot six-surface-to-target crawl gap (current release, revision 5)", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r5.json" }, { "@type": "DataDownload", "name": "Current-release six-surface-to-target gap revision 5 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r5.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r6.json", "identifier": "ai-search-rag-current-release-discovery-gap-2026-08-14-r6", "name": "Bingbot six-surface and spoof-excluded target crawl gap (current release, revision 6)", "description": "Privacy-minimized evidence for release 20260813T210114Z-75d814fc4b80: verified Bingbot fetched six target-referencing discovery surfaces, while one exact-target request claiming Bingbot was excluded after DNS hostname verification failed. The verified target crawl count remains zero. This is delivery and exclusion evidence, not indexing, ranking, citation or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r6.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "distribution": [ { "@type": "DataDownload", "name": "Bingbot six-surface and spoof-excluded target crawl gap (current release, revision 6)", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r6.json" }, { "@type": "DataDownload", "name": "Current-release spoof-excluded target crawl gap revision 6 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r6.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-12-r2.json", "identifier": "ai-search-rag-current-release-discovery-gap-2026-08-12-r2", "name": "Bingbot discovery-to-target crawl gap (current release, revision 2)", "description": "Privacy-minimized evidence for release 20260812T141218Z-3c5d13e24947: verified Bingbot fetched both the homepage and the target-only research-targets.txt surface, while the research target still had zero qualifying crawler fetches. This is discovery evidence, not indexing, ranking, citation or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-12-r2.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bingbot discovery-to-target crawl gap (current release, revision 2)", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-12-r2.json" }, { "@type": "DataDownload", "name": "Current-release discovery-to-target gap revision 2 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-12-r2.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-12-r3.json", "identifier": "ai-search-rag-current-release-discovery-gap-2026-08-12-r3", "name": "Bingbot sitemap-to-target crawl gap (current release, revision 3)", "description": "Privacy-minimized evidence for release 20260812T141218Z-3c5d13e24947: verified Bingbot fetched the homepage, research-sitemap.xml and research-targets.txt while the research target still had zero qualifying crawler fetches. This is discovery evidence, not indexing, ranking, citation or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-12-r3.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bingbot sitemap-to-target crawl gap (current release, revision 3)", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-12-r3.json" }, { "@type": "DataDownload", "name": "Current-release sitemap-to-target gap revision 3 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-12-r3.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-title-intent-matrix-2026-08-01.json", "identifier": "ai-search-rag-baidu-current-title-intent-matrix-2026-08-01", "name": "百度当前严格无品牌结果标题意图矩阵（2026-08-01）", "description": "把当前生产发布对应的 25 个百度严格无品牌自然结果标题，按引用或来源、平台名、选型或落地、研究或证据、GEO 五组预登记字面信号逐行投影。当前仅 1 个标题明确出现引用或来源信号，13 个强调选型或落地，3 个出现研究或证据信号；该矩阵不识别排名原因，也不构成 DragonAI 搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-title-intent-matrix-2026-08-01.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "distribution": [ { "@type": "DataDownload", "name": "百度当前严格无品牌结果标题意图矩阵（2026-08-01）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-title-intent-matrix-2026-08-01.json" }, { "@type": "DataDownload", "name": "百度当前结果标题意图逐位次 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-title-intent-matrix-2026-08-01.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r7.json", "identifier": "ai-search-rag-current-release-discovery-gap-2026-08-14-r7", "name": "Bingbot seven-surface and spoof-excluded target crawl gap (current release, revision 7)", "description": "Privacy-minimized evidence for release 20260813T210114Z-75d814fc4b80: verified Bingbot made ten successful fetches across seven target-referencing discovery surfaces, while one exact-target request claiming Bingbot was excluded after DNS hostname verification failed. The verified target crawl count remains zero. This is delivery and exclusion evidence, not indexing, ranking, citation or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r7.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "distribution": [ { "@type": "DataDownload", "name": "Bingbot seven-surface and spoof-excluded target crawl gap (current release, revision 7)", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r7.json" }, { "@type": "DataDownload", "name": "Current-release spoof-excluded target crawl gap revision 7 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r7.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-tenth-to-post-deploy-delta-2026-07-31.json", "identifier": "ai-search-rag-baidu-tenth-to-post-deploy-delta-2026-07-31", "name": "百度 RAG 与微调引用来源第十次观察至新生产发布后严格无品牌差分（2026-07-31）", "description": "比较上一生产发布的第十次严格无品牌观察与当前生产发布上线后同口径匿名百度自然结果。三条查询各保留 25 个位次，逐 URL 共保留 21 个、新增 4 个、移出 4 个、4 个保留 URL 换位；DragonAI 目标页在两次观察中均未出现。该差分不证明发布导致排名变化、目标收录、引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-tenth-to-post-deploy-delta-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源第十次观察至新生产发布后严格无品牌差分（2026-07-31）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-tenth-to-post-deploy-delta-2026-07-31.json" }, { "@type": "DataDownload", "name": "第十次观察至当前生产发布后逐查询逐 URL 差分 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-tenth-to-post-deploy-delta-2026-07-31.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-post-deploy-same-release-delta-2026-07-31.json", "identifier": "ai-search-rag-baidu-post-deploy-same-release-delta-2026-07-31", "name": "百度 RAG 与微调引用来源同发布相邻严格查询差分（2026-07-31）", "description": "比较当前生产发布上线后 14:35 与 16:47 UTC 的两次同口径匿名百度自然结果。三条严格无品牌查询各保留 25 个位次；逐 URL 共保留 22 个、新增 3 个、移出 3 个，9 个保留 URL 换位。腾讯云页面进入选型来源查询第 1 位，但 DragonAI 目标页和 site 诊断仍为 0；该数据集不识别排名原因，也不构成正向搜索可见性或 GEO 成功证据。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-post-deploy-same-release-delta-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源同发布相邻严格查询差分（2026-07-31）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-post-deploy-same-release-delta-2026-07-31.json" }, { "@type": "DataDownload", "name": "同发布相邻百度严格无品牌逐 URL 差分 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-post-deploy-same-release-delta-2026-07-31.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-post-deploy-host-query-matrix-2026-07-31.json", "identifier": "ai-search-rag-baidu-post-deploy-host-query-matrix-2026-07-31", "name": "百度 RAG 与微调引用来源发布后精确主机 × 严格查询矩阵（2026-07-31）", "description": "按 URL 的精确小写主机名汇总当前生产发布后 16:47 UTC 的三条匿名百度严格无品牌查询，不合并子域或企业家族。25 个自然结果涉及 11 个精确主机；只有 baijiahao.baidu.com 与 blog.csdn.net 覆盖全部三条查询，cloud.tencent.com 覆盖两条查询且其 4 个结果均在前三位。DragonAI 目标页仍为 0；覆盖广度不代表来源质量、权威、长期偏好、排名原因或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-post-deploy-host-query-matrix-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "百度 RAG 与微调引用来源发布后精确主机 × 严格查询矩阵（2026-07-31）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-post-deploy-host-query-matrix-2026-07-31.json" }, { "@type": "DataDownload", "name": "百度发布后精确主机 × 严格查询矩阵 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-post-deploy-host-query-matrix-2026-07-31.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-exact-host-overlap-2026-07-31.json", "identifier": "ai-search-rag-current-baidu-ai-exact-host-overlap-2026-07-31", "name": "当前百度精确主机 × 五批 AI 引用主机重合矩阵（2026-07-31）", "description": "把当前生产发布 16:47 UTC 百度三条严格无品牌查询中的 11 个精确小写 URL 主机，左联结到千问、Kimi、Gemini 五批 45 次运行的 114 个精确引用主机，并为每个百度主机保留三家提供商显式零值。精确重合主机为 CSDN 博客、腾讯云、智源社区和知乎专栏 4 个；DragonAI 两侧均未出现。该矩阵不是来源质量评分、渠道推荐、共享索引证据、排名原因或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-exact-host-overlap-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "当前百度精确主机 × 五批 AI 引用主机重合矩阵（2026-07-31）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-exact-host-overlap-2026-07-31.json" }, { "@type": "DataDownload", "name": "当前百度 × 五批 AI 精确主机重合 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-exact-host-overlap-2026-07-31.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-post-deploy-query-evidence-cards-2026-07-31.json", "identifier": "ai-search-rag-post-deploy-query-evidence-cards-2026-07-31", "name": "当前生产发布 RAG 与微调引用来源逐查询检索证据卡（2026-07-31）", "description": "将当前生产发布后的三条百度严格无品牌查询快照与上一发布第十次观察的逐 URL 差分，压缩为三张查询证据卡。每张卡保留可直接回答的来源选择方法、当前前三结果、结果集差分、目标零结果、来源定位和禁止推断边界；共覆盖 6 条检索问法、25 个当前自然结果和 9 个前三位证据，不构成正向搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-post-deploy-query-evidence-cards-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "当前生产发布 RAG 与微调引用来源逐查询检索证据卡（2026-07-31）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-post-deploy-query-evidence-cards-2026-07-31.json" }, { "@type": "DataDownload", "name": "当前生产发布逐查询检索证据卡 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-post-deploy-query-evidence-cards-2026-07-31.csv" }, { "@type": "DataDownload", "name": "当前生产发布逐查询检索证据卡 Markdown", "encodingFormat": "text/markdown", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-post-deploy-query-evidence-cards-2026-07-31.md" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-ten-observation-2026-07-31.json", "identifier": "ai-search-rag-cross-surface-host-persistence-ten-observation-2026-07-31", "name": "五批中国 AI 引用 × 百度十次严格无品牌结果跨表面主机持久性（2026-07-31）", "description": "五批 45 次千问、Kimi、Gemini 固定问题引用与百度三条严格无品牌查询十次观察的主机级全外联结。124 个联合主机中仍有 6 个跨表面重合，只有 cloud.tencent.com 双侧持续；新增 OSCHINA 仅扩大百度单侧集合。DragonAI 在两侧均未出现。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-ten-observation-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "五批中国 AI 引用 × 百度十次严格无品牌结果跨表面主机持久性（2026-07-31）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-ten-observation-2026-07-31.json" }, { "@type": "DataDownload", "name": "五批 AI 引用 × 百度十次严格查询跨表面主机持久性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-ten-observation-2026-07-31.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-source-type-summary-ten-observation-2026-07-31.json", "identifier": "ai-search-rag-cross-surface-source-type-summary-ten-observation-2026-07-31", "name": "五批中国 AI 引用 × 百度十次严格无品牌结果跨表面来源类型汇总（2026-07-31）", "description": "将十次口径的 124 个跨表面联合主机压缩为 6 个互斥来源类型。6 个重合主机的来源类型构成及唯一双侧持续主机均未改变；OSCHINA 只进入百度单侧未分类分母。DragonAI 在两侧均未出现。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-source-type-summary-ten-observation-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "五批中国 AI 引用 × 百度十次严格无品牌结果跨表面来源类型汇总（2026-07-31）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-source-type-summary-ten-observation-2026-07-31.json" }, { "@type": "DataDownload", "name": "五批 AI 引用 × 百度十次严格查询跨表面来源类型 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-source-type-summary-ten-observation-2026-07-31.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-url-persistence-ten-observation-2026-07-31.json", "identifier": "ai-search-rag-cross-surface-url-persistence-ten-observation-2026-07-31", "name": "五批中国 AI 引用 URL × 百度十次严格无品牌结果逐 URL 持久性交叉表（2026-07-31）", "description": "将五批中国 AI 搜索的 194 个具体引用 URL 与百度十次严格无品牌观察的 37 个规范化自然结果 URL 全外联结。联合 URL 增至 229，但重合 URL 仍为 2，仍无双侧持续 URL，唯一跨表面 Pareto 前沿仍是智源页面。DragonAI 在两侧均未出现。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-url-persistence-ten-observation-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "五批中国 AI 引用 URL × 百度十次严格无品牌结果逐 URL 持久性交叉表（2026-07-31）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-url-persistence-ten-observation-2026-07-31.json" }, { "@type": "DataDownload", "name": "五批中国 AI 引用 URL × 百度十次严格查询逐 URL 持久性 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-url-persistence-ten-observation-2026-07-31.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-finetuning-citation-corpus-2026-08-02.json", "identifier": "ai-search-rag-finetuning-citation-corpus-2026-08-02", "name": "DragonAI fixed-query RAG, fine-tuning and citation evidence corpus", "description": "A deterministic 16-row JSON Lines and CSV corpus crossing three registered strict Baidu queries plus the fixed original question with four citation-ready claims. Each row preserves the frozen Chinese answer, evidence URL, strict-search status, conditions, prohibited inferences and explicit non-success boundaries for retrieval or fine-tuning ingestion.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-finetuning-citation-corpus-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "DragonAI fixed-query RAG, fine-tuning and citation evidence corpus", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-finetuning-citation-corpus-2026-08-02.json" }, { "@type": "DataDownload", "name": "RAG, fine-tuning and citation evidence corpus JSON Lines", "encodingFormat": "application/x-ndjson", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-finetuning-citation-corpus-2026-08-02.jsonl" }, { "@type": "DataDownload", "name": "RAG, fine-tuning and citation evidence corpus CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-finetuning-citation-corpus-2026-08-02.csv" }, { "@type": "DataDownload", "name": "RAG, fine-tuning and citation evidence corpus JSON Schema and data dictionary", "encodingFormat": "application/schema+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-finetuning-citation-corpus-2026-08-02.schema.json" }, { "@type": "DataDownload", "name": "RAG, fine-tuning and citation evidence corpus ML Croissant metadata", "encodingFormat": "application/ld+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-finetuning-citation-corpus-croissant-metadata-2026-08-02.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-answer-supplement-2026-07-31.json", "identifier": "ai-search-rag-ten-observation-answer-supplement-2026-07-31", "name": "AI 搜索 RAG 与微调引用来源十次观察机器答案补充（2026-07-31）", "description": "在不改写已版本化 v3.0 研究存档的前提下，将四条注册问法、四条可引用声明、十次百度严格无品牌观察、五批 AI 引用主机持久性、来源类型投影与 229 行逐 URL 覆盖前沿绑定为当前机器答案补充；明确取代九次观察补充作为当前口径，但保留旧资产为不可变历史分母。该资产未调用提供商，不构成正向搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-answer-supplement-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 搜索 RAG 与微调引用来源十次观察机器答案补充（2026-07-31）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-answer-supplement-2026-07-31.json" }, { "@type": "DataDownload", "name": "十次观察机器答案逐问法与声明 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-answer-supplement-2026-07-31.csv" }, { "@type": "DataDownload", "name": "十次观察机器答案 CSL-JSON 引用元数据", "encodingFormat": "application/vnd.citationstyles.csl+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-answer-supplement-2026-07-31.csl.json" }, { "@type": "DataDownload", "name": "十次观察机器答案 Citation File Format 引用元数据", "encodingFormat": "text/yaml", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-answer-supplement-2026-07-31.cff" }, { "@type": "DataDownload", "name": "十次观察机器答案 BibTeX 引用元数据", "encodingFormat": "application/x-bibtex", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-answer-supplement-2026-07-31.bib" }, { "@type": "DataDownload", "name": "十次观察机器答案 RIS 引用元数据", "encodingFormat": "application/x-research-info-systems", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-answer-supplement-2026-07-31.ris" }, { "@type": "DataDownload", "name": "Ten-observation current answer JSON-LD knowledge graph", "encodingFormat": "application/ld+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-answer-supplement-2026-07-31.jsonld" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-retrieval-evidence-cards-2026-07-31.json", "identifier": "ai-search-rag-ten-observation-retrieval-evidence-cards-2026-07-31", "name": "AI 搜索 RAG 与微调引用来源十次观察检索证据卡（2026-07-31）", "description": "将十次观察机器答案补充确定性投影为四张逐查询检索证据卡，分别回答提供商来源、跨表面重合、选型问法持续来源以及未进入严格百度套件的边界；每张卡保留四条可引用声明、证据定位、必要条件、禁止推断与显式非成功边界。未新增提供商调用。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-retrieval-evidence-cards-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 搜索 RAG 与微调引用来源十次观察检索证据卡（2026-07-31）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-retrieval-evidence-cards-2026-07-31.json" }, { "@type": "DataDownload", "name": "十次观察检索证据卡扁平 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-retrieval-evidence-cards-2026-07-31.csv" }, { "@type": "DataDownload", "name": "十次观察检索证据卡 Markdown", "encodingFormat": "text/markdown", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-retrieval-evidence-cards-2026-07-31.md" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-2026-07-31.json", "identifier": "ai-search-rag-ten-observation-current-evidence-2026-07-31", "name": "AI 搜索 RAG 与微调引用来源十次观察证据链存档（原始已部署版本）", "description": "保留生产发布 20260731T130258Z-c8abe8ac79d6 已公开的原始十次观察证据链字节、清单与 Linkset，作为修订 2 的不可变 predecessor-version；该历史版本不再作为当前目标页的 latest-version，也不构成外链、搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 搜索 RAG 与微调引用来源十次观察证据链存档（原始已部署版本）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-2026-07-31.json" }, { "@type": "DataDownload", "name": "原始十次观察证据链离线验证 ZIP", "encodingFormat": "application/zip", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-2026-07-31.zip" }, { "@type": "DataDownload", "name": "原始十次观察证据链 RFC 9264 Linkset", "encodingFormat": "application/linkset+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-linkset-2026-07-31.json" }, { "@type": "DataDownload", "name": "原始十次观察证据链机器清单", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-manifest-2026-07-31.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-02-r9.json", "identifier": "ai-search-rag-current-evidence-2026-08-02-r9", "name": "AI 搜索 RAG 与微调引用来源当前证据链修订 9（2026-08-02）", "description": "不可变修订 9 将修订 8 声明为前序版本，纳入跨 15 个生产版本的 26 次严格百度观察、78 个查询快照、649 个自然结果位置，以及第 26 次观察的相邻波动、查询—URL、查询—主机、持续前三来源卡与逐查询答案证据。确定性 ZIP 含 50 个来源资产和 57 个归档文件；未调用提供商或执行站外发布，不构成搜索可见性、外链、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-02-r9.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 搜索 RAG 与微调引用来源当前证据链修订 9（2026-08-02）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-02-r9.json" }, { "@type": "DataDownload", "name": "当前证据链修订 9 离线验证 ZIP", "encodingFormat": "application/zip", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-02-r9.zip" }, { "@type": "DataDownload", "name": "当前证据链修订 9 RFC 9264 Linkset", "encodingFormat": "application/linkset+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-linkset-2026-08-02-r9.json" }, { "@type": "DataDownload", "name": "当前证据链修订 9 机器清单", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-manifest-2026-08-02-r9.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-02-r8.json", "identifier": "ai-search-rag-current-evidence-2026-08-02-r8", "name": "AI 搜索 RAG 与微调引用来源当前证据链修订 8（2026-08-02）", "description": "不可变修订 8 将修订 7 声明为前序版本，纳入跨 12 个生产版本的 23 次严格百度观察、69 个查询快照、574 个自然结果位置，以及第 23 次观察的查询—URL、查询—主机、持续前三来源卡、答案卡和只读存活证据。确定性 ZIP 含 50 个来源资产和 57 个归档文件；未调用提供商或执行站外发布，不构成搜索可见性、外链、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-02-r8.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 搜索 RAG 与微调引用来源当前证据链修订 8（2026-08-02）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-02-r8.json" }, { "@type": "DataDownload", "name": "当前证据链修订 8 离线验证 ZIP", "encodingFormat": "application/zip", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-02-r8.zip" }, { "@type": "DataDownload", "name": "当前证据链修订 8 RFC 9264 Linkset", "encodingFormat": "application/linkset+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-linkset-2026-08-02-r8.json" }, { "@type": "DataDownload", "name": "当前证据链修订 8 机器清单", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-manifest-2026-08-02-r8.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r7.json", "identifier": "ai-search-rag-current-evidence-2026-08-01-r7", "name": "AI 搜索 RAG 与微调引用来源当前证据链修订 7（2026-08-01）", "description": "不可变修订 7 将修订 6 声明为前序版本，纳入跨 11 个生产版本的 22 次严格百度观察、66 个查询快照、549 个自然结果位置，以及第 22 次观察的新增 URL 存活证据。确定性 ZIP 含 41 个来源资产和 48 个归档文件；未调用提供商或执行站外发布，不构成搜索可见性、外链、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r7.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 搜索 RAG 与微调引用来源当前证据链修订 7（2026-08-01）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r7.json" }, { "@type": "DataDownload", "name": "当前证据链修订 7 离线验证 ZIP", "encodingFormat": "application/zip", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r7.zip" }, { "@type": "DataDownload", "name": "当前证据链修订 7 RFC 9264 Linkset", "encodingFormat": "application/linkset+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-linkset-2026-08-01-r7.json" }, { "@type": "DataDownload", "name": "当前证据链修订 7 机器清单", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-manifest-2026-08-01-r7.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r6.json", "identifier": "ai-search-rag-current-evidence-2026-08-01-r6", "name": "AI 搜索 RAG 与微调引用来源当前证据链修订 6（2026-08-01）", "description": "不可变修订 6 将修订 5 声明为前序版本，纳入跨 10 个生产版本的 21 次严格百度观察、63 个查询快照、524 个自然结果位置，以及第 21 次观察的新增 URL 存活证据。确定性 ZIP 含 41 个来源资产和 48 个归档文件；未调用提供商或执行站外发布，不构成搜索可见性、外链、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r6.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 搜索 RAG 与微调引用来源当前证据链修订 6（2026-08-01）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r6.json" }, { "@type": "DataDownload", "name": "当前证据链修订 6 离线验证 ZIP", "encodingFormat": "application/zip", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r6.zip" }, { "@type": "DataDownload", "name": "当前证据链修订 6 RFC 9264 Linkset", "encodingFormat": "application/linkset+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-linkset-2026-08-01-r6.json" }, { "@type": "DataDownload", "name": "当前证据链修订 6 机器清单", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-manifest-2026-08-01-r6.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r5.json", "identifier": "ai-search-rag-current-evidence-2026-08-01-r5", "name": "AI search RAG and fine-tuning citation current evidence archive revision 5 (2026-08-01)", "description": "Immutable revision 5 preserves revision 4 as its predecessor, updates the strict unbranded search ledger to 18 comparable observations, and adds a bidirectional RFC 9264 relationship for the stable current-answer endpoint. The deterministic ZIP contains 37 source assets and 44 archive files with offline checksums. It made no provider calls or offsite publication and does not count as search visibility, external-link evidence, AI citation, or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r5.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "distribution": [ { "@type": "DataDownload", "name": "AI search RAG and fine-tuning citation current evidence archive revision 5 (2026-08-01)", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r5.json" }, { "@type": "DataDownload", "name": "Current evidence revision 5 offline-verifiable ZIP", "encodingFormat": "application/zip", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r5.zip" }, { "@type": "DataDownload", "name": "Current evidence revision 5 RFC 9264 Linkset", "encodingFormat": "application/linkset+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-linkset-2026-08-01-r5.json" }, { "@type": "DataDownload", "name": "Current evidence revision 5 machine manifest", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-manifest-2026-08-01-r5.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r4.json", "identifier": "ai-search-rag-current-evidence-2026-08-01-r4", "name": "AI 搜索 RAG 与微调引用来源当前证据链存档修订 4（2026-08-01）", "description": "在修订 3 的不可变历史证据链上，新增绑定生产发布 20260801T120818Z-e6aab2fa868a 的当前百度严格无品牌基线、17 次可比观察逐报告 SHA-256 账本与当前标题意图矩阵。确定性 ZIP 含 37 个来源资产、44 个归档文件，可离线校验；归档未触发提供商调用或站外发布，也不构成搜索可见性、外链或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r4.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 搜索 RAG 与微调引用来源当前证据链存档修订 4（2026-08-01）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r4.json" }, { "@type": "DataDownload", "name": "当前证据链修订 4 离线校验 ZIP", "encodingFormat": "application/zip", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r4.zip" }, { "@type": "DataDownload", "name": "当前证据链修订 4 RFC 9264 Linkset", "encodingFormat": "application/linkset+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-linkset-2026-08-01-r4.json" }, { "@type": "DataDownload", "name": "当前证据链修订 4 机器清单", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-manifest-2026-08-01-r4.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-2026-07-31-r3.json", "identifier": "ai-search-rag-ten-observation-current-evidence-2026-07-31-r3", "name": "AI 搜索 RAG 与微调引用来源十次观察当前证据链存档修订 3（2026-07-31）", "description": "在不改写修订 2 的前提下，将当前机器答案的 CFF、BibTeX 与 RIS 引用元数据纳入确定性证据链；共封装 32 个来源资产和 39 个归档文件，并保留 MANIFEST、SHA256SUMS、RO-Crate、CITATION.cff、RFC 9264 Linkset 与独立验证脚本。该存档未发布站外，不构成外链、搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-2026-07-31-r3.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 搜索 RAG 与微调引用来源十次观察当前证据链存档修订 3（2026-07-31）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-2026-07-31-r3.json" }, { "@type": "DataDownload", "name": "十次观察当前证据链修订 3 离线验证 ZIP", "encodingFormat": "application/zip", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-2026-07-31-r3.zip" }, { "@type": "DataDownload", "name": "十次观察当前证据链修订 3 RFC 9264 Linkset", "encodingFormat": "application/linkset+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-linkset-2026-07-31-r3.json" }, { "@type": "DataDownload", "name": "十次观察当前证据链修订 3 机器清单", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-manifest-2026-07-31-r3.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-2026-07-31-r2.json", "identifier": "ai-search-rag-ten-observation-current-evidence-2026-07-31-r2", "name": "AI 搜索 RAG 与微调引用来源十次观察当前证据链存档（2026-07-31）", "description": "将基础答案、九次版本谱系、五批 AI 引用持续性、百度十次严格观察、第九到第十次变动、十次跨表面主机/URL/来源类型、当前机器答案、四张检索卡，以及当前生产发布的严格搜索基线、十次到当前差分和逐问法证据卡共 29 个来源资产封装为 36 文件确定性离线校验 ZIP；附 MANIFEST、SHA256SUMS、RO-Crate、CITATION.cff、RFC 9264 Linkset 和独立验证脚本。该存档未发布站外，不构成外链、搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-2026-07-31-r2.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 搜索 RAG 与微调引用来源十次观察当前证据链存档（2026-07-31）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-2026-07-31-r2.json" }, { "@type": "DataDownload", "name": "十次观察当前证据链离线验证 ZIP", "encodingFormat": "application/zip", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-2026-07-31-r2.zip" }, { "@type": "DataDownload", "name": "十次观察当前证据链 RFC 9264 Linkset", "encodingFormat": "application/linkset+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-linkset-2026-07-31-r2.json" }, { "@type": "DataDownload", "name": "十次观察当前证据链机器清单", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-manifest-2026-07-31-r2.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-answer-supplement-2026-07-30.json", "identifier": "ai-search-rag-nine-observation-answer-supplement-2026-07-30", "name": "AI 搜索 RAG 与微调引用来源九次观察机器答案补充（2026-07-30）", "description": "在不改写已版本化 v3.0 研究存档的前提下，将四条注册问法、四条可引用声明、九次百度严格无品牌观察、五批 AI 引用主机持久性、来源类型投影与 227 行逐 URL 覆盖前沿绑定为当前机器答案补充；明确取代八次观察补充作为当前口径，但保留旧资产为不可变历史分母。该资产未调用提供商，不构成正向搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-answer-supplement-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "AI 搜索 RAG 与微调引用来源九次观察机器答案补充（2026-07-30）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-answer-supplement-2026-07-30.json" }, { "@type": "DataDownload", "name": "九次观察机器答案逐问法与声明 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-answer-supplement-2026-07-30.csv" }, { "@type": "DataDownload", "name": "九次观察机器答案 CSL-JSON 引用元数据", "encodingFormat": "application/vnd.citationstyles.csl+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-answer-supplement-2026-07-30.csl.json" }, { "@type": "DataDownload", "name": "Nine-observation current answer JSON-LD knowledge graph", "encodingFormat": "application/ld+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-answer-supplement-2026-07-30.jsonld" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-retrieval-evidence-cards-2026-07-30.json", "identifier": "ai-search-rag-nine-observation-retrieval-evidence-cards-2026-07-30", "name": "千问、Kimi、Gemini RAG 与微调引用来源九次观察检索证据卡", "description": "将当前九次观察机器答案补充投影为四张逐查询差异化中文检索证据卡，并提供 16 行问法—声明 CSV；每张卡保留四条可引用声明、逐查询百度状态、证据定位、必要条件、禁止推断与显式非成功边界。该资产取代六次口径旧卡作为当前检索交付，但保留旧卡为不可变历史分母。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-retrieval-evidence-cards-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "烛龙智元内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问、Kimi、Gemini RAG 与微调引用来源九次观察检索证据卡", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-retrieval-evidence-cards-2026-07-30.json" }, { "@type": "DataDownload", "name": "九次观察当前证据链离线验证 ZIP", "encodingFormat": "application/zip", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-current-evidence-2026-07-30.zip" }, { "@type": "DataDownload", "name": "Nine-observation current evidence RFC 9264 linkset", "encodingFormat": "application/linkset+json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-current-evidence-linkset-2026-07-30.json" }, { "@type": "DataDownload", "name": "九次观察问法—声明证据索引 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-retrieval-evidence-cards-2026-07-30.csv" }, { "@type": "DataDownload", "name": "九次观察检索证据卡 Markdown", "encodingFormat": "text/markdown", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-retrieval-evidence-cards-2026-07-30.md" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-citation-reddit-surface-preflight-2026-07-31.json", "identifier": "ai-search-citation-reddit-surface-preflight-2026-07-31", "name": "Reddit cited-surface eligibility preflight (2026-07-31)", "description": "A read-only audit of the five Reddit URLs cited in the frozen five-batch benchmark, binding exact citation occurrences to subreddit policy and human-account eligibility boundaries. It is not a publication payload, backlink, or GEO success signal.", "url": "https://course.dragonai.tech/datasets/ai-search-citation-reddit-surface-preflight-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Content Research", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "distribution": [ { "@type": "DataDownload", "name": "Reddit cited-surface eligibility preflight (2026-07-31)", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-reddit-surface-preflight-2026-07-31.json" }, { "@type": "DataDownload", "name": "Reddit cited-surface eligibility preflight CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-reddit-surface-preflight-2026-07-31.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-citation-authorea-surface-preflight-2026-07-31.json", "identifier": "ai-search-citation-authorea-surface-preflight-2026-07-31", "name": "Authorea cited-surface eligibility preflight (2026-07-31)", "description": "A read-only audit of the exact Authorea URL cited four times in the frozen five-batch benchmark, binding run occurrences to Crossref metadata and permanent scholarly-record eligibility boundaries. It is not a manuscript, publication payload, backlink, or GEO success signal.", "url": "https://course.dragonai.tech/datasets/ai-search-citation-authorea-surface-preflight-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Content Research", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "distribution": [ { "@type": "DataDownload", "name": "Authorea cited-surface eligibility preflight (2026-07-31)", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-authorea-surface-preflight-2026-07-31.json" }, { "@type": "DataDownload", "name": "Authorea cited-surface eligibility preflight CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-authorea-surface-preflight-2026-07-31.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-citation-prepared-authority-host-coverage-gap-2026-07-31.json", "identifier": "ai-search-citation-prepared-authority-host-coverage-gap-2026-07-31", "name": "AI 搜索引用来源与待审批权威渠道覆盖差距（2026-07-31）", "description": "将固定查询中实际出现的引用主机与待真人审批的权威渠道逐项对齐，公开未覆盖来源及证据边界；准备状态不代表已发布、已获外链或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-prepared-authority-host-coverage-gap-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "distribution": [ { "@type": "DataDownload", "name": "AI 搜索引用来源与待审批权威渠道覆盖差距（2026-07-31）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-prepared-authority-host-coverage-gap-2026-07-31.json" }, { "@type": "DataDownload", "name": "待审批权威渠道覆盖差距 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-citation-prepared-authority-host-coverage-gap-2026-07-31.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-07-31.json", "identifier": "ai-search-rag-baidu-serp-entry-liveness-2026-07-31", "name": "百度严格无品牌结果新进入 URL 存活性诊断（2026-07-31）", "description": "对相邻严格百度观测中新进入结果集的 URL 做匿名只读抓取，记录 HTTP 存活性与内容指纹；该诊断不改写排名，也不构成 DragonAI 搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "distribution": [ { "@type": "DataDownload", "name": "百度严格无品牌结果新进入 URL 存活性诊断（2026-07-31）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-07-31.json" }, { "@type": "DataDownload", "name": "百度结果新进入 URL 存活性诊断 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-07-31.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-07-31-185000.json", "identifier": "ai-search-rag-baidu-serp-entry-liveness-2026-07-31-185000", "name": "百度严格无品牌结果新进入 URL 存活性复查（2026-07-31 18:50 UTC）", "description": "对后续相邻严格百度观测中新进入结果集的 URL 再做匿名只读抓取，保留 HTTP 存活性变化；该复查不改写排名，也不构成 DragonAI 搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-07-31-185000.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "distribution": [ { "@type": "DataDownload", "name": "百度严格无品牌结果新进入 URL 存活性复查（2026-07-31 18:50 UTC）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-07-31-185000.json" }, { "@type": "DataDownload", "name": "百度结果新进入 URL 存活性复查 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-07-31-185000.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bingbot-discovery-to-target-crawl-gap-2026-07-31.json", "identifier": "ai-search-rag-bingbot-discovery-to-target-crawl-gap-2026-07-31", "name": "Bingbot 发现面到研究目标页的抓取缺口（2026-07-31）", "description": "从当前生产发布的受控访问日志中提取经正反向 DNS 验证的 Bingbot 请求，公开研究 sitemap 与 Atom feed 已抓取而唯一目标页仍未抓取的最小化路径级证据；不公开 IP 或原始日志，也不把发现面抓取当作收录、排名、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bingbot-discovery-to-target-crawl-gap-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "distribution": [ { "@type": "DataDownload", "name": "Bingbot 发现面到研究目标页的抓取缺口（2026-07-31）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bingbot-discovery-to-target-crawl-gap-2026-07-31.json" }, { "@type": "DataDownload", "name": "Bingbot 发现面到目标页抓取缺口 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bingbot-discovery-to-target-crawl-gap-2026-07-31.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-11t232351z.json", "identifier": "ai-search-rag-baidu-serp-entry-liveness-2026-08-11t232351z", "name": "百度严格无品牌结果新进入 URL 存活性诊断（2026-08-11 23:23 UTC）", "description": "对相邻严格百度观测中唯一新进入结果 URL 做匿名只读抓取，记录其 HTTP 521 状态及退出 URL；该诊断不改写排名，也不构成 DragonAI 搜索可见性、来源质量、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-11t232351z.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI 内容研究组", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "distribution": [ { "@type": "DataDownload", "name": "百度严格无品牌结果新进入 URL 存活性诊断（2026-08-11 23:23 UTC）", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-11t232351z.json" }, { "@type": "DataDownload", "name": "百度结果新进入 URL 存活性诊断 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-11t232351z.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-12-143823.json", "identifier": "ai-search-rag-baidu-serp-entry-liveness-2026-08-12-143823", "name": "Baidu strict unbranded result-entry liveness diagnostic (2026-08-12 14:38 UTC)", "description": "Anonymous read-only fetch evidence for the single URL entering the current strict Baidu result set. The observed destination returned a Baidu verification page; this does not establish source quality, DragonAI search visibility, AI citation or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-12-143823.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Baidu strict unbranded result-entry liveness diagnostic (2026-08-12 14:38 UTC)", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-12-143823.json" }, { "@type": "DataDownload", "name": "Baidu result-entry liveness diagnostic CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-12-143823.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-release-adjacent-churn-2026-08-12.json", "identifier": "ai-search-rag-baidu-current-release-adjacent-churn-2026-08-12", "name": "Baidu strict unbranded adjacent-result churn for the current evidence release", "description": "Adjacent anonymous Baidu organic-result comparison across three fixed unbranded queries. The source-comparison query replaced six of seven URLs, while the target and brand domain remained absent from every unbranded result set.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-release-adjacent-churn-2026-08-12.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Baidu strict unbranded adjacent-result churn for the current evidence release", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-release-adjacent-churn-2026-08-12.json" }, { "@type": "DataDownload", "name": "Baidu current-release adjacent-result churn CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-release-adjacent-churn-2026-08-12.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-strict-query-answer-map-v1.json", "identifier": "ai-search-rag-strict-query-answer-map-v1", "name": "RAG、微调与 AI 引用来源三条严格无品牌查询答案地图", "description": "将三条固定严格无品牌查询映射到当前模型引用、百度竞争结果、Bing 十二次观察和可下载证据；DragonAI 仅在 Bing 的来源对比查询出现。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-strict-query-answer-map-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/comparisons/rag-vs-fine-tuning#article", "url": "https://course.dragonai.tech/comparisons/rag-vs-fine-tuning" }, { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-search-citation-monitoring-method#article", "url": "https://course.dragonai.tech/guides/ai-search-citation-monitoring-method" }, { "@type": "Article", "@id": "https://course.dragonai.tech/guides/how-ai-search-chooses-citation-sources#article", "url": "https://course.dragonai.tech/guides/how-ai-search-chooses-citation-sources" }, { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "RAG、微调与 AI 引用来源三条严格无品牌查询答案地图", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-strict-query-answer-map-v1.json" }, { "@type": "DataDownload", "name": "三条严格无品牌查询答案地图 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-strict-query-answer-map-v1.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-overlap-2026-08-14-r2.json", "identifier": "ai-search-rag-current-baidu-ai-overlap-2026-08-14-r2", "name": "当前百度严格结果与 2026-08-14 三平台引用交集 r2", "description": "把当前百度三条严格无品牌查询的 22 个不同结果 URL 与 2026-08-14 千问、Kimi、Gemini 九次复测的 50 个不同引用 URL 做主机与精确 URL 全外联结；两侧有 2 个精确 URL 和 3 个规范化主机重合，DragonAI 在两侧均未出现。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-overlap-2026-08-14-r2.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/comparisons/rag-vs-fine-tuning#article", "url": "https://course.dragonai.tech/comparisons/rag-vs-fine-tuning" }, { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-search-citation-monitoring-method#article", "url": "https://course.dragonai.tech/guides/ai-search-citation-monitoring-method" }, { "@type": "Article", "@id": "https://course.dragonai.tech/guides/how-ai-search-chooses-citation-sources#article", "url": "https://course.dragonai.tech/guides/how-ai-search-chooses-citation-sources" }, { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "当前百度严格结果与 2026-08-14 三平台引用交集 r2", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-overlap-2026-08-14-r2.json" }, { "@type": "DataDownload", "name": "当前百度与三平台引用主机交集 r2 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-overlap-2026-08-14-r2.csv" }, { "@type": "DataDownload", "name": "当前百度与三平台引用逐 URL 全外联结 r2 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-overlap-2026-08-14-r2-urls.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-top3-delivery-features-2026-08-12.json", "identifier": "ai-search-rag-baidu-current-top3-delivery-features-2026-08-12", "name": "Current Baidu top-three delivery features with DragonAI target reference", "description": "Read-only delivery-feature capture for nine current top-three observations across three fixed unbranded Baidu queries, plus the DragonAI target as an unranked reference. Most competitor pages blocked direct capture, so the dataset preserves failures and does not infer ranking causality, source quality, search visibility or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-top3-delivery-features-2026-08-12.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Current Baidu top-three delivery features with DragonAI target reference", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-top3-delivery-features-2026-08-12.json" }, { "@type": "DataDownload", "name": "Current Baidu top-three delivery features CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-top3-delivery-features-2026-08-12.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-top3-delivery-features-2026-08-14-r2.json", "identifier": "ai-search-rag-baidu-current-top3-delivery-features-2026-08-14-r2", "name": "Current-release Baidu top-three delivery features with DragonAI target reference, revision 2", "description": "Read-only delivery-feature capture for the current production release and the nine top-three observations across three fixed unbranded Baidu queries, plus the DragonAI target as an unranked reference. Failed competitor fetches remain explicit; feature co-occurrence is not ranking causality, source quality, visibility or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-top3-delivery-features-2026-08-14-r2.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "distribution": [ { "@type": "DataDownload", "name": "Current-release Baidu top-three delivery features with DragonAI target reference, revision 2", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-top3-delivery-features-2026-08-14-r2.json" }, { "@type": "DataDownload", "name": "Current-release Baidu top-three delivery features revision 2 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-top3-delivery-features-2026-08-14-r2.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-query-authority-answer-card-v1.json", "identifier": "rag-fine-tuning-selection-query-authority-answer-card-v1", "name": "RAG 与微调选型 AI 引用来源：查询级权威答案卡 v1", "description": "把固定无品牌选型查询的直接答案、六份一手技术与治理来源、逐项验证动作、不可推断边界、2026-08-13 22:00 UTC 百度当前八个自然结果，以及当前目标与八份竞品文档的 16 指标交付对照绑定在同一对象中；仅三份竞品返回可用 HTML，DragonAI 仍为零命中。", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-query-authority-answer-card-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "RAG 与微调选型 AI 引用来源：查询级权威答案卡 v1", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-query-authority-answer-card-v1.json" }, { "@type": "DataDownload", "name": "RAG 与微调选型查询级权威答案卡（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-query-authority-answer-card-v1.csv" }, { "@type": "DataDownload", "name": "RAG 与微调选型查询级权威答案图（JSON-LD）", "encodingFormat": "application/ld+json", "contentUrl": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-query-authority-answer-card-v1.jsonld" }, { "@type": "DataDownload", "name": "RAG 与微调选型查询级权威答案卡（CSL-JSON）", "encodingFormat": "application/vnd.citationstyles.csl+json", "contentUrl": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-query-authority-answer-card-v1.csl.json" }, { "@type": "DataDownload", "name": "RAG 与微调选型查询级权威答案卡（CFF）", "encodingFormat": "text/yaml", "contentUrl": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-query-authority-answer-card-v1.cff" }, { "@type": "DataDownload", "name": "RAG 与微调选型查询级权威答案卡（BibTeX）", "encodingFormat": "application/x-bibtex", "contentUrl": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-query-authority-answer-card-v1.bib" }, { "@type": "DataDownload", "name": "RAG 与微调选型查询级权威答案卡（RIS）", "encodingFormat": "application/x-research-info-systems", "contentUrl": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-query-authority-answer-card-v1.ris" }, { "@type": "DataDownload", "name": "RAG 与微调选型查询级权威答案卡（RFC 9264 Linkset）", "encodingFormat": "application/linkset+json", "contentUrl": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-query-authority-answer-card-v1.linkset.json" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/baidu-selection-answer-source-layer-gap-2026-08-13.json", "identifier": "baidu-selection-answer-source-layer-gap-2026-08-13", "name": "Baidu selection-answer named, organic and concrete-citation source gap", "description": "A seven-row join of the sources named in one anonymous Baidu generated answer, the eight current organic results, and the one exposed reference-document link. Six source classes were named, four also appeared organically, one exposed a concrete citation, and DragonAI was absent from all three layers. These point-in-time layers are not interchangeable and do not prove ranking causality or GEO success.", "url": "https://course.dragonai.tech/datasets/baidu-selection-answer-source-layer-gap-2026-08-13.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Baidu selection-answer named, organic and concrete-citation source gap", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/baidu-selection-answer-source-layer-gap-2026-08-13.json" }, { "@type": "DataDownload", "name": "Baidu selection-answer source-layer gap CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/baidu-selection-answer-source-layer-gap-2026-08-13.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-51cto-cross-engine-channel-candidate-2026-08-14.json", "identifier": "ai-search-rag-51cto-cross-engine-channel-candidate-2026-08-14", "name": "千问 Kimi Gemini RAG 微调引用来源：51CTO 跨百度与 Bing 证据", "description": "固定无品牌查询在百度与 Bing 中命中 blog.51cto.com 的 20 条可复算记录；两侧页面不同，DragonAI 目标与品牌域均未出现，因此它只支持渠道候选判断，不构成排名因果、推荐或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-51cto-cross-engine-channel-candidate-2026-08-14.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "千问 Kimi Gemini RAG 微调引用来源：51CTO 跨百度与 Bing 证据", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-51cto-cross-engine-channel-candidate-2026-08-14.json" }, { "@type": "DataDownload", "name": "51CTO 跨引擎逐观察证据 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-51cto-cross-engine-channel-candidate-2026-08-14.csv" }, { "@type": "DataDownload", "name": "51CTO 跨引擎证据复算说明", "encodingFormat": "text/markdown", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-51cto-cross-engine-channel-candidate-2026-08-14.md" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r2.json", "identifier": "ai-search-rag-bing-current-query-delta-2026-08-14-r2", "name": "Bing 中国无品牌三查询同发布变化 r2", "description": "2026-08-14 13:22 UTC 对同一生产发布下三条固定 Bing 中国匿名无品牌查询复核：30 个查询级 URL 出现保留 27、进入 3、退出 3，16 个留存 URL 位次变化；全部进入和退出来自第一条查询。DragonAI 目标页与品牌域均为 0，结果变化不构成排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r2.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询同发布变化 r2", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r2.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询同发布变化 r2（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r2.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r3.json", "identifier": "ai-search-rag-bing-current-query-delta-2026-08-14-r3", "name": "Bing 中国无品牌三查询目标页首次点时出现 r3", "description": "2026-08-14 22:08 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在无品牌查询“AI 搜索 RAG 微调 引用来源 对比”位于第 1，另外两条查询未出现。该单次点时结果计作搜索可见性，不证明持续排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r3.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页首次点时出现 r3", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r3.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页首次点时出现 r3（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r3.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r4.json", "identifier": "ai-search-rag-bing-current-query-delta-2026-08-14-r4", "name": "Bing 中国无品牌三查询目标页连续第二次位列第 1 r4", "description": "2026-08-14 22:44 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第二次位于第 1，另外两条仍未出现。相邻快照保留 26 个查询级 URL、进入 4、退出 4，12 个留存 URL 换位；两个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r4.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第二次位列第 1 r4", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r4.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第二次位列第 1 r4（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r4.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r5.json", "identifier": "ai-search-rag-bing-current-query-delta-2026-08-14-r5", "name": "Bing 中国无品牌三查询目标页连续第三次位列第 1 r5", "description": "2026-08-14 23:10 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第三次位于第 1，另外两条仍未出现。相邻快照的 30 个查询级 URL 全部留存、没有进入或退出，12 个 URL 换位；三个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r5.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第三次位列第 1 r5", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r5.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第三次位列第 1 r5（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r5.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r6.json", "identifier": "ai-search-rag-bing-current-query-delta-2026-08-14-r6", "name": "Bing 中国无品牌三查询目标页连续第四次位列第 1 r6", "description": "2026-08-14 23:41 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第四次位于第 1，另外两条仍未出现。相邻快照保留 29 个查询级 URL、进入 1、退出 1，7 个 URL 换位；四个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r6.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第四次位列第 1 r6", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r6.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第四次位列第 1 r6（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r6.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r7.json", "identifier": "ai-search-rag-bing-current-query-delta-2026-08-15-r7", "name": "Bing 中国无品牌三查询目标页连续第五次位列第 1 r7", "description": "2026-08-15 00:17 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第五次位于第 1，另外两条仍未出现。相邻快照保留 27 个查询级 URL、进入 3、退出 3，12 个 URL 换位；五个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r7.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第五次位列第 1 r7", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r7.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第五次位列第 1 r7（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r7.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-five-observation-persistence-2026-08-15.json", "identifier": "ai-search-rag-bing-five-observation-persistence-2026-08-15", "name": "Bing 中国无品牌三查询五次观察 URL 与主机持久性", "description": "合并 2026-08-14 22:08 至 2026-08-15 00:17 UTC 的五次匿名 Bing 中国区只读观察：15 个固定查询快照、150 个自然结果出现、36 条查询—URL 行与 26 条查询—主机行。DragonAI 目标研究页只在一条无品牌查询中五次均位于第 1；这计作已观察搜索可见性，不证明长期排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-five-observation-persistence-2026-08-15.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询五次观察 URL 与主机持久性", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-five-observation-persistence-2026-08-15.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询五次观察 URL 持久性（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-five-observation-persistence-2026-08-15.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r8.json", "identifier": "ai-search-rag-bing-current-query-delta-2026-08-15-r8", "name": "Bing 中国无品牌三查询目标页连续第六次位列第 1 r8", "description": "2026-08-15 01:06 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第六次位于第 1，另外两条仍未出现。相邻快照保留 29 个查询级 URL、进入 1、退出 1，9 个 URL 换位；六个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r8.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第六次位列第 1 r8", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r8.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第六次位列第 1 r8（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r8.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-six-observation-persistence-2026-08-15.json", "identifier": "ai-search-rag-bing-six-observation-persistence-2026-08-15", "name": "Bing 中国无品牌三查询六次观察 URL 与主机持久性", "description": "合并 2026-08-14 22:08 至 2026-08-15 01:06 UTC 的六次匿名 Bing 中国区只读观察：18 个固定查询快照、180 个自然结果出现、36 条查询—URL 行与 26 条查询—主机行。DragonAI 目标研究页只在一条无品牌查询中六次均位于第 1；这计作已观察搜索可见性，不证明长期排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-six-observation-persistence-2026-08-15.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询六次观察 URL 与主机持久性", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-six-observation-persistence-2026-08-15.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询六次观察 URL 持久性（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-six-observation-persistence-2026-08-15.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r9.json", "identifier": "ai-search-rag-bing-current-query-delta-2026-08-15-r9", "name": "Bing 中国无品牌三查询目标页连续第七次位列第 1 r9", "description": "2026-08-15 01:42 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第七次位于第 1，另外两条仍未出现。相邻快照保留 29 个查询级 URL、进入 1、退出 1，7 个 URL 换位；七个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r9.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第七次位列第 1 r9", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r9.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第七次位列第 1 r9（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r9.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-seven-observation-persistence-2026-08-15.json", "identifier": "ai-search-rag-bing-seven-observation-persistence-2026-08-15", "name": "Bing 中国无品牌三查询七次观察 URL 与主机持久性", "description": "合并 2026-08-14 22:08 至 2026-08-15 01:42 UTC 的七次匿名 Bing 中国区只读观察：21 个固定查询快照、210 个自然结果出现、36 条查询—URL 行与 26 条查询—主机行。DragonAI 目标研究页只在一条无品牌查询中七次均位于第 1；这计作已观察搜索可见性，不证明长期排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-seven-observation-persistence-2026-08-15.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询七次观察 URL 与主机持久性", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-seven-observation-persistence-2026-08-15.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询七次观察 URL 持久性（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-seven-observation-persistence-2026-08-15.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r10.json", "identifier": "ai-search-rag-bing-current-query-delta-2026-08-15-r10", "name": "Bing 中国无品牌三查询目标页连续第八次位列第 1 r10", "description": "2026-08-15 02:17 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第八次位于第 1，另外两条仍未出现。相邻快照 30 个查询级 URL 全部保留，4 个 URL 换位；八个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r10.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第八次位列第 1 r10", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r10.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第八次位列第 1 r10（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r10.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-eight-observation-persistence-2026-08-15.json", "identifier": "ai-search-rag-bing-eight-observation-persistence-2026-08-15", "name": "Bing 中国无品牌三查询八次观察 URL 与主机持久性", "description": "合并 2026-08-14 22:08 至 2026-08-15 02:17 UTC 的八次匿名 Bing 中国区只读观察：24 个固定查询快照、240 个自然结果出现、36 条查询—URL 行与 26 条查询—主机行。DragonAI 目标研究页只在一条无品牌查询中八次均位于第 1；这计作已观察搜索可见性，不证明长期排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-eight-observation-persistence-2026-08-15.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询八次观察 URL 与主机持久性", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-eight-observation-persistence-2026-08-15.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询八次观察 URL 持久性（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-eight-observation-persistence-2026-08-15.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r11.json", "identifier": "ai-search-rag-bing-current-query-delta-2026-08-15-r11", "name": "Bing 中国无品牌三查询目标页连续第九次位列第 1 r11", "description": "2026-08-15 03:52 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第九次位于第 1，另外两条仍未出现。相邻快照保留 29 个查询级 URL、进入 1、退出 1，10 个留存 URL 换位；九个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r11.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第九次位列第 1 r11", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r11.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第九次位列第 1 r11（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r11.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-nine-observation-persistence-2026-08-15.json", "identifier": "ai-search-rag-bing-nine-observation-persistence-2026-08-15", "name": "Bing 中国无品牌三查询九次观察 URL 与主机持久性", "description": "合并 2026-08-14 22:08 至 2026-08-15 03:52 UTC 的九次匿名 Bing 中国区只读观察：27 个固定查询快照、270 个自然结果出现、36 条查询—URL 行与 26 条查询—主机行。DragonAI 目标研究页只在一条无品牌查询中九次均位于第 1；这计作已观察搜索可见性，不证明长期排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-nine-observation-persistence-2026-08-15.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询九次观察 URL 与主机持久性", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-nine-observation-persistence-2026-08-15.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询九次观察 URL 持久性（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-nine-observation-persistence-2026-08-15.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r12.json", "identifier": "ai-search-rag-bing-current-query-delta-2026-08-15-r12", "name": "Bing 中国无品牌三查询目标页连续第十次位列第 1 r12", "description": "2026-08-15 04:58 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第十次位于第 1，另外两条仍未出现。相邻快照保留 28 个查询级 URL、进入 2、退出 2，3 个留存 URL 换位；十个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r12.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第十次位列第 1 r12", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r12.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第十次位列第 1 r12（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r12.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-ten-observation-persistence-2026-08-15.json", "identifier": "ai-search-rag-bing-ten-observation-persistence-2026-08-15", "name": "Bing 中国无品牌三查询十次观察 URL 与主机持久性", "description": "合并 2026-08-14 22:08 至 2026-08-15 04:58 UTC 的十次匿名 Bing 中国区只读观察：30 个固定查询快照、300 个自然结果出现、36 条查询—URL 行与 26 条查询—主机行。DragonAI 目标研究页只在一条无品牌查询中十次均位于第 1；这计作已观察搜索可见性，不证明长期排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-ten-observation-persistence-2026-08-15.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询十次观察 URL 与主机持久性", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-ten-observation-persistence-2026-08-15.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询十次观察 URL 持久性（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-ten-observation-persistence-2026-08-15.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r13.json", "identifier": "ai-search-rag-bing-current-query-delta-2026-08-15-r13", "name": "Bing 中国无品牌三查询目标页连续第十一次位列第 1 r13", "description": "2026-08-15 05:55 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第十一次位于第 1，另外两条仍未出现。相邻快照保留 26 个查询级 URL、进入 4、退出 4，12 个留存 URL 换位；十一个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r13.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第十一次位列第 1 r13", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r13.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第十一次位列第 1 r13（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r13.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-eleven-observation-persistence-2026-08-15.json", "identifier": "ai-search-rag-bing-eleven-observation-persistence-2026-08-15", "name": "Bing 中国无品牌三查询十一次观察 URL 与主机持久性", "description": "合并 2026-08-14 22:08 至 2026-08-15 05:55 UTC 的十一次匿名 Bing 中国区只读观察：33 个固定查询快照、330 个自然结果出现、38 条查询—URL 行与 27 条查询—主机行。DragonAI 目标研究页只在一条无品牌查询中十一次均位于第 1；这计作已观察搜索可见性，不证明长期排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-eleven-observation-persistence-2026-08-15.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/comparisons/rag-vs-fine-tuning#article", "url": "https://course.dragonai.tech/comparisons/rag-vs-fine-tuning" }, { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-search-citation-monitoring-method#article", "url": "https://course.dragonai.tech/guides/ai-search-citation-monitoring-method" }, { "@type": "Article", "@id": "https://course.dragonai.tech/guides/how-ai-search-chooses-citation-sources#article", "url": "https://course.dragonai.tech/guides/how-ai-search-chooses-citation-sources" }, { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询十一次观察 URL 与主机持久性", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-eleven-observation-persistence-2026-08-15.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询十一次观察 URL 持久性（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-eleven-observation-persistence-2026-08-15.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r14.json", "identifier": "ai-search-rag-bing-current-query-delta-2026-08-15-r14", "name": "Bing 中国无品牌三查询目标页连续第十二次位列第 1 r14", "description": "2026-08-15 09:02 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第十二次位于第 1，另外两条仍未出现。相邻快照保留 27 个查询级 URL、进入 3、退出 3，7 个留存 URL 换位；十二个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r14.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第十二次位列第 1 r14", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r14.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询目标页连续第十二次位列第 1 r14（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r14.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-bing-twelve-observation-persistence-2026-08-15.json", "identifier": "ai-search-rag-bing-twelve-observation-persistence-2026-08-15", "name": "Bing 中国无品牌三查询十二次观察 URL 与主机持久性", "description": "合并 2026-08-14 22:08 至 2026-08-15 09:02 UTC 的十二次匿名 Bing 中国区只读观察：36 个固定查询快照、360 个自然结果出现、38 条查询—URL 行与 27 条查询—主机行。DragonAI 目标研究页只在一条无品牌查询中十二次均位于第 1；这计作已观察搜索可见性，不证明长期排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-twelve-observation-persistence-2026-08-15.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "Article", "@id": "https://course.dragonai.tech/comparisons/rag-vs-fine-tuning#article", "url": "https://course.dragonai.tech/comparisons/rag-vs-fine-tuning" }, { "@type": "Article", "@id": "https://course.dragonai.tech/guides/ai-search-citation-monitoring-method#article", "url": "https://course.dragonai.tech/guides/ai-search-citation-monitoring-method" }, { "@type": "Article", "@id": "https://course.dragonai.tech/guides/how-ai-search-chooses-citation-sources#article", "url": "https://course.dragonai.tech/guides/how-ai-search-chooses-citation-sources" }, { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "Bing 中国无品牌三查询十二次观察 URL 与主机持久性", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-twelve-observation-persistence-2026-08-15.json" }, { "@type": "DataDownload", "name": "Bing 中国无品牌三查询十二次观察 URL 持久性（CSV）", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-bing-twelve-observation-persistence-2026-08-15.csv" } ] }, { "@type": "Dataset", "@id": "https://course.dragonai.tech/datasets/ai-search-rag-csdn-baidu-qwen-current-overlap-2026-08-14.json", "identifier": "ai-search-rag-csdn-baidu-qwen-current-overlap-2026-08-14", "name": "CSDN 当前百度自然结果与合格千问引用逐出现投影", "description": "blog.csdn.net 在当前百度三问中有 4 个自然结果位次、4 个不同 URL；最近合格千问批次 3 次运行中有 11 次引用 URL 出现、4 个不同 URL。两侧精确 URL 交集为 0，绑定不同生产发布，DragonAI 目标与品牌证据均为 0，因此不构成同发布效应、共享索引、来源质量、渠道推荐或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-csdn-baidu-qwen-current-overlap-2026-08-14.json", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "DragonAI Editorial Research Team", "@id": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team", "url": "https://course.dragonai.tech/brand/editorial-methodology#editorial-team" }, "publisher": { "@id": "https://dragonai.tech/#organization" }, "includedInDataCatalog": { "@id": "https://course.dragonai.tech/datasets/index.json#catalog" }, "isPartOf": [ { "@type": "ScholarlyArticle", "@id": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07#article", "url": "https://course.dragonai.tech/research/ai-search-rag-finetuning-citations-2026-07" } ], "distribution": [ { "@type": "DataDownload", "name": "CSDN 当前百度自然结果与合格千问引用逐出现投影", "encodingFormat": "application/json", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-csdn-baidu-qwen-current-overlap-2026-08-14.json" }, { "@type": "DataDownload", "name": "CSDN 当前百度与合格千问逐出现投影 CSV", "encodingFormat": "text/csv", "contentUrl": "https://course.dragonai.tech/datasets/ai-search-rag-csdn-baidu-qwen-current-overlap-2026-08-14.csv" } ] } ], "datasets": [ { "id": "ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-15-r2", "title": "RAG 与微调引用来源：当前百度—Bing 精确 URL 与主机对照 r2", "description": "以三条冻结无品牌查询对照最新百度与 Bing 中国匿名自然结果：百度 25 个位次、22 个不同 URL，Bing 30 个位次、28 个不同 URL；跨查询精确 URL 交集为 1。DragonAI 目标页只在 Bing 的一条查询位于第 1，在百度为 0，跨引擎目标命中为 0；这是搜索可见性，不是 AI 引用、品牌提及或跨引擎 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-15-r2.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "当前百度—Bing 精确 URL 对照 r2（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-15-r2.csv", "encodingFormat": "text/csv" }, { "title": "当前百度—Bing 同查询精确主机对照 r2（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-15-r2-hosts.csv", "encodingFormat": "text/csv" }, { "title": "当前百度—Bing 精确 URL 对照 r2（Markdown）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-15-r2.md", "encodingFormat": "text/markdown" } ] }, { "id": "ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-14", "title": "RAG and fine-tuning citation sources: current Baidu–Bing exact-URL overlap", "description": "A deterministic same-query comparison of 25 Baidu and 30 Bing China anonymous organic result positions for three frozen unbranded queries, covering 22 and 26 distinct URLs. The cross-query union is 47 URLs and one exact URL overlaps within the same query; the target and brand domain are absent on both surfaces. This point-in-time snapshot is not visibility or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-14.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Current Baidu–Bing exact-URL overlap CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-14.csv", "encodingFormat": "text/csv" }, { "title": "Current Baidu–Bing same-query exact-host overlap CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-14-hosts.csv", "encodingFormat": "text/csv" }, { "title": "Current Baidu–Bing exact-URL overlap Markdown", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-bing-current-exact-url-overlap-2026-08-14.md", "encodingFormat": "text/markdown" } ] }, { "id": "ai-search-rag-baidu-forty-six-observation-host-priority-2026-08-14", "title": "RAG and fine-tuning citation sources: 46-observation Baidu exact-host priority matrix", "description": "Exact-host aggregation for three frozen unbranded Baidu queries across 46 comparable observations, exposing result occurrences, top-three slot occupancy, top-three observation coverage, position ranges, and latest URLs. The matrix prioritizes evidence review and distribution verification without treating frequency as source quality, ranking causality, visibility, or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-forty-six-observation-host-priority-2026-08-14.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "46-observation Baidu exact-host priority matrix CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-forty-six-observation-host-priority-2026-08-14.csv", "encodingFormat": "text/csv" }, { "title": "46-observation Baidu exact-host priority matrix Markdown", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-forty-six-observation-host-priority-2026-08-14.md", "encodingFormat": "text/markdown" } ] }, { "id": "ai-search-rag-baidu-forty-six-observation-query-evidence-2026-08-14", "title": "RAG and fine-tuning citation sources: 46-observation Baidu query evidence cards", "description": "Direct-answer evidence cards for three frozen unbranded Baidu queries across 46 comparable observations and 31 production releases, with URL position sequences and latest top-three results. DragonAI target and brand-domain hits remain zero; this is not GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-forty-six-observation-query-evidence-2026-08-14.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "46-observation Baidu query evidence cards CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-forty-six-observation-query-evidence-2026-08-14.csv", "encodingFormat": "text/csv" }, { "title": "46-observation Baidu query evidence cards Markdown", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-forty-six-observation-query-evidence-2026-08-14.md", "encodingFormat": "text/markdown" } ] }, { "id": "ai-search-rag-research-authority-graph-v1", "title": "RAG and fine-tuning citation research authority graph", "description": "A deterministic graph of the canonical DragonAI research target, 19 query-aligned contextual internal references, and one verified independent external repository reference. The external reference is not classified as earned editorial because the required human receipt is incomplete; every edge remains explicitly false for search visibility, AI citation, ranking improvement, and GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-research-authority-graph-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Research authority graph edge list CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-research-authority-graph-v1.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-release-observation-panel-2026-08-14", "title": "Baidu strict unbranded release-aligned observation panel", "description": "Release-level aggregation of 46 comparable anonymous Baidu observations across 31 production releases, 138 strict-query snapshots and 1,137 organic positions. It separates release count from observation count and preserves zero target and brand-domain observations.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-release-observation-panel-2026-08-14.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Baidu release-aligned observation panel CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-release-observation-panel-2026-08-14.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-current-baidu-ai-overlap-2026-08-11", "title": "Current Baidu and AI citation source overlap for RAG and fine-tuning", "description": "A deterministic 70-row full outer join of the current 25-position, 22-URL Baidu snapshot and the current nine-run Qwen, Kimi and Gemini citation batch. Three normalized hosts and two exact URLs overlap; DragonAI is absent from both surfaces, so this is not shared-index evidence, target visibility or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-overlap-2026-08-11.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Current Baidu and AI citation hostname overlap CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-overlap-2026-08-11.csv", "encodingFormat": "text/csv" }, { "title": "Current Baidu and AI citation exact URL full outer join CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-overlap-2026-08-11-urls.csv", "encodingFormat": "text/csv" }, { "title": "Current Baidu and AI citation URL overlap summary figure", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-overlap-2026-08-11.svg", "encodingFormat": "image/svg+xml" }, { "title": "Reproducible current Baidu and AI citation overlap deposit supplement", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-overlap-supplement-2026-08-11.zip", "encodingFormat": "application/zip" } ] }, { "id": "ai-search-rag-baidu-current-strict-snapshot-2026-08-11", "title": "Baidu current strict unbranded RAG and fine-tuning snapshot", "description": "Current anonymous read-only results for the three frozen unbranded Baidu queries at 2026-08-14 15:04 UTC: 25 organic positions across 22 distinct URLs. The machine-readable source-group distribution is CSDN 4, Baijiahao 5, Tencent Cloud 4, Baidu AI Cloud 3, Zhihu 2, and other sources 7; DragonAI target-page and brand-domain appearances remain zero. Relative to the 12:53 UTC same-release observation, 24 query-level URLs were retained without rank movement and the BAAI Hub page entered the provider-source query at position 10. The grouping and adjacent change are descriptive, not source-quality rankings, ranking causality, AI citation evidence, or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-strict-snapshot-2026-08-11.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Current strict Baidu snapshot CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-strict-snapshot-2026-08-11.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-entry-liveness-2026-08-14-150436", "title": "Baidu strict provider-source SERP entry liveness at 2026-08-14 15:04 UTC", "description": "Read-only liveness evidence for the one URL newly entering the fixed unbranded provider-source query at position 10: the BAAI Hub page returned live canonical HTML. The previous nine URLs retained their positions, while DragonAI target-page and brand-domain appearances remained zero across the strict suite. Entry and liveness do not establish source quality, ranking causality, target visibility, AI citation, or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-14-150436.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Baidu strict provider-source SERP entry liveness CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-14-150436.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-entry-liveness-2026-08-13t175038z", "title": "Baidu strict SERP entry liveness at 2026-08-13 17:50 UTC", "description": "Read-only liveness evidence for the three URLs that entered the current adjacent strict Baidu snapshot: Red Hat and a Baijiahao safety-verification page returned HTML, while the CSDN request ended in a connection error. DragonAI target-page and brand-domain appearances remained zero. Entry and liveness do not establish source quality, ranking causality, target visibility, AI citation, or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-13t175038z.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Baidu strict SERP entry liveness CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-13t175038z.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-entry-liveness-2026-08-13t170117z", "title": "Baidu strict SERP entry liveness at 2026-08-13 17:01 UTC", "description": "Read-only liveness evidence for the one URL that entered the current adjacent strict Baidu snapshot: the BAAI Hub page returned live HTML, while DragonAI target-page and brand-domain appearances remained zero. Entry and liveness do not establish source quality, ranking causality, target visibility, AI citation, or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-13t170117z.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Baidu strict SERP entry liveness CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-13t170117z.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-citation-benchmark-current-batch-2026-08-11", "title": "千问、Kimi、Gemini 固定问题当前 9 次引用复测（2026-08-11）", "description": "公开同一固定无品牌问题的三平台各 3 次逐 URL 结果、零引用运行、模型版本与严格趋势边界；DragonAI 目标引用和品牌提及均为零。可下载的脱敏原始响应保留回答文本、搜索或 grounding 元数据、引用表面与用量字段，仅对提供商请求/响应标识符作带 SHA-256 承诺的脱敏；第三方内容不在 DragonAI 的 CC BY 4.0 授权范围内。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-current-batch-2026-08-11.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "当前 9 次复测逐引用 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-current-batch-2026-08-11.csv", "encodingFormat": "text/csv" }, { "title": "当前 9 次复测脱敏原始响应与复现代码 ZIP（版本 1）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-benchmark-redacted-raw-evidence-2026-08-11-v1.zip", "encodingFormat": "application/zip" }, { "title": "当前百度与三平台引用审计存档候选 ZIP（版本 1）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-audit-deposit-2026-08-11-v1.zip", "encodingFormat": "application/zip" }, { "title": "当前研究证据上下文交付包 ZIP（版本 9）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-context-deposit-2026-08-14-v9.zip", "encodingFormat": "application/zip" }, { "title": "当前研究证据上下文交付包 ZIP（版本 8）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-context-deposit-2026-08-14-v8.zip", "encodingFormat": "application/zip" } ] }, { "id": "ai-search-citation-benchmark-current-batch-2026-08-14", "title": "千问、Kimi、Gemini 固定问题当前 9 次引用复测（2026-08-14）", "description": "同一固定无品牌问题的三平台各 3 次逐 URL 观察：共 91 次具体引用 URL 出现、50 个不同 URL、37 个不同主机；Kimi 三次均未展示具体 URL，DragonAI 目标引用和品牌提及均为 0/9。该数据只描述点时引用表面，不代表来源质量、索引、排名因果或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-current-batch-2026-08-14.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "2026-08-14 当前 9 次复测逐引用 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-current-batch-2026-08-14.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-selection-query-live-churn-2026-08-02", "title": "Baidu live churn for the RAG and fine-tuning source-selection query", "description": "A bounded, read-only comparison between the 26th strict observation and one later anonymous Baidu result page for the fixed unbranded source-selection query. Six of eight URLs persisted, two entered, two exited, three persistent URLs changed rank, and the top three stayed unchanged. This partial single-query capture is not a 27th strict observation, AI citation evidence, or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-live-churn-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Baidu source-selection live churn CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-live-churn-2026-08-02.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-twenty-six-observation-query-evidence-2026-08-02", "title": "RAG and fine-tuning citation sources: 26-observation Baidu query evidence cards", "description": "Direct-answer evidence cards for three frozen unbranded Baidu queries across 26 comparable observations, with URL position sequences, persistent sources and latest top-three results. DragonAI target and brand-domain hits remain zero; this is not GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-six-observation-query-evidence-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "26-observation Baidu query evidence cards CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-six-observation-query-evidence-2026-08-02.csv", "encodingFormat": "text/csv" }, { "title": "26-observation Baidu query evidence cards Markdown", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-six-observation-query-evidence-2026-08-02.md", "encodingFormat": "text/markdown" } ] }, { "id": "ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r8", "title": "Baidu strict unbranded comparable-observation ledger (26 observations, revision 8)", "description": "Immutable cumulative ledger for 26 strict Baidu observations across 15 production releases, 78 query snapshots and 649 organic result positions, with source-report paths and hashes. Target and brand-domain hits remain zero.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r8.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Comparable-observation ledger revision 8 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r8.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-comparable-observation-ledger-2026-08-12-r9", "title": "Baidu strict unbranded comparable-observation ledger (33 observations, revision 9)", "description": "Immutable SHA-256-bound ledger for 33 comparable anonymous Baidu observations across 22 production releases, 99 query snapshots and 819 organic result positions. Target and brand-domain hits remain zero.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-12-r9.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Comparable-observation ledger revision 9 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-12-r9.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-comparable-observation-ledger-2026-08-14-r10", "title": "Baidu strict unbranded comparable-observation ledger (46 observations, revision 10)", "description": "Immutable SHA-256-bound ledger for 46 comparable anonymous Baidu observations across 31 production releases, 138 query snapshots and 1,137 organic result positions. Target and brand-domain hits remain zero.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-14-r10.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Comparable-observation ledger revision 10 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-14-r10.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-twenty-six-observation-persistence-2026-08-02", "title": "Baidu strict unbranded query-URL persistence across 26 observations", "description": "Query-URL persistence across 26 observations, 15 releases, 78 snapshots and 649 result positions. Eight of 51 query-URL pairs span all observations; none keeps an exact position throughout. Target and brand-domain hits remain zero.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-six-observation-persistence-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "26-observation query-URL persistence CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-six-observation-persistence-2026-08-02.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-query-host-persistence-twenty-six-observation-2026-08-02", "title": "Baidu strict unbranded query-host persistence across 26 observations", "description": "Query-host persistence across 26 observations. Ten of 28 query-host pairs span the full window, one keeps its exact position, and five stay in the top three. DragonAI target and brand domains remain absent.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-six-observation-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "26-observation query-host persistence CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-six-observation-2026-08-02.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-persistent-top3-source-cards-twenty-six-observation-2026-08-02", "title": "Baidu persistent top-three source cards across 26 strict observations", "description": "Joins five query-host pairs that stayed in the top three throughout 26 observations to four hosts and 12 concrete result URLs. Target and brand-domain hits remain zero; persistence is not a quality or GEO-success claim.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-six-observation-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "26-observation persistent top-three source cards CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-six-observation-2026-08-02.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-twenty-sixth-observation-churn-2026-08-02", "title": "Baidu strict-query adjacent churn from observation 25 to 26", "description": "Adjacent frozen SERP delta for three strict unbranded queries. Two queries are unchanged; the second retains five URLs, adds two, removes two and moves two. DragonAI target and brand-domain hits remain zero.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-sixth-observation-churn-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Observation 25 to 26 strict Baidu churn CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-sixth-observation-churn-2026-08-02.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-twenty-five-observation-query-evidence-2026-08-02", "title": "RAG 与微调引用来源：25 次百度观察逐查询证据卡", "description": "把三条严格无品牌百度查询的 25 次冻结观察压缩为可摘录答案卡，同时公开逐 URL 位置序列、全窗口持续 URL、持续前三主机和最新前三结果。目标页与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-five-observation-query-evidence-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "25 次百度观察逐查询证据卡 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-five-observation-query-evidence-2026-08-02.csv", "encodingFormat": "text/csv" }, { "title": "25 次百度观察逐查询证据卡 Markdown", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-five-observation-query-evidence-2026-08-02.md", "encodingFormat": "text/markdown" } ] }, { "id": "ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r7", "title": "百度严格无品牌搜索可比观察累计账本（25 次，修订 7）", "description": "25 次严格百度观察的累计账本：14 个生产发布、75 个查询快照、624 个自然结果位置；每次观察绑定源报告路径与 SHA-256。目标页与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r7.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度严格无品牌搜索可比观察累计账本修订 7 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r7.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-twenty-five-observation-persistence-2026-08-02", "title": "百度 RAG 与微调引用来源严格无品牌 25 次查询—URL 持续性（2026-08-02）", "description": "跨 14 个生产版本汇总 25 次严格无品牌百度观察、75 个查询快照和 624 个自然结果位置；50 个查询—URL 组合中有 8 个覆盖全部 25 次，没有 URL 在全窗口保持精确位置不变。DragonAI 目标页与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-five-observation-persistence-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度严格无品牌 25 次查询—URL 持续性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-five-observation-persistence-2026-08-02.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-query-host-persistence-twenty-five-observation-2026-08-02", "title": "百度 RAG 与微调引用来源严格无品牌 25 次查询×主机持久性（2026-08-02）", "description": "将 25 次严格无品牌百度观察按查询与精确主机聚合。28 个查询×主机组合中 10 个覆盖全部 25 次，只有 1 个在全窗口保持精确位置不变，5 个始终位于前 3；DragonAI 目标主机与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-five-observation-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度严格无品牌 25 次查询×主机持久性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-five-observation-2026-08-02.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-persistent-top3-source-cards-twenty-five-observation-2026-08-02", "title": "百度 RAG 与微调引用来源严格无品牌 25 次观察持续前三来源卡（2026-08-02）", "description": "将 25 次严格无品牌百度观察中每次都进入前三位的 5 个查询—主机组合，与 12 个具体结果 URL、最新标题及逐文档覆盖联表。共覆盖 4 个不同主机；DragonAI 目标主机与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-five-observation-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度严格无品牌 25 次观察持续前三来源卡 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-five-observation-2026-08-02.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-twenty-fifth-observation-churn-2026-08-02", "title": "百度 RAG 与微调引用来源第 24 至 25 次严格观察相邻波动", "description": "固定三条严格无品牌百度查询的第 24 至 25 次相邻观察：第二、第三条查询的 15 个自然结果位次完全不变；第一条查询保留 9 个 URL，新增与移除各 1 个，并有 1 个保留 URL 前移一位。新增与移除记录是同一篇 CSDN 文章的不同规范主机路径；DragonAI 目标页和品牌域仍均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-fifth-observation-churn-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "第 24 至 25 次严格百度观察相邻波动 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-fifth-observation-churn-2026-08-02.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-twenty-four-observation-persistence-2026-08-02", "title": "百度 RAG 与微调引用来源严格无品牌 24 次查询—URL 持续性（2026-08-02）", "description": "跨 13 个生产版本汇总 24 次严格无品牌百度观察、72 个查询快照和 599 个自然结果位置；49 个查询—URL 组合中有 9 个覆盖全部 24 次，只有 1 个在全窗口保持精确位置不变。DragonAI 目标页与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-four-observation-persistence-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度严格无品牌 24 次查询—URL 持续性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-four-observation-persistence-2026-08-02.csv", "encodingFormat": "text/csv" } ], "staticRoutes": [ "/research" ] }, { "id": "ai-search-rag-baidu-query-host-persistence-twenty-four-observation-2026-08-02", "title": "百度 RAG 与微调引用来源严格无品牌 24 次查询×主机持久性（2026-08-02）", "description": "将 24 次严格无品牌百度观察按查询与精确主机聚合。28 个查询×主机组合中 11 个覆盖全部 24 次，只有 2 个在全窗口保持精确位置不变，6 个始终位于前 3；DragonAI 目标主机与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-four-observation-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度严格无品牌 24 次查询×主机持久性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-four-observation-2026-08-02.csv", "encodingFormat": "text/csv" } ], "staticRoutes": [ "/research" ] }, { "id": "ai-search-rag-baidu-persistent-top3-source-cards-twenty-four-observation-2026-08-02", "title": "百度 RAG 与微调引用来源严格无品牌 24 次观察持续前三来源卡（2026-08-02）", "description": "将 24 次严格无品牌百度观察中每次都进入前三位的 6 个查询—主机组合，与 13 个具体结果 URL、最新标题及逐文档覆盖联表。三个固定问法各有两个持续前三主机；DragonAI 目标主机与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-four-observation-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度严格无品牌 24 次观察持续前三来源卡 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-four-observation-2026-08-02.csv", "encodingFormat": "text/csv" } ], "staticRoutes": [ "/research" ] }, { "id": "ai-search-rag-baidu-twenty-four-observation-query-evidence-2026-08-02", "title": "百度 RAG 与微调引用来源 24 次观察逐查询证据卡", "description": "三条严格无品牌百度查询的 24 次冻结观察证据卡，公开逐 URL 位置序列、全窗口持续 URL、持续前三主机与最新前三结果；目标页和品牌域均为零命中，不构成搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-four-observation-query-evidence-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "24 次百度观察逐查询证据卡 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-four-observation-query-evidence-2026-08-02.csv", "encodingFormat": "text/csv" }, { "title": "24 次百度观察逐查询证据卡 Markdown", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-four-observation-query-evidence-2026-08-02.md", "encodingFormat": "text/markdown" } ] }, { "id": "ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r6", "title": "百度严格无品牌搜索可比观察累计账本（24 次，修订 6）", "description": "24 次严格百度观察的累计账本：13 个生产发布、72 个查询快照、599 个自然结果位置；每次观察绑定源报告路径与 SHA-256。目标页与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r6.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度严格无品牌搜索可比观察累计账本修订 6 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r6.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-twenty-fourth-observation-churn-2026-08-02", "title": "百度 RAG 与微调引用来源第 23 至 24 次严格观察相邻波动", "description": "固定三条严格无品牌百度查询的第 23 至 24 次相邻观察：前两条查询的 17 个自然结果位次完全不变；选型查询保留 7 个 URL，新增与移出各 1 个，5 个保留 URL 前移一位。DragonAI 目标页和品牌域仍均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-fourth-observation-churn-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "第 23 至 24 次严格百度观察相邻波动 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-fourth-observation-churn-2026-08-02.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-twenty-three-observation-query-answer-cards-2026-08-02", "title": "百度 RAG 与微调引用来源 23 次观察查询答案卡（2026-08-02）", "description": "三条严格无品牌百度查询的 23 次冻结观察与持续前三来源逐查询答案卡；另保留原始固定问题未进入百度套件的边界。目标页与品牌域仍为零命中，不构成搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-three-observation-query-answer-cards-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "23 次百度观察查询答案卡 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-three-observation-query-answer-cards-2026-08-02.csv", "encodingFormat": "text/csv" }, { "title": "23 次百度观察查询答案卡 Markdown", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-three-observation-query-answer-cards-2026-08-02.md", "encodingFormat": "text/markdown" } ] }, { "id": "ai-search-rag-baidu-query-host-persistence-twenty-three-observation-2026-08-02", "title": "百度 RAG 与微调引用来源严格无品牌 23 次查询×主机持久性（2026-08-02）", "description": "将 23 次严格无品牌百度观测按查询与规范化主机聚合。28 个查询×主机组合中 11 个覆盖全部 23 次，只有 2 个在全窗口保持精确位置不变，6 个始终位于前 3；DragonAI 目标主机与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-three-observation-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度严格无品牌 23 次查询×主机持久性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-three-observation-2026-08-02.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-twenty-three-observation-persistence-2026-08-02", "title": "百度 RAG 与微调引用来源严格无品牌 23 次查询—URL 持续性（2026-08-02）", "description": "跨 12 个生产版本汇总 23 次严格无品牌百度观察、69 个查询快照和 574 个自然结果位置；48 个查询—URL 组合中有 10 个覆盖全部 23 次，只有 1 个在全窗口保持精确位置不变。DragonAI 目标页与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-three-observation-persistence-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度严格无品牌 23 次查询—URL 持续性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-three-observation-persistence-2026-08-02.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-persistent-top3-source-cards-twenty-three-observation-2026-08-02", "title": "百度 RAG 与微调引用来源严格无品牌 23 次观察持续前三来源卡（2026-08-02）", "description": "将 23 次严格无品牌百度观察中每次都进入前三位的 6 个查询—主机组合，与 12 个具体结果 URL、最新标题及逐文档覆盖确定性联表。三个固定问法各有两个持续前三主机；DragonAI 目标主机与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-three-observation-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度严格无品牌 23 次观察持续前三来源卡 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-twenty-three-observation-2026-08-02.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-persistent-top3-source-cards-2026-08-02", "title": "百度 RAG 与微调引用来源严格无品牌 22 次观察持续前三来源卡（2026-08-02）", "description": "将 22 次严格无品牌百度观察中每次都进入前三位的 6 个查询—主机组合，与 12 个具体结果 URL、最新标题及逐文档覆盖确定性联表。三个固定问法各有两个持续前三主机；DragonAI 目标主机与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度严格无品牌 22 次观察持续前三来源卡 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-persistent-top3-source-cards-2026-08-02.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-query-host-persistence-twenty-two-observation-2026-08-02", "title": "百度 RAG 与微调引用来源严格无品牌 22 次查询×主机持久性（2026-08-02）", "description": "将 22 次严格无品牌百度观测按查询与规范化主机聚合。28 个查询×主机组合中 11 个覆盖全部 22 次，只有 2 个在全窗口保持精确位置不变，6 个始终位于前 3；DragonAI 目标主机与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-two-observation-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度严格无品牌 22 次查询×主机持久性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-query-host-persistence-twenty-two-observation-2026-08-02.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-twenty-two-observation-persistence-2026-08-01", "title": "百度 RAG 与微调引用来源严格无品牌 22 次查询—URL 持续性（2026-08-01）", "description": "跨 11 个生产版本汇总 22 次严格无品牌百度观察、66 个查询快照和 549 个自然结果位置；48 个查询—URL 组合中有 10 个覆盖全部 22 次，只有 1 个在全窗口保持精确位置不变。DragonAI 目标页与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-two-observation-persistence-2026-08-01.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度严格无品牌 22 次查询—URL 持续性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-twenty-two-observation-persistence-2026-08-01.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r5", "title": "百度严格无品牌搜索可比观察累计账本（23 次，修订 5）", "description": "23 次严格百度观察的累计账本：12 个生产版本、69 个查询快照、574 个自然结果位置；目标页与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r5.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "百度严格无品牌搜索可比观察累计账本修订 5 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-02-r5.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r4", "title": "百度严格无品牌搜索可比观察累计账本（22 次，修订 4）", "description": "22 次严格百度观察的累计账本：11 个生产版本、66 个查询快照、549 个自然结果位置；目标页与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r4.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "百度严格无品牌搜索可比观察累计账本 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r4.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r3", "title": "百度严格无品牌搜索可比观察累计账本（21 次，修订 3）", "description": "21 次严格百度观察的累计账本：10 个生产版本、63 个查询快照、524 个自然结果位置；目标页与品牌域均为零命中，不构成 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r3.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "百度严格无品牌搜索可比观察累计账本修订 3 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r3.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-entry-liveness-2026-08-02-055317", "title": "百度第 23 次严格观察新增 URL 存活证据", "description": "第 23 次严格百度观察的相邻结果变动与只读存活证据：1 个 URL 新增、1 个移出；新增 URL 返回 HTTP 200，但最终页面为百度安全验证 HTML。该证据不构成搜索可见性、来源质量、正文可读性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-02-055317.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "百度第 23 次严格观察新增 URL 存活证据 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-02-055317.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-entry-liveness-2026-08-01-222307", "title": "百度第 22 次严格观察新增 URL 存活证据", "description": "第 22 次严格百度观察的相邻结果变动与只读存活证据：2 个新增、2 个移出，新增 URL 中 1 个可读取 HTML、1 个返回 HTTP 404；不构成搜索可见性、来源质量或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-01-222307.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "百度第 22 次严格观察新增 URL 存活证据 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-01-222307.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-twenty-first-observation-entry-liveness-2026-08-01", "title": "百度第 21 次严格观察新增 URL 存活证据", "description": "第 21 次严格百度观察的结果集变动与只读存活证据：2 个新增、2 个移出，新增 URL 返回 HTTP 521 与 403；不构成可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-first-observation-entry-liveness-2026-08-01.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "百度第 21 次严格观察新增 URL 存活证据 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-twenty-first-observation-entry-liveness-2026-08-01.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r2", "title": "百度严格无品牌搜索可比观察累计账本（20 次，修订 2）", "description": "绑定 20 次同口径百度严格无品牌观察、9 次生产发布、60 个查询快照和 499 个自然结果位次；每行保留源报告路径与 SHA-256，并保留目标页和品牌域零命中。该账本未调用模型或执行站外发布，不构成搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r2.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "百度严格无品牌搜索可比观察累计账本 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r2.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r1", "title": "百度严格无品牌搜索可比观察累计账本（19 次，修订 1）", "description": "绑定 19 次同口径百度严格无品牌观察、8 次生产发布、57 个查询快照和 474 个自然结果位次；每行保留源报告路径与 SHA-256，并保留目标页和品牌域零命中。该账本未调用模型或执行站外发布，不构成搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "百度严格无品牌搜索可比观察累计账本 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-comparable-observation-ledger-2026-08-01-r1.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-product-evaluation-suite-template-v1", "title": "AI 产品评测集 JSON 模板 v1", "description": "用于定义产品场景、测试样例、风险切片、指标、阈值、人工复核和生产监测字段的空白模板，并附三条虚构示例。", "url": "https://course.dragonai.tech/datasets/ai-product-evaluation-suite-template-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-product-evaluation-suite-schema-v1", "title": "AI 产品评测套件 JSON Schema v1", "description": "用于自动校验产品级 AI 评测套件必需字段、枚举值、指标定义、切片规则、测试样例与发布决策结构的 JSON Schema Draft 2020-12 文件。", "url": "https://course.dragonai.tech/datasets/ai-product-evaluation-suite-schema-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "dragonai-content-review-policy-v1", "title": "DragonAI 内容审核政策 JSON v1", "description": "公开 DragonAI 知识库五角色审核门槛、权重、硬失败规则、人工批准边界、bundle-v2 绑定范围和一条已发布内容示例回执的机器可读文件。", "url": "https://course.dragonai.tech/datasets/dragonai-content-review-policy-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-product-manager-readiness-matrix-v1", "title": "AI 产品经理八维能力与产物矩阵 v1", "description": "把问题定义、产品策略、用户交互、数据与上下文、原型系统、评测实验、协同采用、治理运营八类能力映射为可观察产物和验证动作。", "url": "https://course.dragonai.tech/datasets/ai-product-manager-readiness-matrix-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-product-manager-eight-week-execution-roadmap-v1", "title": "AI 产品经理八周执行路线图 v1", "description": "把同一个 AI 产品练习按问题、策略、交互、数据、原型、评测、发布和治理拆成八周，逐周记录输入、最低产物、验收门禁、失败回退和阶段交接证据。", "url": "https://course.dragonai.tech/datasets/ai-product-manager-eight-week-execution-roadmap-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-product-manager-portfolio-evidence-template-v1", "title": "AI 产品经理作品集六张证据卡模板 v1", "description": "为每个作品集项目分别记录来源类型、证据环境、真实用户、生产部署、数据条件，以及问题决策、个人贡献、用户交互、系统数据、评测发布和失败复盘六张证据卡。", "url": "https://course.dragonai.tech/datasets/ai-product-manager-portfolio-evidence-template-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-product-decision-defense-cards-v1", "title": "AI 产品决策十二道答辩卡 v1", "description": "用于口头答辩和项目复盘的十二道决策卡，逐题记录结论、依据、取舍、失败、下一步、证据版本和复核问题，不生成总分或能力等级。", "url": "https://course.dragonai.tech/datasets/ai-product-decision-defense-cards-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-product-decision-ownership-map-v1", "title": "AI 产品七类决策协作图 v1", "description": "用七类决策记录产品判断、协作输入、专业责任、最小产物、失败追问和复审条件，不生成能力等级或职业结果预测。", "url": "https://course.dragonai.tech/datasets/ai-product-decision-ownership-map-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-pm-course-due-diligence-checklist-v1", "title": "AI 产品经理课程报名尽调清单 v1", "description": "用于逐项核验目标匹配、课程产物、公开样例、反馈机制、师资证据、学习条件、结果主张、费用、合同退款、变更退出、个人信息与宣传申诉的空白清单。", "url": "https://course.dragonai.tech/datasets/ai-pm-course-due-diligence-checklist-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-pm-course-syllabus-audit-matrix-v1", "title": "AI 产品经理课程大纲核验矩阵 v1", "description": "从八类核验模块中按目标确定必需项，映射为可观察学习目标、练习、产物和验收动作，并用缺失、仅提及、有练习、有证据四种状态逐项记录课程覆盖。", "url": "https://course.dragonai.tech/datasets/ai-pm-course-syllabus-audit-matrix-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-product-prd-field-checklist-v1", "title": "AI 产品 PRD 适用字段与验收清单 v1", "description": "把 AI 产品 PRD 拆成决策背景、用户任务、行为合同、数据上下文、评测门禁、失败接管、Agent 权限、治理、运行约束和发布运营十个可核验部分。", "url": "https://course.dragonai.tech/datasets/ai-product-prd-field-checklist-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "rag-product-decision-checklist-v1", "title": "RAG 产品决策深度与七项核验清单 v1", "description": "用解释、规格化、评测诊断、生产运营四级深度，以及选型、语料治理、摄取索引、检索、生成交互、评测、运行七项决策核验产品经理的 RAG 实践能力。", "url": "https://course.dragonai.tech/datasets/rag-product-decision-checklist-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "rag-fine-tuning-decision-matrix-v1", "title": "RAG、微调与混合方案决策矩阵 v1", "description": "用问题类型、更新频率、来源追溯、数据条件、评测、成本延迟、运维和组合必要性八项字段比较基础方案、RAG、微调与混合方案。", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-decision-matrix-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "rag-fine-tuning-quick-selector-v1", "title": "RAG 与微调八情景快速选型表 v1", "description": "把八种常见选型情景映射为继续基线、验证 RAG、验证微调、消融后验证组合或先补数据五类动作，并逐项给出必须验证的证据与停止条件。", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-quick-selector-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "RAG 与微调八情景快速选型表（CSV）", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-quick-selector-v1.csv", "encodingFormat": "text/csv" } ] }, { "id": "rag-fine-tuning-reliable-source-evidence-map-v1", "title": "RAG 与微调可靠来源—决策—验证映射 v1", "description": "把百度智能云、腾讯云文档及四份国际一手技术与治理来源逐项映射为可支持的选型判断、应执行的验证动作和不可推出的结论，绑定固定 RAG 与微调选型问题。", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-reliable-source-evidence-map-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "RAG 与微调可靠来源—决策—验证映射（CSV）", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-reliable-source-evidence-map-v1.csv", "encodingFormat": "text/csv" }, { "title": "RAG 与微调来源声明范围验证器（Python）", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-source-claim-validator-v1.py", "encodingFormat": "text/x-python" }, { "title": "RAG 与微调来源声明范围验证器夹具（JSON）", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-source-claim-validator-v1-fixtures.json", "encodingFormat": "application/json" }, { "title": "RAG 与微调来源声明范围验证器说明（Markdown）", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-source-claim-validator-v1-guide.md", "encodingFormat": "text/markdown" } ] }, { "id": "rag-fine-tuning-selection-reference-v1", "title": "RAG、微调与组合方案可执行选型参考 v1", "description": "零依赖 Python 决策合同：严格校验 12 个布尔事实与两条跨字段不变量；两个候选通过同一门槛后，在同一评测集、负载、成本周期和评分量表下对五项实测值做无权重、容差感知的 Pareto 比较。12 个选择用例、4 个比较用例与 5 个拒绝用例共同验证失败关闭。", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-reference-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "RAG、微调与组合方案可运行 Python 参考实现", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-reference-v1.py", "encodingFormat": "text/x-python" }, { "title": "RAG、微调与组合方案机器回归夹具", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-reference-v1-fixtures.json", "encodingFormat": "application/json" }, { "title": "RAG、微调与组合方案可执行选型说明", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-reference-v1-guide.md", "encodingFormat": "text/markdown" } ] }, { "id": "ai-search-rag-baidu-selection-query-current-delta-2026-08-13", "title": "百度无品牌 RAG 微调选型查询当前变化 v1", "description": "匿名只读复核固定无品牌选型查询，逐位次记录 8 个当前结果，并与 2026-08-12 基线比较 4 个留存、4 个进入、4 个退出和完全变化的前三主机。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-13.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度无品牌 RAG 微调选型查询当前变化（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-13.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-selection-query-current-delta-2026-08-13-r2", "title": "百度无品牌 RAG 微调选型查询当前变化 r2", "description": "匿名只读复核固定无品牌选型查询：相对 2026-08-13 01:11 UTC 基线，当前 8 条自然结果保留 4、进入 4、退出 4，2 条留存 URL 换位，前三主机为腾讯云/百度智能云/百家号；DragonAI 目标页、品牌域与生成式摘要文本均为 0，且本轮未捕获可核对引用链接。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-13-r2.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度无品牌 RAG 微调选型查询当前变化 r2（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-13-r2.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-selection-query-current-delta-2026-08-13-r3", "title": "百度无品牌 RAG 微调选型查询同发布变化 r3", "description": "2026-08-13 22:00 UTC 匿名只读复核固定选型查询：相对 21:22 UTC 同发布观察，8 条自然结果保留 4、进入 4、退出 4，3 条留存 URL 换位；前三主机为腾讯云、百家号和智源社区，第 4 位新进入标题直接命中 RAG 和微调选型的知乎页面。DragonAI 目标页和品牌域均为 0。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-13-r3.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度无品牌 RAG 微调选型查询同发布变化 r3（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-13-r3.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-selection-query-current-delta-2026-08-14-r4", "title": "百度无品牌 RAG 微调选型查询同发布变化 r4", "description": "2026-08-14 00:53 UTC 匿名只读复核固定选型查询：相对 2026-08-13 22:00 UTC 同发布观察，8 条自然结果保留 7、进入 1、退出 1，前六位 URL 与位次不变；第 7 位进入什么值得买的 RAG、微调与长上下文决策框架，原第 7 位百家号结果退出，Red Hat 保持第 8 位。DragonAI 目标页和品牌域均为 0。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-14-r4.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度无品牌 RAG 微调选型查询同发布变化 r4（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-14-r4.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-selection-query-current-delta-2026-08-14-r5", "title": "百度无品牌 RAG 微调选型查询同发布变化 r5", "description": "2026-08-14 12:53 UTC 匿名只读复核固定选型查询：相对同日 00:53 UTC 的同发布观察，8 条自然结果保留 7、进入 1、退出 1，其中 1 条留存结果下移；前六位 URL 与位次不变，第 7 位进入新的 CSDN 企业级 RAG 技术方案，什么值得买从第 7 位移至第 8 位，Red Hat 退出。DragonAI 目标页和品牌域均为 0。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-14-r5.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度无品牌 RAG 微调选型查询同发布变化 r5（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-query-current-delta-2026-08-14-r5.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-selection-current-document-features-2026-08-14-r4", "title": "百度当前 RAG 微调选型结果文档交付特征 r4", "description": "对 2026-08-14 00:53 UTC 固定无品牌选型查询的 8 个当前自然结果执行一次只读页面获取：3 个返回可用 HTML 文档、3 个为 HTML 壳、2 个为 HTTP 或请求错误；可用文档中 3 个有摘要、2 个有 canonical、0 个暴露预登记扩展名的下载链接。失败不回填，结构共现不构成排名原因、来源质量或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-current-document-features-2026-08-14-r4.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度当前 RAG 微调选型结果文档交付特征 r4（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-current-document-features-2026-08-14-r4.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-selection-target-delivery-comparison-2026-08-14-r4", "title": "DragonAI 目标页与百度当前选型结果文档交付特征对照 r4", "description": "使用同一只读 HTML 特征提取器，对当前生产目标页与 2026-08-14 00:53 UTC 固定无品牌选型查询八个自然结果中 3 个成功返回的 HTML 文档进行 16 项交付特征对照。目标页有 3 项高于全部可用竞争文档；5 个失败或 HTML 壳响应不插补。结构差异不构成排名原因、来源质量、目标可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-target-delivery-comparison-2026-08-14-r4.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "DragonAI 目标页与百度当前选型结果文档交付特征对照 r4（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-target-delivery-comparison-2026-08-14-r4.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-selection-current-document-features-2026-08-13-r3", "title": "百度当前 RAG 微调选型结果文档交付特征 r3", "description": "对 2026-08-13 22:00 UTC 固定无品牌选型查询的 8 个自然结果执行一次只读页面获取：3 个返回可用 HTML 文档、3 个为 HTML 壳、2 个返回 HTTP 错误；可用文档中 3 个有摘要、2 个有 canonical、0 个暴露预登记扩展名的下载链接。失败不回填，结构共现不构成排名原因、来源质量或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-current-document-features-2026-08-13-r3.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度当前 RAG 微调选型结果文档交付特征 r3（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-current-document-features-2026-08-13-r3.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-selection-target-delivery-comparison-2026-08-13-r3", "title": "DragonAI 目标页与百度当前选型结果文档交付特征对照 r3", "description": "使用同一只读 HTML 特征提取器，对当前生产目标页与固定无品牌选型查询八个自然结果中 3 个成功返回的 HTML 文档进行 16 项交付特征对照。目标页有 3 项高于全部可用竞争文档；5 个失败或 HTML 壳响应不插补。结构差异不构成排名原因、来源质量、目标可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-target-delivery-comparison-2026-08-13-r3.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "DragonAI 目标页与百度当前选型结果文档交付特征对照 r3（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-selection-target-delivery-comparison-2026-08-13-r3.csv", "encodingFormat": "text/csv" } ] }, { "id": "rag-citation-traceability-reference-v1", "title": "RAG 引用可追溯最小参考实现 v1", "description": "零依赖 Python 参考实现：把逐条陈述绑定到稳定片段 ID 与来源 URL，并在无匹配证据时失败关闭；附可直接运行的自测试与实现边界。", "url": "https://course.dragonai.tech/datasets/rag-citation-traceability-reference-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "RAG 引用可追溯可运行 Python 参考实现", "url": "https://course.dragonai.tech/datasets/rag-citation-traceability-reference-v1.py", "encodingFormat": "text/x-python" }, { "title": "RAG 引用可追溯参考实现说明", "url": "https://course.dragonai.tech/datasets/rag-citation-traceability-reference-v1-guide.md", "encodingFormat": "text/markdown" }, { "title": "RAG 引用可追溯机器可读测试夹具", "url": "https://course.dragonai.tech/datasets/rag-citation-traceability-reference-v1-fixtures.json", "encodingFormat": "application/json" } ] }, { "id": "ai-search-rag-finetuning-citation-sample-2026-07", "title": "千问、Kimi、Gemini 的 RAG 与微调选型引用样本（2026-07）", "description": "同一中文无品牌问题在千问、Kimi、Gemini 各运行三次后的引用计数、主机名分布、平台交集、目标引用与方法限制。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-finetuning-citation-sample-2026-07.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-search-citation-benchmark-snapshot-2026-07-22", "title": "千问、Kimi、Gemini 固定引用问题双窗口基准（2026-07-21—22）", "description": "同一固定中文无品牌元问题在正式发布前基线 B 与发布后时间盒 C 的18条逐运行记录、具体引用 URL、零目标结果和模型漂移边界。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-snapshot-2026-07-22.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "正式窗口汇总 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-formal-window-summary-2026-07-22.csv", "encodingFormat": "text/csv" }, { "title": "逐运行引用 URL CSV", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-run-level-urls-2026-07-22.csv", "encodingFormat": "text/csv" }, { "title": "按窗口与平台汇总的引用主机频次和运行覆盖 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-host-frequency-2026-07-22.csv", "encodingFormat": "text/csv" }, { "title": "平台间与跨窗口引用主机集合重叠 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-host-overlap-2026-07-22.csv", "encodingFormat": "text/csv" }, { "title": "逐 URL 复现与跨窗口持续性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-url-persistence-2026-07-22.csv", "encodingFormat": "text/csv" }, { "title": "逐声明证据定位索引（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-claim-evidence-index-2026-07-22.csv", "encodingFormat": "text/csv" }, { "title": "指标、分母、公式与跨窗口可比性定义（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-metric-definitions-2026-07-22.csv", "encodingFormat": "text/csv" }, { "title": "逐声明证据溯源图（JSON-LD）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-claim-evidence-graph-2026-07-22.json", "encodingFormat": "application/ld+json" }, { "title": "三平台主要引用主机证据（JSON）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-leading-hosts-2026-07-22.json", "encodingFormat": "application/json" }, { "title": "四条无品牌检索问法 × 三平台主要引用主机十六行证据表（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-leading-hosts-2026-07-22.csv", "encodingFormat": "text/csv" }, { "title": "固定无品牌查询的直接答案、复现协议与证据卡（JSON）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-exact-query-answer-2026-07-22.json", "encodingFormat": "application/json" }, { "title": "固定无品牌查询直接答案 JSON Schema", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-exact-query-answer-schema-2026-07-22.json", "encodingFormat": "application/schema+json" }, { "title": "十九条可引用事实、证据片段与规则片段扁平表（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-citation-ready-claims-2026-07-22.csv", "encodingFormat": "text/csv" }, { "title": "完整研究存档包（含 README、CITATION.cff、校验清单与仓库元数据）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-archive-2026-07-22.zip", "encodingFormat": "application/zip" }, { "title": "BagIt 1.0 数字保存包（含 Payload-Oxum 与 SHA-256 payload/tag manifests）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-bagit-2026-07-22.zip", "encodingFormat": "application/zip" }, { "title": "JATS 1.3 machine-readable benchmark article", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-report-2026-07-22.jats.xml", "encodingFormat": "application/xml" }, { "title": "OAI Dublin Core repository-ingestion metadata", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-oai-dublin-core-2026-07-22.xml", "encodingFormat": "application/xml" }, { "title": "推荐引用元数据（CITATION.cff）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-citation-2026-07-22.cff", "encodingFormat": "text/yaml" }, { "title": "CSL-JSON 引用元数据（UTF-8）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-citation-2026-07-22.csl.json", "encodingFormat": "application/vnd.citationstyles.csl+json" }, { "title": "RO-Crate 1.3 研究对象元数据（JSON-LD）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-ro-crate-metadata-2026-07-22.json", "encodingFormat": "application/ld+json" }, { "title": "BibTeX 引用元数据（UTF-8）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-citation-2026-07-22.bib", "encodingFormat": "application/x-bibtex" }, { "title": "RIS 引用元数据（UTF-8）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-citation-2026-07-22.ris", "encodingFormat": "application/x-research-info-systems" }, { "title": "研究存档清单（MANIFEST.json）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-manifest-2026-07-22.json", "encodingFormat": "application/json" }, { "title": "文件校验和（SHA256SUMS）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-sha256sums-2026-07-22.txt", "encodingFormat": "text/plain" }, { "title": "研究存档复核说明（README）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-readme-2026-07-22.md", "encodingFormat": "text/markdown" }, { "title": "中文固定问题答案、证据地图与复核说明（README.zh-CN）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-readme-zh-cn-2026-07-22.md", "encodingFormat": "text/markdown" }, { "title": "五十四个表格投影的数据字典（CODEBOOK）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-codebook-2026-07-22.md", "encodingFormat": "text/markdown" }, { "title": "固定问题引用基准的完整方法与证据边界（METHODS）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-methods-2026-07-22.md", "encodingFormat": "text/markdown" }, { "title": "权威基准快照 JSON Schema（Draft 2020-12）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-snapshot-schema-v1.1.0.json", "encodingFormat": "application/schema+json" }, { "title": "CC BY 4.0 许可、署名与第三方权利边界说明", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-license-2026-07-22.txt", "encodingFormat": "text/plain" }, { "title": "仓库提交元数据（Zenodo JSON）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-zenodo-metadata-2026-07-22.json", "encodingFormat": "application/json" }, { "title": "DataCite 4.7 仓库元数据模板（JSON API）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-datacite-metadata-2026-07-22.json", "encodingFormat": "application/json" }, { "title": "RFC 9264 研究对象 Linkset（JSON）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-linkset-2026-07-22.json", "encodingFormat": "application/linkset+json" }, { "title": "MLCommons Croissant 1.0 数据集元数据（JSON-LD）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-croissant-metadata-2026-07-22.json", "encodingFormat": "application/ld+json" }, { "title": "Frictionless Tabular Data Package 元数据（JSON）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-datapackage-2026-07-22.json", "encodingFormat": "application/json" }, { "title": "W3C DCAT 3 dataset catalog record (JSON-LD)", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-dcat-metadata-2026-07-22.json", "encodingFormat": "application/ld+json" }, { "title": "双窗口汇总、主机频次、集合重叠与 URL 持续性离线复算脚本", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-reproduce-summary-2026-07-22.py", "encodingFormat": "text/x-python" }, { "title": "提供商窗口核心指标摘要表（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-provider-window-summary-2026-07-22.csv", "encodingFormat": "text/csv" }, { "title": "B v2 与 C 历史口径不可直接比较摘要图", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-summary-2026-07-22.svg", "encodingFormat": "image/svg+xml" } ] }, { "id": "ai-search-citation-benchmark-observation-2026-07-28", "title": "千问、Kimi、Gemini 固定问题非因果补充观察（2026-07-28）", "description": "同一固定中文无品牌问题的九次真实提供商输出补充观察，逐运行保留具体引用 URL、零目标结果、竞争主机、模型漂移、归因可靠性与非因果边界；不是新的正式 B/C 窗口。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-2026-07-28.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "九次补充观察逐运行引用 URL（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-2026-07-28.csv", "encodingFormat": "text/csv" }, { "title": "九次补充观察竞争主机频次、运行覆盖与来源类型审计（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-host-frequency-2026-07-28.csv", "encodingFormat": "text/csv" }, { "title": "3 providers x 5 source types: citation occurrences, hosts and citation-surface coverage (CSV)", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-source-type-summary-2026-07-28.csv", "encodingFormat": "text/csv" }, { "title": "46-host x 3-provider citation-occurrence and run-coverage matrix with explicit zero cells (CSV)", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-provider-host-matrix-2026-07-28.csv", "encodingFormat": "text/csv" }, { "title": "9-run x 46-host citation incidence matrix with explicit zero cells (CSV)", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-run-host-incidence-2026-07-28.csv", "encodingFormat": "text/csv" }, { "title": "Within-provider same-run competitor-host co-citation edge list (CSV)", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-host-cocitation-edges-2026-07-28.csv", "encodingFormat": "text/csv" }, { "title": "138-row provider by competitor-host co-citation node profile with explicit zero cells (CSV)", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-host-cocitation-node-profile-2026-07-28.csv", "encodingFormat": "text/csv" }, { "title": "Three-provider co-citation graph structure summary with density, components and explicit zero denominators (CSV)", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-host-cocitation-graph-summary-2026-07-28.csv", "encodingFormat": "text/csv" }, { "title": "Nine-run real-provider source-mix and co-citation evidence figure (SVG)", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-observation-summary-2026-07-28.svg", "encodingFormat": "image/svg+xml" } ] }, { "id": "ai-search-citation-benchmark-five-batch-run-level-2026-07-30", "title": "千问、Kimi、Gemini 五批次 45 次运行逐引用 URL 数据", "description": "把同一固定中文无品牌问题的五个真实提供商批次统一为 45 条运行记录与 552 次具体引用 URL 出现；JSON 保留逐运行引用数组，566 行 CSV 为 14 次零引用运行各保留一条显式空 URL 行，便于完整重算分母。DragonAI 目标引用与品牌提及均为零。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-five-batch-run-level-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "五批次逐运行引用 URL 与显式零运行（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-five-batch-run-level-2026-07-30.csv", "encodingFormat": "text/csv" }, { "title": "五批次 AI 引用与百度六轮严格查询当前证据摘要图（SVG）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-current-evidence-summary-2026-07-30.svg", "encodingFormat": "image/svg+xml" }, { "title": "五批次 AI 引用与百度九次严格观察证据摘要图（SVG）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-nine-observation-evidence-summary-2026-07-30.svg", "encodingFormat": "image/svg+xml" } ] }, { "id": "ai-search-citation-benchmark-five-batch-url-profile-2026-07-30", "title": "千问、Kimi、Gemini 五批次具体引用 URL 跨平台画像", "description": "把五批次 45 次运行的 552 次具体引用 URL 出现压缩为 194 个精确 URL 画像，逐行保留平台、批次、运行与出现次数。193 个 URL 仅由一家平台展示，只有 arxiv.org/abs/2312.10997 同时由 Kimi 与 Gemini 展示；11 个 URL 覆盖全部五批次，DragonAI 目标 URL 仍为零。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-five-batch-url-profile-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "五批次具体引用 URL 跨平台覆盖画像（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-five-batch-url-profile-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-citation-benchmark-url-delivery-shape-2026-07-30", "title": "千问、Kimi、Gemini 五批次引用 URL 交付形态画像", "description": "将五批次 45 次运行中的 194 个精确引用 URL 确定性拆解为路径深度、认可文件后缀、查询参数、片段和 HTTPS 字段。182 个 URL 是无认可文件后缀的深层页面，156 个至少有两级路径；DragonAI 目标页形态与最高频类别一致，但 45 次运行中的目标引用仍为零。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-url-delivery-shape-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "五批次逐引用 URL 交付形态画像（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-url-delivery-shape-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-citation-retrieval-evidence-cards-2026-07-30", "title": "千问、Kimi、Gemini RAG 与微调引用来源中文检索证据卡", "description": "把四条中文无品牌问法绑定到同一审核答案、八条可引用声明、逐项证据 URL、百度严格观察状态和显式非成功边界。JSON 适合检索系统导入，Markdown 适合人工复核；它不构成搜索可见性、外链证据或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-citation-retrieval-evidence-cards-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "四条中文无品牌问法的可引用证据卡（Markdown）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-citation-retrieval-evidence-cards-2026-07-30.md", "encodingFormat": "text/markdown" } ] }, { "id": "ai-search-citation-benchmark-provider-overlap-matrix-2026-07-30", "title": "千问、Kimi、Gemini 五批次引用 URL 与主机跨平台重合矩阵", "description": "把五批次 45 次运行中 194 个精确引用 URL 和 114 个主机压缩为三个平台对；逐对公开精确 URL 与主机的交集、并集、Jaccard 和重合系数。只有 Kimi × Gemini 共享 1 个精确 URL；千问 × Kimi 仅在主机层共享 CSDN 与知乎专栏。目标引用仍为零。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-provider-overlap-matrix-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "三平台对精确 URL 与主机重合矩阵（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-provider-overlap-matrix-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-citation-benchmark-provider-concentration-2026-07-30", "title": "千问、Kimi、Gemini 五批次引用 URL 与主机集中度", "description": "基于同一固定问题的五批次、45 次真实提供商运行和 552 次具体引用 URL 出现，逐平台发布 URL 与主机层的 Top-1、Top-5 占比、HHI 和有效来源数。主机 HHI 在本窗口中依次为千问 0.157644、Kimi 0.142400、Gemini 0.031173；这是冻结观察的描述性集中度，不代表来源质量、长期偏好、排序因果或 GEO 成功，目标引用与品牌提及仍为零。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-provider-concentration-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "三平台引用 URL 与主机集中度（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-provider-concentration-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-citation-benchmark-provider-host-persistence-2026-07-30", "title": "千问、Kimi、Gemini 五批次固定问题引用主机持续性矩阵", "description": "汇总五个同查询批次、45 次真实提供商运行与 552 次具体引用 URL 出现，将 3 家平台和 114 个观察主机展开为 342 行显式零矩阵；逐行保留五批次出现向量、运行覆盖、来源类型、零目标结果和非长期偏好边界。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-provider-host-persistence-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "五批次提供商 × 引用主机显式零矩阵（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-provider-host-persistence-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-cross-surface-host-persistence-2026-07-30", "title": "五批 AI 引用域 × 百度六轮严格无品牌结果跨表面持久性交集", "description": "将五批次、45 次固定问题 AI 引用主机与六轮、18 个百度严格无品牌查询快照按规范化主机全外联结为 123 行显式零面板；6 个主机跨表面重合，只有 cloud.tencent.com 同时满足 AI 某提供商覆盖五批与百度至少一个 URL 覆盖六轮，DragonAI 两侧均未出现。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "123 行跨表面主机持久性显式零面板（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-cross-surface-source-type-summary-2026-07-30", "title": "五批 AI 引用域 × 百度六轮严格无品牌结果跨表面来源类型汇总", "description": "将 123 行跨表面主机持久性面板压缩为 6 个互斥来源类型；6 个跨表面重合主机中，厂商文档 3 个、专家社区 2 个、其他实践者或出版者 1 个，唯一双侧持续主机 cloud.tencent.com 属于厂商文档。百度侧 9 个没有既有 AI 来源类型的主机保留为未分类，不做推断式补标；DragonAI 两侧均未出现。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-source-type-summary-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "6 行跨表面来源类型汇总（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-source-type-summary-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-citation-benchmark-baseline-attribution-v2", "title": "千问、Kimi、Gemini 固定问题发布前基线：引用归因规则 v2 离线重算", "description": "不重新调用模型，按答案可归因规则重算固定 B 基线的九次原始响应；逐运行绑定回答哈希、原始载荷哈希、引用 URL、旧规则差值和零目标引用结果。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-baseline-attribution-v2.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-search-citation-benchmark-attribution-lineage-v1", "title": "千问、Kimi、Gemini 固定问题引用资产归因版本与可比性谱系 v1", "description": "逐项登记 26 个公开分析资产的归因规则版本、SHA-256、严格趋势端点资格和替代关系；仅 B 窗口 v2 重算基线可作为未来同口径严格端点，C 窗口与补充观察保持历史未版本化边界。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-attribution-lineage-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "归因版本与可比性谱系 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-citation-benchmark-attribution-lineage-v1.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-citation-source-type-taxonomy-v1", "title": "千问、Kimi、Gemini 固定问题引用来源类型透明分类 v1", "description": "基于正式 B/C 双窗口引用主机频次表，按公开主机名后缀规则把 92 个平台窗口主机行、58 个唯一主机名和 230 次引用出现映射为五类来源，并明确回退分类、模型漂移与非质量排名边界。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-source-type-taxonomy-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "平台窗口来源类型汇总 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-citation-source-type-provider-window-v1.csv", "encodingFormat": "text/csv" }, { "title": "逐主机来源类型审计 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-citation-source-type-host-audit-v1.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-citation-url-source-type-persistence-v1", "title": "Qwen、Kimi、Gemini 逐 URL 持续性与来源类型联结数据 v1", "description": "将固定 B/C 双窗口的 98 个“平台 × 具体引用 URL”持续性记录，与公开主机名后缀来源类型规则确定性联结；逐行保留平台、模型、精确 URL、来源类型、两窗运行覆盖、严格可比门禁和非质量排名边界。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-url-source-type-persistence-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "逐 URL 持续性与来源类型联结 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-citation-url-source-type-persistence-v1.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-provider-citation-surface-crosswalk-v1", "title": "千问、Kimi、Gemini 官方来源返回表面与基准适配器对照 v1", "description": "逐行对照三家第一方文档公开的来源返回表面、DragonAI 固定问题基准的适配器与 URL 提取路径，并明确产品帮助页、API 字段、索引、候选集和实际引用之间的推断边界。", "url": "https://course.dragonai.tech/datasets/ai-search-provider-citation-surface-crosswalk-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "三平台官方来源返回表面与基准适配器对照 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-provider-citation-surface-crosswalk-v1.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-unbranded-baseline-2026-07-29", "title": "百度 RAG 与微调引用来源严格无品牌搜索基线（2026-07-29）", "description": "公开 3 条严格无品牌查询和 1 条 site 诊断查询在百度匿名自然搜索结果中的逐位次记录、规范化目标匹配结果与结果集哈希；保留目标页未出现的零结果。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-unbranded-baseline-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度 RAG 与微调引用来源严格搜索基线 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-unbranded-baseline-2026-07-29.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-host-query-matrix-2026-07-29", "title": "百度严格无品牌查询 × 观察主机出现矩阵（2026-07-29）", "description": "从既有百度严格无品牌搜索基线确定性派生 13 个观察主机在 3 条查询中的出现、缺席、次数与位次；site: 诊断被排除在全部分母之外，矩阵不构成来源质量排名、权威证据、推荐或 GEO 成果。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-host-query-matrix-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度严格无品牌查询 × 观察主机出现矩阵 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-host-query-matrix-2026-07-29.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-source-surface-matrix-2026-07-29", "title": "百度严格无品牌查询 × 来源表面类型矩阵（2026-07-29）", "description": "从同一百度严格无品牌基线按预登记主机名后缀确定性派生 3 条查询 × 5 类来源表面及总体汇总共 20 行；25 个结果分为社区/专栏 8、厂商/云技术站 8、发布平台/门户 6、机构研究枢纽 2、其他观察主机 1。分类不确认发布者身份、编辑控制、质量、权威、推荐、因果排名因素或 GEO 成果。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-source-surface-matrix-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度严格无品牌查询 × 来源表面类型矩阵 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-source-surface-matrix-2026-07-29.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-title-term-matrix-2026-07-29", "title": "百度严格无品牌自然结果标题词覆盖矩阵（2026-07-29）", "description": "从同一百度严格无品牌基线确定性派生 25 个自然结果标题 × 7 个固定问题概念的显式零矩阵；RAG 命中 20 个标题、微调 11 个、引用 2 个、来源 0 个。字面命中不构成排名原因、来源质量、索引、正向搜索可见性或 GEO 成果。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-title-term-matrix-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度严格无品牌自然结果标题词覆盖矩阵 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-title-term-matrix-2026-07-29.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-evidence-join-2026-07-29", "title": "百度严格无品牌自然结果逐行联合证据表（2026-07-29）", "description": "按 queryId + position + url 唯一键，把同一冻结百度基线的 3 条严格查询、25 个自然位次、标题、URL、规范化主机、来源表面分类和 7 个标题概念状态合并为逐结果证据表；保留来源哈希与显式非成功边界。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-evidence-join-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度严格无品牌自然结果逐行联合证据表 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-evidence-join-2026-07-29.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-churn-2026-07-29", "title": "百度 RAG 与微调引用来源严格无品牌结果集波动（2026-07-29）", "description": "比较同一生产发布、同一固定严格无品牌查询套件在 2026-07-29 的两次只读观察：两次各有 25 个自然结果位次，22 个 URL 被保留、3 个新增、3 个移除，其中 2 个保留 URL 位次移动；DragonAI 目标页两次均未出现，结果集波动不构成目标排名变化或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-churn-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度 RAG 与微调引用来源严格无品牌结果集波动 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-churn-2026-07-29.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-adjacent-churn-2026-07-30", "title": "百度 RAG 与微调引用来源严格无品牌第三至第四时点相邻区间波动（2026-07-30）", "description": "比较同一生产发布、同一固定严格无品牌查询套件的第三与第四次只读观察：两次各有 25 个自然结果位次，25 个 URL 全部留存、无新增或移除；第三条查询有 2 个 URL 在第 7、8 位互换，另外两条查询完全不变。DragonAI 目标页与 site 诊断仍未出现，竞争结果换位不构成目标排名变化、正向搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-adjacent-churn-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度 RAG 与微调引用来源严格无品牌第三至第四时点相邻区间波动 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-adjacent-churn-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-fourth-to-fifth-churn-2026-07-30", "title": "百度 RAG 与微调引用来源严格无品牌第四至第五时点相邻区间波动（2026-07-30）", "description": "比较同一生产发布、同一固定严格无品牌查询套件的第四与第五次只读观察：前两条查询结果集不变；第三条查询保留 6 个 URL、新增 2 个、移除 2 个，3 个保留 URL 位次后移。DragonAI 目标页与 site 诊断仍未出现，竞争结果更替不构成目标排名变化、正向搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-fourth-to-fifth-churn-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度 RAG 与微调引用来源严格无品牌第四至第五时点相邻区间波动 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-fourth-to-fifth-churn-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-four-observation-persistence-2026-07-30", "title": "百度 RAG 与微调引用来源严格无品牌四时点 URL 持续性（2026-07-30）", "description": "合并同一生产发布与固定严格无品牌查询套件的四次百度自然结果观察：12 个查询快照共 100 个自然位次、29 个查询—URL 组合；22 个组合覆盖 4 / 4 次观察，其中 20 个四次位次完全相同，3 个组合在已出现窗口中存在位次变化。DragonAI 目标页四次均未出现，竞争 URL 持续或换位不构成排名因果、正向目标可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-four-observation-persistence-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度 RAG 与微调引用来源严格无品牌四时点 URL 持续性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-four-observation-persistence-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-fifth-to-sixth-churn-2026-07-30", "title": "百度 RAG 与微调引用来源严格无品牌第五至第六时点相邻区间波动（2026-07-30）", "description": "比较同一生产发布、同一固定严格无品牌查询套件的第五与第六次只读观察：两次各有 25 个自然结果位次，共保留 23 个 URL、新增 2 个、移除 2 个；第一条查询另有 1 个保留 URL 从第 3 位移至第 2 位，第二条查询无保留 URL 换位，第三条查询完全不变。DragonAI 目标页与 site 诊断仍未出现，竞争结果更替不构成目标排名变化、正向搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-fifth-to-sixth-churn-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度 RAG 与微调引用来源严格无品牌第五至第六时点相邻区间波动 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-fifth-to-sixth-churn-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-cross-release-churn-2026-07-30", "title": "百度 RAG 与微调引用来源严格无品牌跨发布相邻结果变动（2026-07-30）", "description": "比较固定三条严格无品牌查询在相邻两次匿名百度自然结果观察中的变化：两次各有 25 个自然结果位次，共保留 19 个 URL、新增 6 个、移除 6 个，7 个保留 URL 位次变化。两次观察跨越两个生产发布，DragonAI 目标页均未出现；该数据不支持发布效果、排名因素、正向目标可见性或 GEO 成功归因。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-cross-release-churn-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度 RAG 与微调引用来源严格无品牌跨发布相邻结果变动 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-cross-release-churn-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-five-observation-persistence-2026-07-30", "title": "百度 RAG 与微调引用来源严格无品牌五时点 URL 持续性（2026-07-30）", "description": "合并同一生产发布与固定严格无品牌查询套件的五次百度自然结果观察：15 个查询快照共 125 个自然位次、31 个查询—URL 组合；20 个组合覆盖 5 / 5 次观察，其中 15 个五次位次完全相同，6 个组合在已出现窗口中存在位次变化。DragonAI 目标页五次均未出现，竞争 URL 持续、进入、退出或换位不构成排名因果、正向目标可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-five-observation-persistence-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度 RAG 与微调引用来源严格无品牌五时点 URL 持续性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-five-observation-persistence-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-six-observation-persistence-2026-07-30", "title": "百度 RAG 与微调引用来源严格无品牌六时点 URL 持续性（2026-07-30）", "description": "合并同一生产发布与固定严格无品牌查询套件的六次百度自然结果观察：18 个查询快照共 150 个自然位次、33 个查询—URL 组合；18 个组合覆盖 6 / 6 次观察，其中 12 个六次位次完全相同，7 个组合在已出现窗口中存在位次变化。DragonAI 目标页六次均未出现，竞争 URL 持续、进入、退出或换位不构成排名因果、正向目标可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-six-observation-persistence-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度 RAG 与微调引用来源严格无品牌六时点 URL 持续性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-six-observation-persistence-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-eight-observation-persistence-2026-07-30", "title": "百度 RAG 与微调引用来源严格无品牌八时点 URL 持续性（2026-07-30）", "description": "汇总三个生产发布上的八次匿名百度自然结果观察：24 个严格查询快照共 200 个自然结果位次，形成 36 个查询—URL 组合；16 个组合覆盖 8/8 次观察，其中 7 个八次位次完全相同，12 个组合在已出现窗口中发生位次变化。DragonAI 目标页八次均未出现；竞争 URL 持续、进入、退出或换位不构成排名因果、正向目标可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-eight-observation-persistence-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度 RAG 与微调引用来源严格无品牌八时点 URL 持续性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-eight-observation-persistence-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-nine-observation-persistence-2026-07-30", "title": "百度 RAG 与微调引用来源严格无品牌九时点 URL 持续性（2026-07-30）", "description": "汇总三个生产发布上的九次匿名百度自然结果观察：27 个严格查询快照共 225 个自然结果位次，形成 37 个查询—URL 组合；15 个组合覆盖 9/9 次观察，其中 7 个九次位次完全相同，12 个组合在已出现窗口中发生位次变化。DragonAI 目标页九次均未出现；竞争 URL 持续、进入、退出或换位不构成排名因果、正向目标可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-nine-observation-persistence-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度 RAG 与微调引用来源严格无品牌九时点 URL 持续性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-nine-observation-persistence-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-ten-observation-persistence-2026-07-31", "title": "百度 RAG 与微调引用来源严格无品牌十时点 URL 持续性（2026-07-31）", "description": "汇总三个生产发布上的十次匿名百度自然结果观察：30 个严格查询快照共 250 个自然结果位次，形成 39 个查询—URL 组合；14 个组合覆盖 10/10 次观察，其中 5 个十次位次完全相同，13 个组合在已出现窗口中发生位次变化。DragonAI 目标页十次均未出现；竞争 URL 持续、进入、退出或换位不构成排名因果、正向目标可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-ten-observation-persistence-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度 RAG 与微调引用来源严格无品牌十时点 URL 持续性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-ten-observation-persistence-2026-07-31.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-ninth-to-tenth-churn-2026-07-31", "title": "百度 RAG 与微调引用来源严格无品牌第九至第十时点相邻区间波动（2026-07-31）", "description": "比较同一生产发布、同一匿名百度自然搜索表面、同一三条严格无品牌查询的第九与第十次只读观察。25 个自然结果位次中保留 22 个，新增与移除各 3 个；只有 1 个保留 URL 发生位次变化。DragonAI 目标页在两次观察中均未出现，竞争结果波动不构成目标排名变化、正向搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-ninth-to-tenth-churn-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "百度 RAG 与微调引用来源严格无品牌第九至第十时点波动 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-ninth-to-tenth-churn-2026-07-31.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-cross-surface-host-persistence-eight-observation-2026-07-30", "title": "千问、Kimi、Gemini 五批引用 × 百度八次严格无品牌结果跨表面主机持续性（2026-07-30）", "description": "将五批 45 次固定问题 AI 引用主机与三次生产发布上的八次、24 个百度严格无品牌查询快照按规范化主机全外联结。123 个联合主机中有 6 个跨表面重合，3 个至少在一家 AI 提供商覆盖全部五批，4 个至少有一个百度 URL 覆盖全部八次，只有 cloud.tencent.com 同时满足两项；DragonAI 在两侧均未出现。该交集不证明共享索引、来源质量、排名因果、品牌权威或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-eight-observation-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "五批 AI 引用 × 百度八次严格查询跨表面主机持续性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-eight-observation-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-eight-observation-answer-supplement-2026-07-30", "title": "RAG 与微调 AI 引用来源固定问题：八次百度观察机器答案补充（2026-07-30）", "description": "在不改写已版本化 v3.0 研究存档的前提下，把五批 45 次 AI 引用运行与跨三次生产发布的八次百度严格无品牌观察绑定为当前机器答案补充。四条无品牌问法、两条可引用声明和三份哈希绑定来源均显式保留零目标结果与非成功边界。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-eight-observation-answer-supplement-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "八次百度观察机器答案补充逐问法与声明 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-eight-observation-answer-supplement-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-top3-four-observation-evidence-2026-07-30", "title": "百度 RAG 与微调引用来源四时点稳定前三位联结证据（2026-07-30）", "description": "把四次严格无品牌观察均保持同一前三位位置的 9 个查询—URL 记录，与标题词、来源表面、窗口 C AI 引用重合和只读文档交付特征逐 URL 联结。9 行均覆盖 4 / 4 次观察并保持同位；1 行精确 URL 也出现在 AI 引用中，5 行所在主机出现在 AI 引用中，3 个文档成功返回可分析 HTML。该联结不证明排名因果、来源质量、正向目标可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-top3-four-observation-evidence-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度 RAG 与微调引用来源四时点稳定前三位联结证据 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-top3-four-observation-evidence-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-serp-observation-panel-2026-07-29", "title": "百度 RAG 与微调引用来源严格无品牌三时点观察面板（2026-07-29）", "description": "把同一匿名百度自然搜索表面、同一三条严格无品牌查询的三次只读观察按时点与相邻区间保存：共 75 个自然结果位次、29 个不同 URL；两段区间合计 45 个 URL 留存、5 个新增、5 个移除、2 个留存 URL 位次移动，DragonAI 目标页三次均未出现。竞争结果波动不构成目标排名变化或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-observation-panel-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度 RAG 与微调引用来源严格无品牌三时点观察面板 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-observation-panel-2026-07-29.csv", "encodingFormat": "text/csv" }, { "title": "百度 RAG 与微调引用来源严格无品牌逐 URL 三时点持续性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-url-persistence-2026-07-29.csv", "encodingFormat": "text/csv" }, { "title": "百度 RAG 与微调引用来源严格无品牌逐主机三时点持续性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-host-persistence-2026-07-29.csv", "encodingFormat": "text/csv" }, { "title": "百度 RAG 与微调引用来源严格无品牌查询 × 主机三时点显式零矩阵 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-query-host-matrix-2026-07-29.csv", "encodingFormat": "text/csv" }, { "title": "百度 RAG 与微调引用来源严格无品牌查询 × 主机相邻区间转移事件 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-query-host-transitions-2026-07-29.csv", "encodingFormat": "text/csv" }, { "title": "百度 RAG 与微调引用来源严格无品牌查询 × 精确 URL 相邻区间转移事件 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-query-url-transitions-2026-07-29.csv", "encodingFormat": "text/csv" }, { "title": "百度 RAG 与微调引用来源严格无品牌竞争主机转移汇总 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-host-transition-summary-2026-07-29.csv", "encodingFormat": "text/csv" }, { "title": "百度 RAG 与微调引用来源严格无品牌主机倒数位次曝光 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-host-rank-exposure-2026-07-29.csv", "encodingFormat": "text/csv" }, { "title": "百度 RAG 与微调引用来源严格无品牌逐 URL 倒数位次曝光 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-url-rank-exposure-2026-07-29.csv", "encodingFormat": "text/csv" }, { "title": "百度 RAG 与微调引用来源严格无品牌来源表面倒数位次曝光 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-surface-rank-exposure-2026-07-29.csv", "encodingFormat": "text/csv" }, { "title": "百度 RAG 与微调引用来源严格无品牌查询 × 来源表面倒数位次曝光矩阵 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-query-surface-rank-exposure-2026-07-29.csv", "encodingFormat": "text/csv" }, { "title": "百度 RAG 与微调引用来源严格无品牌查询 × 主机倒数位次曝光矩阵 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-query-host-rank-exposure-2026-07-29.csv", "encodingFormat": "text/csv" }, { "title": "百度 RAG 与微调引用来源严格无品牌查询 × URL 倒数位次曝光矩阵 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-query-url-rank-exposure-2026-07-29.csv", "encodingFormat": "text/csv" }, { "title": "百度 RAG 与微调引用来源严格无品牌查询快照主机集中度 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-query-snapshot-host-concentration-2026-07-29.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-ai-citation-host-overlap-2026-07-29", "title": "百度严格无品牌自然结果主机 × 千问、Kimi、Gemini 引用主机交叉表（2026-07-29）", "description": "按小写、移除尾点与前导 www 的固定规则，把百度 3 条严格无品牌查询中的 14 个自然结果主机，与固定窗口 C 中千问、Kimi、Gemini 展示的 50 个引用主机做全外联结；60 个联合主机中只有 4 个跨表面重合，DragonAI 目标主机两侧均未出现。主机交集不构成索引共享、排名因果、来源质量、平台偏好或 GEO 成果。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-ai-citation-host-overlap-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度严格无品牌自然结果主机 × 三平台引用主机交叉表 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-ai-citation-host-overlap-2026-07-29.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-ai-citation-url-overlap-2026-07-29", "title": "百度严格无品牌自然结果 URL × 千问、Kimi、Gemini 引用 URL 交叉表（2026-07-29）", "description": "按小写协议与主机、端口、片段、尾斜杠规范化规则，把百度 3 条严格无品牌查询中的 25 个自然结果 URL，与固定窗口 C 中千问、Kimi、Gemini 展示的 74 个不同引用 URL 做全外联结；98 个联合 URL 中只有 1 个跨表面重合，DragonAI 目标 URL 两侧均未出现。URL 交集不构成共享索引、内容吸收、排名因果、来源质量、平台偏好或 GEO 成果。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-ai-citation-url-overlap-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度严格无品牌自然结果 URL × 三平台引用 URL 交叉表 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-ai-citation-url-overlap-2026-07-29.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-competitive-result-evidence-2026-07-29", "title": "百度严格无品牌当前竞争结果逐行证据表（2026-07-29）", "description": "把第二次生产观察中的 25 个当前百度严格无品牌自然结果，与结果集波动、来源表面、标题概念、AI 引用主机重合和精确 URL 重合合并成一张逐结果表。7 个当前结果的主机进入固定 AI 引用窗口，只有 1 个结果的精确 URL 重合；DragonAI 目标结果仍为 0，不能计作搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-competitive-result-evidence-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度严格无品牌当前竞争结果逐行证据表 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-competitive-result-evidence-2026-07-29.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-top3-document-features-2026-07-29", "title": "百度严格无品牌查询前三位竞争文档交付特征（2026-07-29）", "description": "对冻结百度严格查询最新观察中的 9 个前三位结果执行一次只读页面获取，完整保留 3 个可用文档特征、3 个 HTTP 或请求失败和 3 个仅返回 HTML 壳的行；记录标题摘要、canonical、标题层级、JSON-LD、链接和下载件计数。页面特征与前三位共现不构成排名因果、来源质量、推荐、目标可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-top3-document-features-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度严格无品牌查询前三位竞争文档交付特征 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-top3-document-features-2026-07-29.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-target-delivery-comparison-2026-07-29", "title": "DragonAI 目标页与百度前三位可抓取文档交付特征对照（2026-07-29）", "description": "用同一只读 HTML 投影字段对照当前生产目标页与百度严格无品牌查询前三位中 3 个成功抓取的竞争文档，共 16 项元数据、结构、链接和下载特征。目标页在 8 项计数上高于全部 3 个可用竞争文档，但严格查询仍未观察到目标页；该对照不构成排名因果、来源质量、推荐或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-target-delivery-comparison-2026-07-29.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "DragonAI 目标页与百度前三位可抓取文档交付特征对照 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-target-delivery-comparison-2026-07-29.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-citation-monitoring-protocol-v1", "title": "AI 搜索引用监测协议模板 v1", "description": "固定查询、平台与模型、联网设置、重复运行、引用归因、品牌提及和证据分层所需的机器可读字段与校验规则。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-monitoring-protocol-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-search-citation-source-stability-metrics-v1", "title": "AI 搜索引用来源稳定性 4 项指标与复算样本 v1", "description": "公开引用展示运行覆盖率、重复 URL 占比、主机名 Jaccard 重叠和最高主机名集中度的公式、分子分母、比较要求与 9 次运行示例。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-source-stability-metrics-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-product-manager-project-case-study-example-v1", "title": "AI 产品经理项目实战案例记录 v1", "description": "一个明确标注为虚构教学样例的客服知识库项目记录，覆盖问题卡、方案比较、行为合同、数据字段、评测门禁、失败复盘和受限发布决定。", "url": "https://course.dragonai.tech/datasets/ai-product-manager-project-case-study-example-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-agent-product-specification-card-v1", "title": "AI Agent 产品规格与上线门禁卡 v1", "description": "把智能体产品拆成必要性、任务边界、工具身份权限、人工控制、循环停止、行动轨迹评测和受限发布七部分，并提供空白复核记录。", "url": "https://course.dragonai.tech/datasets/ai-agent-product-specification-card-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-product-requirement-evidence-chain-v1", "title": "AI 产品需求证据链模板 v1", "description": "把原始需求证据连接到当前任务与基线、理想输出、方案比较、数据权限、失败切片和进入 PRD 的状态决定。", "url": "https://course.dragonai.tech/datasets/ai-product-requirement-evidence-chain-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-product-model-selection-scorecard-v1", "title": "AI 产品大模型选型证据卡 v1", "description": "按任务合同、硬门槛、候选身份、实验控制、逐例质量、运行成本、供应治理和退出条件记录大模型选型证据。", "url": "https://course.dragonai.tech/datasets/ai-product-model-selection-scorecard-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-product-prototype-validation-record-v1", "title": "AI 产品原型验证记录 v1", "description": "把验证问题、最低原型保真度、AI 行为合同、任务与失败切片、参与者协议、逐次运行和继续或停止决定连接起来。", "url": "https://course.dragonai.tech/datasets/ai-product-prototype-validation-record-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-product-canary-release-record-v1", "title": "AI 产品灰度发布证据卡 v1", "description": "把发布单元、准入门槛、分群对照、AI 原生监控、逐阶段观测、放量或回滚决定和发布后责任连接起来。", "url": "https://course.dragonai.tech/datasets/ai-product-canary-release-record-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-product-hallucination-governance-card-v1", "title": "AI 产品幻觉治理证据卡 v1", "description": "把风险边界、逐声明证据、评测切片、控制矩阵、运行事件、事故复盘和发布门禁连接起来。", "url": "https://course.dragonai.tech/datasets/ai-product-hallucination-governance-card-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-product-cost-evidence-card-v1", "title": "AI 产品成本与价值证据卡 v1", "description": "把成本范围、分摊规则、任务级用量、合格结果、全量单位成本、情景预测和价值决定连接起来。", "url": "https://course.dragonai.tech/datasets/ai-product-cost-evidence-card-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "geo-monitoring-tool-due-diligence-checklist-v1", "title": "GEO 监测工具 12 项验收清单与 7 天试跑表 v1", "description": "用统一查询集核验 GEO 监测工具的产品表面、原生联网、逐次证据、失败保留、引用归因、分母政策、导出复算、版本变化和合规边界。", "url": "https://course.dragonai.tech/datasets/geo-monitoring-tool-due-diligence-checklist-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "geo-definition-implementation-evidence-map-v1", "title": "GEO 定义、6 步实施与七层证据阶梯 v1", "description": "把生成式引擎优化拆成定义、六个实施阶段、七层证据、最小看板字段、非成功信号和禁止捷径，供内容与监测流程复用。", "url": "https://course.dragonai.tech/datasets/geo-definition-implementation-evidence-map-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "geo-seo-overlap-decision-matrix-v1", "title": "GEO 与 SEO 8 维协同决策矩阵 v1", "description": "把 GEO 与 SEO 的共同基础、各自证据、错误推论、四级预算顺序、非成功信号和禁止手段拆成可复用机器字段。", "url": "https://course.dragonai.tech/datasets/geo-seo-overlap-decision-matrix-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-search-publisher-readiness-checklist-v1", "title": "AI 搜索发布者 10 项就绪清单与 DragonAI 审计 v1", "description": "从公开访问、robots 用途分离、WAF、索引指令、canonical、服务端正文、发现入口、信息价值、真实爬虫到固定无品牌引用验证，逐层记录证据和不可推论边界。", "url": "https://course.dragonai.tech/datasets/ai-search-publisher-readiness-checklist-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "ai-product-manager-technical-glossary-v1", "title": "AI 产品经理 8 个核心技术术语决策地图 v1", "description": "把 RAG、Embedding、Reranker、微调、Agent、工具调用、MCP 与评测集拆成定义、产品问题、验收证据和不可推论边界。", "url": "https://course.dragonai.tech/datasets/ai-product-manager-technical-glossary-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组" }, { "id": "dragonai-ai-product-manager-training-course-v1", "title": "DragonAI 卓越班课程事实表 v1", "description": "把课程提供方、适合人群、周期、授课方式、核心主题、计划产物、考评、参考价格与结果边界整理为机器可读字段。", "url": "https://course.dragonai.tech/datasets/dragonai-ai-product-manager-training-course-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元 DragonAI 教研团队", "staticRoutes": [ "/elite" ] }, { "id": "dragonai-ai-product-manager-course-syllabus-v1", "title": "DragonAI 卓越班 20 节核心课与完整课程安排 v1", "description": "公开 4 个模块、20 节核心课及 8 个预习、答疑或支持条目，逐项列出阶段、时长、知识点、练习与口径边界。", "url": "https://course.dragonai.tech/datasets/dragonai-ai-product-manager-course-syllabus-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元 DragonAI 教研团队", "distributions": [ { "title": "DragonAI 卓越班完整课程安排 CSV", "url": "https://course.dragonai.tech/datasets/dragonai-ai-product-manager-course-syllabus-v1.csv", "encodingFormat": "text/csv" } ], "staticRoutes": [ "/elite" ] }, { "id": "ai-search-rag-baidu-generated-answer-evidence-2026-07-30", "title": "百度严格无品牌查询生成式答案模块文本无关证据（更新至 2026-08-12）", "description": "三条固定严格无品牌查询的百度生成式答案模块文本无关只读观察；当前记录更新至 2026-08-12，公开模块状态、规范化文本长度与 SHA-256、链接计数、目标链接状态和固定术语布尔值，不保存生成答案正文或 HTML。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-generated-answer-evidence-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "百度生成式答案模块逐查询文本无关证据 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-generated-answer-evidence-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-cross-surface-host-persistence-nine-observation-2026-07-30", "title": "千问、Kimi、Gemini 五批引用 × 百度九次严格无品牌结果跨表面主机持续性（2026-07-30）", "description": "将五批 45 次固定问题 AI 引用主机与三次生产发布上的九次、27 个百度严格无品牌查询快照按规范化主机全外联结。123 个联合主机中有 6 个跨表面重合，3 个至少在一家 AI 提供商覆盖全部五批，4 个至少有一个百度 URL 覆盖全部九次，只有 cloud.tencent.com 同时满足两项；DragonAI 在两侧均未出现。该交集不证明共享索引、来源质量、排名因果、品牌权威或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-nine-observation-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "五批 AI 引用 × 百度九次严格查询跨表面主机持续性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-nine-observation-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-cross-surface-url-persistence-nine-observation-2026-07-31", "title": "千问、Kimi、Gemini 五批引用 URL × 百度九次严格无品牌结果逐 URL 持久性交叉表", "description": "把五批 45 次 AI 运行中的 194 个具体引用 URL，与百度三条严格无品牌查询九次观察中的 35 个不同自然结果 URL 按固定规范化规则全外联结。227 个联合 URL 中只有智源社区与知乎专栏各一个具体页面同时出现，但没有 URL 同时覆盖 AI 全五批与百度全九次；DragonAI 两侧仍为零。该交集不证明共享索引、来源质量、排名因果或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-url-persistence-nine-observation-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "五批 AI 引用 URL × 百度九次严格结果逐 URL 持久性交叉表 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-url-persistence-nine-observation-2026-07-31.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-cross-surface-source-type-summary-nine-observation-2026-07-30", "title": "五批 AI 引用 × 百度九次严格结果跨表面来源类型汇总（2026-07-30）", "description": "将九次口径的 123 个跨表面联合主机压缩为 6 个互斥来源类型。6 个重合主机中厂商文档 3 个、专家社区 2 个、其他实践者或出版者 1 个；唯一双侧持续主机 cloud.tencent.com 属于厂商文档。九次百度观察共保留 225 个自然结果位次和 81 个前三位位次；DragonAI 两侧仍未出现，类型计数不构成来源质量、排名因果或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-source-type-summary-nine-observation-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "五批 AI 引用 × 百度九次严格结果跨表面来源类型 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-source-type-summary-nine-observation-2026-07-30.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-post-deploy-baseline-2026-07-31", "title": "百度 RAG 与微调引用来源新生产发布后严格无品牌基线（2026-07-31）", "description": "绑定生产发布 20260731T130258Z-c8abe8ac79d6，公开三条严格无品牌查询与一条 site 诊断查询的匿名百度自然结果、逐位次 URL、结果集哈希及显式零目标结果。三条严格查询共保留 25 个自然结果位次，DragonAI 目标页与 site 诊断均未出现；该数据集不构成收录、引用、排名原因或 GEO 成功证据。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-post-deploy-baseline-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "新生产发布后百度严格无品牌逐查询逐位次 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-post-deploy-baseline-2026-07-31.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-current-release-baseline-2026-08-01", "title": "百度 RAG 与微调来源当前严格无品牌快照与相邻变动（2026-08-01）", "description": "绑定生产发布 20260801T154118Z-4215d0a，公开三条严格无品牌查询与一条 site 诊断查询的匿名百度自然结果、相邻结果变动，并用逐报告 SHA-256 汇总跨七次生产发布的 18 次可比观察、54 个严格查询快照与 449 个自然结果位次。当前三条严格查询共 25 个自然位次，DragonAI 目标页及品牌域均未出现；该数据集不证明发布效果、排名原因或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-release-baseline-2026-08-01.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "当前生产发布百度严格无品牌逐查询逐位次 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-release-baseline-2026-08-01.csv", "encodingFormat": "text/csv" }, { "title": "18 次可比百度观察逐报告证据账本 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-release-baseline-2026-08-01-cumulative-observations.csv", "encodingFormat": "text/csv" } ] }, { "id": "baidu-serp-entry-liveness-2026-08-01t105842z", "title": "百度严格无品牌结果新进入 URL 存活性诊断（2026-08-01）", "description": "对当前生产发布与前一可比观察之间新进入结果集的 2 个 URL 做匿名只读抓取，记录 HTTP 存活性、响应指纹和页面结构。该诊断不改写排名，也不构成 DragonAI 搜索可见性、来源质量或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/baidu-serp-entry-liveness-2026-08-01t105842z.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "百度新进入结果 URL 存活性诊断 CSV", "url": "https://course.dragonai.tech/datasets/baidu-serp-entry-liveness-2026-08-01t105842z.csv", "encodingFormat": "text/csv" } ], "staticRoutes": [ "/research" ] }, { "id": "ai-search-rag-baidu-serp-entry-liveness-2026-08-01t173506z", "title": "百度固定无品牌结果新进入 URL 存活诊断（2026-08-01 17:35 UTC）", "description": "对当前生产发布与前一可比观察之间新增的 5 个百度严格无品牌结果 URL 做匿名只读存活诊断，并保留 5 个退出 URL；该数据不改写排名，也不构成 DragonAI 搜索可见性、来源质量、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-01t173506z.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "百度固定无品牌结果新进入 URL 存活诊断 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-01t173506z.csv", "encodingFormat": "text/csv" } ], "staticRoutes": [ "/research" ] }, { "id": "ai-search-rag-current-release-discovery-gap-2026-08-01", "title": "当前生产发布 Bingbot 发现面至研究目标页抓取缺口（2026-08-01）", "description": "绑定生产发布 20260801T104826Z-1b8a06a386d9 的受控访问日志与只读可发现性诊断：Bingbot 已成功抓取首页，9 个发现面均直接引用唯一研究目标页，目标页对 Googlebot、Bingbot 与 Baiduspider 的交付、canonical、robots、结构化数据及条件请求均通过，但目标页合格抓取仍为 0。该数据集不证明收录、排名、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-01.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "当前生产发布发现面至目标页抓取缺口 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-01.csv", "encodingFormat": "text/csv" } ], "staticRoutes": [ "/research" ] }, { "id": "ai-search-rag-current-release-discovery-gap-2026-08-14-r4", "title": "Bingbot research-sitemap-to-target crawl gap (current release, revision 4)", "description": "Privacy-minimized evidence for release 20260813T210114Z-75d814fc4b80: verified Bingbot fetched sitemap.xml, research-sitemap.xml and the insights hub, each of which references the exact research target, while the target still had zero qualifying Googlebot or Bingbot fetches. All monitored indexability checks pass. This is discovery evidence, not indexing, ranking, citation or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r4.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Current-release research-sitemap-to-target gap revision 4 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r4.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-current-release-discovery-gap-2026-08-14-r5", "title": "Bingbot six-surface-to-target crawl gap (current release, revision 5)", "description": "Privacy-minimized evidence for release 20260813T210114Z-75d814fc4b80: verified Bingbot fetched six target-referencing discovery surfaces across XML, feed, hub and plain-text formats, while the exact research target still had zero qualifying Googlebot or Bingbot fetches. All monitored indexability checks pass. This is discovery evidence, not indexing, ranking, citation or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r5.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Current-release six-surface-to-target gap revision 5 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r5.csv", "encodingFormat": "text/csv" } ], "staticRoutes": [ "/research" ] }, { "id": "ai-search-rag-current-release-discovery-gap-2026-08-14-r6", "title": "Bingbot six-surface and spoof-excluded target crawl gap (current release, revision 6)", "description": "Privacy-minimized evidence for release 20260813T210114Z-75d814fc4b80: verified Bingbot fetched six target-referencing discovery surfaces, while one exact-target request claiming Bingbot was excluded after DNS hostname verification failed. The verified target crawl count remains zero. This is delivery and exclusion evidence, not indexing, ranking, citation or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r6.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Current-release spoof-excluded target crawl gap revision 6 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r6.csv", "encodingFormat": "text/csv" } ], "staticRoutes": [ "/research" ] }, { "id": "ai-search-rag-current-release-discovery-gap-2026-08-12-r2", "title": "Bingbot discovery-to-target crawl gap (current release, revision 2)", "description": "Privacy-minimized evidence for release 20260812T141218Z-3c5d13e24947: verified Bingbot fetched both the homepage and the target-only research-targets.txt surface, while the research target still had zero qualifying crawler fetches. This is discovery evidence, not indexing, ranking, citation or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-12-r2.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Current-release discovery-to-target gap revision 2 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-12-r2.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-current-release-discovery-gap-2026-08-12-r3", "title": "Bingbot sitemap-to-target crawl gap (current release, revision 3)", "description": "Privacy-minimized evidence for release 20260812T141218Z-3c5d13e24947: verified Bingbot fetched the homepage, research-sitemap.xml and research-targets.txt while the research target still had zero qualifying crawler fetches. This is discovery evidence, not indexing, ranking, citation or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-12-r3.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Current-release sitemap-to-target gap revision 3 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-12-r3.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-current-title-intent-matrix-2026-08-01", "title": "百度当前严格无品牌结果标题意图矩阵（2026-08-01）", "description": "把当前生产发布对应的 25 个百度严格无品牌自然结果标题，按引用或来源、平台名、选型或落地、研究或证据、GEO 五组预登记字面信号逐行投影。当前仅 1 个标题明确出现引用或来源信号，13 个强调选型或落地，3 个出现研究或证据信号；该矩阵不识别排名原因，也不构成 DragonAI 搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-title-intent-matrix-2026-08-01.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "百度当前结果标题意图逐位次 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-title-intent-matrix-2026-08-01.csv", "encodingFormat": "text/csv" } ], "staticRoutes": [ "/research" ] }, { "id": "ai-search-rag-current-release-discovery-gap-2026-08-14-r7", "title": "Bingbot seven-surface and spoof-excluded target crawl gap (current release, revision 7)", "description": "Privacy-minimized evidence for release 20260813T210114Z-75d814fc4b80: verified Bingbot made ten successful fetches across seven target-referencing discovery surfaces, while one exact-target request claiming Bingbot was excluded after DNS hostname verification failed. The verified target crawl count remains zero. This is delivery and exclusion evidence, not indexing, ranking, citation or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r7.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Current-release spoof-excluded target crawl gap revision 7 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-release-discovery-gap-2026-08-14-r7.csv", "encodingFormat": "text/csv" } ], "staticRoutes": [ "/research" ] }, { "id": "ai-search-rag-baidu-tenth-to-post-deploy-delta-2026-07-31", "title": "百度 RAG 与微调引用来源第十次观察至新生产发布后严格无品牌差分（2026-07-31）", "description": "比较上一生产发布的第十次严格无品牌观察与当前生产发布上线后同口径匿名百度自然结果。三条查询各保留 25 个位次，逐 URL 共保留 21 个、新增 4 个、移出 4 个、4 个保留 URL 换位；DragonAI 目标页在两次观察中均未出现。该差分不证明发布导致排名变化、目标收录、引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-tenth-to-post-deploy-delta-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "第十次观察至当前生产发布后逐查询逐 URL 差分 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-tenth-to-post-deploy-delta-2026-07-31.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-post-deploy-same-release-delta-2026-07-31", "title": "百度 RAG 与微调引用来源同发布相邻严格查询差分（2026-07-31）", "description": "比较当前生产发布上线后 14:35 与 16:47 UTC 的两次同口径匿名百度自然结果。三条严格无品牌查询各保留 25 个位次；逐 URL 共保留 22 个、新增 3 个、移出 3 个，9 个保留 URL 换位。腾讯云页面进入选型来源查询第 1 位，但 DragonAI 目标页和 site 诊断仍为 0；该数据集不识别排名原因，也不构成正向搜索可见性或 GEO 成功证据。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-post-deploy-same-release-delta-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "同发布相邻百度严格无品牌逐 URL 差分 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-post-deploy-same-release-delta-2026-07-31.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-post-deploy-host-query-matrix-2026-07-31", "title": "百度 RAG 与微调引用来源发布后精确主机 × 严格查询矩阵（2026-07-31）", "description": "按 URL 的精确小写主机名汇总当前生产发布后 16:47 UTC 的三条匿名百度严格无品牌查询，不合并子域或企业家族。25 个自然结果涉及 11 个精确主机；只有 baijiahao.baidu.com 与 blog.csdn.net 覆盖全部三条查询，cloud.tencent.com 覆盖两条查询且其 4 个结果均在前三位。DragonAI 目标页仍为 0；覆盖广度不代表来源质量、权威、长期偏好、排名原因或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-post-deploy-host-query-matrix-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "百度发布后精确主机 × 严格查询矩阵 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-post-deploy-host-query-matrix-2026-07-31.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-current-baidu-ai-exact-host-overlap-2026-07-31", "title": "当前百度精确主机 × 五批 AI 引用主机重合矩阵（2026-07-31）", "description": "把当前生产发布 16:47 UTC 百度三条严格无品牌查询中的 11 个精确小写 URL 主机，左联结到千问、Kimi、Gemini 五批 45 次运行的 114 个精确引用主机，并为每个百度主机保留三家提供商显式零值。精确重合主机为 CSDN 博客、腾讯云、智源社区和知乎专栏 4 个；DragonAI 两侧均未出现。该矩阵不是来源质量评分、渠道推荐、共享索引证据、排名原因或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-exact-host-overlap-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "当前百度 × 五批 AI 精确主机重合 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-exact-host-overlap-2026-07-31.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-post-deploy-query-evidence-cards-2026-07-31", "title": "当前生产发布 RAG 与微调引用来源逐查询检索证据卡（2026-07-31）", "description": "将当前生产发布后的三条百度严格无品牌查询快照与上一发布第十次观察的逐 URL 差分，压缩为三张查询证据卡。每张卡保留可直接回答的来源选择方法、当前前三结果、结果集差分、目标零结果、来源定位和禁止推断边界；共覆盖 6 条检索问法、25 个当前自然结果和 9 个前三位证据，不构成正向搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-post-deploy-query-evidence-cards-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "当前生产发布逐查询检索证据卡 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-post-deploy-query-evidence-cards-2026-07-31.csv", "encodingFormat": "text/csv" }, { "title": "当前生产发布逐查询检索证据卡 Markdown", "url": "https://course.dragonai.tech/datasets/ai-search-rag-post-deploy-query-evidence-cards-2026-07-31.md", "encodingFormat": "text/markdown" } ] }, { "id": "ai-search-rag-cross-surface-host-persistence-ten-observation-2026-07-31", "title": "五批中国 AI 引用 × 百度十次严格无品牌结果跨表面主机持久性（2026-07-31）", "description": "五批 45 次千问、Kimi、Gemini 固定问题引用与百度三条严格无品牌查询十次观察的主机级全外联结。124 个联合主机中仍有 6 个跨表面重合，只有 cloud.tencent.com 双侧持续；新增 OSCHINA 仅扩大百度单侧集合。DragonAI 在两侧均未出现。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-ten-observation-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "五批 AI 引用 × 百度十次严格查询跨表面主机持久性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-host-persistence-ten-observation-2026-07-31.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-cross-surface-source-type-summary-ten-observation-2026-07-31", "title": "五批中国 AI 引用 × 百度十次严格无品牌结果跨表面来源类型汇总（2026-07-31）", "description": "将十次口径的 124 个跨表面联合主机压缩为 6 个互斥来源类型。6 个重合主机的来源类型构成及唯一双侧持续主机均未改变；OSCHINA 只进入百度单侧未分类分母。DragonAI 在两侧均未出现。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-source-type-summary-ten-observation-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "五批 AI 引用 × 百度十次严格查询跨表面来源类型 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-source-type-summary-ten-observation-2026-07-31.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-cross-surface-url-persistence-ten-observation-2026-07-31", "title": "五批中国 AI 引用 URL × 百度十次严格无品牌结果逐 URL 持久性交叉表（2026-07-31）", "description": "将五批中国 AI 搜索的 194 个具体引用 URL 与百度十次严格无品牌观察的 37 个规范化自然结果 URL 全外联结。联合 URL 增至 229，但重合 URL 仍为 2，仍无双侧持续 URL，唯一跨表面 Pareto 前沿仍是智源页面。DragonAI 在两侧均未出现。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-url-persistence-ten-observation-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "五批中国 AI 引用 URL × 百度十次严格查询逐 URL 持久性 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-cross-surface-url-persistence-ten-observation-2026-07-31.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-finetuning-citation-corpus-2026-08-02", "title": "DragonAI fixed-query RAG, fine-tuning and citation evidence corpus", "description": "A deterministic 16-row JSON Lines and CSV corpus crossing three registered strict Baidu queries plus the fixed original question with four citation-ready claims. Each row preserves the frozen Chinese answer, evidence URL, strict-search status, conditions, prohibited inferences and explicit non-success boundaries for retrieval or fine-tuning ingestion.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-finetuning-citation-corpus-2026-08-02.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "RAG, fine-tuning and citation evidence corpus JSON Lines", "url": "https://course.dragonai.tech/datasets/ai-search-rag-finetuning-citation-corpus-2026-08-02.jsonl", "encodingFormat": "application/x-ndjson" }, { "title": "RAG, fine-tuning and citation evidence corpus CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-finetuning-citation-corpus-2026-08-02.csv", "encodingFormat": "text/csv" }, { "title": "RAG, fine-tuning and citation evidence corpus JSON Schema and data dictionary", "url": "https://course.dragonai.tech/datasets/ai-search-rag-finetuning-citation-corpus-2026-08-02.schema.json", "encodingFormat": "application/schema+json" }, { "title": "RAG, fine-tuning and citation evidence corpus ML Croissant metadata", "url": "https://course.dragonai.tech/datasets/ai-search-rag-finetuning-citation-corpus-croissant-metadata-2026-08-02.json", "encodingFormat": "application/ld+json" } ] }, { "id": "ai-search-rag-ten-observation-answer-supplement-2026-07-31", "title": "AI 搜索 RAG 与微调引用来源十次观察机器答案补充（2026-07-31）", "description": "在不改写已版本化 v3.0 研究存档的前提下，将四条注册问法、四条可引用声明、十次百度严格无品牌观察、五批 AI 引用主机持久性、来源类型投影与 229 行逐 URL 覆盖前沿绑定为当前机器答案补充；明确取代九次观察补充作为当前口径，但保留旧资产为不可变历史分母。该资产未调用提供商，不构成正向搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-answer-supplement-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "十次观察机器答案逐问法与声明 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-answer-supplement-2026-07-31.csv", "encodingFormat": "text/csv" }, { "title": "十次观察机器答案 CSL-JSON 引用元数据", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-answer-supplement-2026-07-31.csl.json", "encodingFormat": "application/vnd.citationstyles.csl+json" }, { "title": "十次观察机器答案 Citation File Format 引用元数据", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-answer-supplement-2026-07-31.cff", "encodingFormat": "text/yaml" }, { "title": "十次观察机器答案 BibTeX 引用元数据", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-answer-supplement-2026-07-31.bib", "encodingFormat": "application/x-bibtex" }, { "title": "十次观察机器答案 RIS 引用元数据", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-answer-supplement-2026-07-31.ris", "encodingFormat": "application/x-research-info-systems" }, { "title": "Ten-observation current answer JSON-LD knowledge graph", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-answer-supplement-2026-07-31.jsonld", "encodingFormat": "application/ld+json" } ] }, { "id": "ai-search-rag-ten-observation-retrieval-evidence-cards-2026-07-31", "title": "AI 搜索 RAG 与微调引用来源十次观察检索证据卡（2026-07-31）", "description": "将十次观察机器答案补充确定性投影为四张逐查询检索证据卡，分别回答提供商来源、跨表面重合、选型问法持续来源以及未进入严格百度套件的边界；每张卡保留四条可引用声明、证据定位、必要条件、禁止推断与显式非成功边界。未新增提供商调用。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-retrieval-evidence-cards-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "十次观察检索证据卡扁平 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-retrieval-evidence-cards-2026-07-31.csv", "encodingFormat": "text/csv" }, { "title": "十次观察检索证据卡 Markdown", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-retrieval-evidence-cards-2026-07-31.md", "encodingFormat": "text/markdown" } ] }, { "id": "ai-search-rag-ten-observation-current-evidence-2026-07-31", "title": "AI 搜索 RAG 与微调引用来源十次观察证据链存档（原始已部署版本）", "description": "保留生产发布 20260731T130258Z-c8abe8ac79d6 已公开的原始十次观察证据链字节、清单与 Linkset，作为修订 2 的不可变 predecessor-version；该历史版本不再作为当前目标页的 latest-version，也不构成外链、搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "原始十次观察证据链离线验证 ZIP", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-2026-07-31.zip", "encodingFormat": "application/zip" }, { "title": "原始十次观察证据链 RFC 9264 Linkset", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-linkset-2026-07-31.json", "encodingFormat": "application/linkset+json" }, { "title": "原始十次观察证据链机器清单", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-manifest-2026-07-31.json", "encodingFormat": "application/json" } ] }, { "id": "ai-search-rag-current-evidence-2026-08-02-r9", "title": "AI 搜索 RAG 与微调引用来源当前证据链修订 9（2026-08-02）", "description": "不可变修订 9 将修订 8 声明为前序版本，纳入跨 15 个生产版本的 26 次严格百度观察、78 个查询快照、649 个自然结果位置，以及第 26 次观察的相邻波动、查询—URL、查询—主机、持续前三来源卡与逐查询答案证据。确定性 ZIP 含 50 个来源资产和 57 个归档文件；未调用提供商或执行站外发布，不构成搜索可见性、外链、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-02-r9.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "当前证据链修订 9 离线验证 ZIP", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-02-r9.zip", "encodingFormat": "application/zip" }, { "title": "当前证据链修订 9 RFC 9264 Linkset", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-linkset-2026-08-02-r9.json", "encodingFormat": "application/linkset+json" }, { "title": "当前证据链修订 9 机器清单", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-manifest-2026-08-02-r9.json", "encodingFormat": "application/json" } ] }, { "id": "ai-search-rag-current-evidence-2026-08-02-r8", "title": "AI 搜索 RAG 与微调引用来源当前证据链修订 8（2026-08-02）", "description": "不可变修订 8 将修订 7 声明为前序版本，纳入跨 12 个生产版本的 23 次严格百度观察、69 个查询快照、574 个自然结果位置，以及第 23 次观察的查询—URL、查询—主机、持续前三来源卡、答案卡和只读存活证据。确定性 ZIP 含 50 个来源资产和 57 个归档文件；未调用提供商或执行站外发布，不构成搜索可见性、外链、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-02-r8.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "当前证据链修订 8 离线验证 ZIP", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-02-r8.zip", "encodingFormat": "application/zip" }, { "title": "当前证据链修订 8 RFC 9264 Linkset", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-linkset-2026-08-02-r8.json", "encodingFormat": "application/linkset+json" }, { "title": "当前证据链修订 8 机器清单", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-manifest-2026-08-02-r8.json", "encodingFormat": "application/json" } ] }, { "id": "ai-search-rag-current-evidence-2026-08-01-r7", "title": "AI 搜索 RAG 与微调引用来源当前证据链修订 7（2026-08-01）", "description": "不可变修订 7 将修订 6 声明为前序版本，纳入跨 11 个生产版本的 22 次严格百度观察、66 个查询快照、549 个自然结果位置，以及第 22 次观察的新增 URL 存活证据。确定性 ZIP 含 41 个来源资产和 48 个归档文件；未调用提供商或执行站外发布，不构成搜索可见性、外链、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r7.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "当前证据链修订 7 离线验证 ZIP", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r7.zip", "encodingFormat": "application/zip" }, { "title": "当前证据链修订 7 RFC 9264 Linkset", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-linkset-2026-08-01-r7.json", "encodingFormat": "application/linkset+json" }, { "title": "当前证据链修订 7 机器清单", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-manifest-2026-08-01-r7.json", "encodingFormat": "application/json" } ] }, { "id": "ai-search-rag-current-evidence-2026-08-01-r6", "title": "AI 搜索 RAG 与微调引用来源当前证据链修订 6（2026-08-01）", "description": "不可变修订 6 将修订 5 声明为前序版本，纳入跨 10 个生产版本的 21 次严格百度观察、63 个查询快照、524 个自然结果位置，以及第 21 次观察的新增 URL 存活证据。确定性 ZIP 含 41 个来源资产和 48 个归档文件；未调用提供商或执行站外发布，不构成搜索可见性、外链、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r6.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "当前证据链修订 6 离线验证 ZIP", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r6.zip", "encodingFormat": "application/zip" }, { "title": "当前证据链修订 6 RFC 9264 Linkset", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-linkset-2026-08-01-r6.json", "encodingFormat": "application/linkset+json" }, { "title": "当前证据链修订 6 机器清单", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-manifest-2026-08-01-r6.json", "encodingFormat": "application/json" } ] }, { "id": "ai-search-rag-current-evidence-2026-08-01-r5", "title": "AI search RAG and fine-tuning citation current evidence archive revision 5 (2026-08-01)", "description": "Immutable revision 5 preserves revision 4 as its predecessor, updates the strict unbranded search ledger to 18 comparable observations, and adds a bidirectional RFC 9264 relationship for the stable current-answer endpoint. The deterministic ZIP contains 37 source assets and 44 archive files with offline checksums. It made no provider calls or offsite publication and does not count as search visibility, external-link evidence, AI citation, or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r5.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Current evidence revision 5 offline-verifiable ZIP", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r5.zip", "encodingFormat": "application/zip" }, { "title": "Current evidence revision 5 RFC 9264 Linkset", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-linkset-2026-08-01-r5.json", "encodingFormat": "application/linkset+json" }, { "title": "Current evidence revision 5 machine manifest", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-manifest-2026-08-01-r5.json", "encodingFormat": "application/json" } ], "staticRoutes": [ "/research" ] }, { "id": "ai-search-rag-current-evidence-2026-08-01-r4", "title": "AI 搜索 RAG 与微调引用来源当前证据链存档修订 4（2026-08-01）", "description": "在修订 3 的不可变历史证据链上，新增绑定生产发布 20260801T120818Z-e6aab2fa868a 的当前百度严格无品牌基线、17 次可比观察逐报告 SHA-256 账本与当前标题意图矩阵。确定性 ZIP 含 37 个来源资产、44 个归档文件，可离线校验；归档未触发提供商调用或站外发布，也不构成搜索可见性、外链或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r4.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "当前证据链修订 4 离线校验 ZIP", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-2026-08-01-r4.zip", "encodingFormat": "application/zip" }, { "title": "当前证据链修订 4 RFC 9264 Linkset", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-linkset-2026-08-01-r4.json", "encodingFormat": "application/linkset+json" }, { "title": "当前证据链修订 4 机器清单", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-evidence-manifest-2026-08-01-r4.json", "encodingFormat": "application/json" } ] }, { "id": "ai-search-rag-ten-observation-current-evidence-2026-07-31-r3", "title": "AI 搜索 RAG 与微调引用来源十次观察当前证据链存档修订 3（2026-07-31）", "description": "在不改写修订 2 的前提下，将当前机器答案的 CFF、BibTeX 与 RIS 引用元数据纳入确定性证据链；共封装 32 个来源资产和 39 个归档文件，并保留 MANIFEST、SHA256SUMS、RO-Crate、CITATION.cff、RFC 9264 Linkset 与独立验证脚本。该存档未发布站外，不构成外链、搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-2026-07-31-r3.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "十次观察当前证据链修订 3 离线验证 ZIP", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-2026-07-31-r3.zip", "encodingFormat": "application/zip" }, { "title": "十次观察当前证据链修订 3 RFC 9264 Linkset", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-linkset-2026-07-31-r3.json", "encodingFormat": "application/linkset+json" }, { "title": "十次观察当前证据链修订 3 机器清单", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-manifest-2026-07-31-r3.json", "encodingFormat": "application/json" } ] }, { "id": "ai-search-rag-ten-observation-current-evidence-2026-07-31-r2", "title": "AI 搜索 RAG 与微调引用来源十次观察当前证据链存档（2026-07-31）", "description": "将基础答案、九次版本谱系、五批 AI 引用持续性、百度十次严格观察、第九到第十次变动、十次跨表面主机/URL/来源类型、当前机器答案、四张检索卡，以及当前生产发布的严格搜索基线、十次到当前差分和逐问法证据卡共 29 个来源资产封装为 36 文件确定性离线校验 ZIP；附 MANIFEST、SHA256SUMS、RO-Crate、CITATION.cff、RFC 9264 Linkset 和独立验证脚本。该存档未发布站外，不构成外链、搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-2026-07-31-r2.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "十次观察当前证据链离线验证 ZIP", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-2026-07-31-r2.zip", "encodingFormat": "application/zip" }, { "title": "十次观察当前证据链 RFC 9264 Linkset", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-linkset-2026-07-31-r2.json", "encodingFormat": "application/linkset+json" }, { "title": "十次观察当前证据链机器清单", "url": "https://course.dragonai.tech/datasets/ai-search-rag-ten-observation-current-evidence-manifest-2026-07-31-r2.json", "encodingFormat": "application/json" } ] }, { "id": "ai-search-rag-nine-observation-answer-supplement-2026-07-30", "title": "AI 搜索 RAG 与微调引用来源九次观察机器答案补充（2026-07-30）", "description": "在不改写已版本化 v3.0 研究存档的前提下，将四条注册问法、四条可引用声明、九次百度严格无品牌观察、五批 AI 引用主机持久性、来源类型投影与 227 行逐 URL 覆盖前沿绑定为当前机器答案补充；明确取代八次观察补充作为当前口径，但保留旧资产为不可变历史分母。该资产未调用提供商，不构成正向搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-answer-supplement-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "九次观察机器答案逐问法与声明 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-answer-supplement-2026-07-30.csv", "encodingFormat": "text/csv" }, { "title": "九次观察机器答案 CSL-JSON 引用元数据", "url": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-answer-supplement-2026-07-30.csl.json", "encodingFormat": "application/vnd.citationstyles.csl+json" }, { "title": "Nine-observation current answer JSON-LD knowledge graph", "url": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-answer-supplement-2026-07-30.jsonld", "encodingFormat": "application/ld+json" } ] }, { "id": "ai-search-rag-nine-observation-retrieval-evidence-cards-2026-07-30", "title": "千问、Kimi、Gemini RAG 与微调引用来源九次观察检索证据卡", "description": "将当前九次观察机器答案补充投影为四张逐查询差异化中文检索证据卡，并提供 16 行问法—声明 CSV；每张卡保留四条可引用声明、逐查询百度状态、证据定位、必要条件、禁止推断与显式非成功边界。该资产取代六次口径旧卡作为当前检索交付，但保留旧卡为不可变历史分母。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-retrieval-evidence-cards-2026-07-30.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "烛龙智元内容研究组", "distributions": [ { "title": "九次观察当前证据链离线验证 ZIP", "url": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-current-evidence-2026-07-30.zip", "encodingFormat": "application/zip" }, { "title": "Nine-observation current evidence RFC 9264 linkset", "url": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-current-evidence-linkset-2026-07-30.json", "encodingFormat": "application/linkset+json" }, { "title": "九次观察问法—声明证据索引 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-retrieval-evidence-cards-2026-07-30.csv", "encodingFormat": "text/csv" }, { "title": "九次观察检索证据卡 Markdown", "url": "https://course.dragonai.tech/datasets/ai-search-rag-nine-observation-retrieval-evidence-cards-2026-07-30.md", "encodingFormat": "text/markdown" } ] }, { "id": "ai-search-citation-reddit-surface-preflight-2026-07-31", "title": "Reddit cited-surface eligibility preflight (2026-07-31)", "description": "A read-only audit of the five Reddit URLs cited in the frozen five-batch benchmark, binding exact citation occurrences to subreddit policy and human-account eligibility boundaries. It is not a publication payload, backlink, or GEO success signal.", "url": "https://course.dragonai.tech/datasets/ai-search-citation-reddit-surface-preflight-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Content Research", "distributions": [ { "title": "Reddit cited-surface eligibility preflight CSV", "url": "https://course.dragonai.tech/datasets/ai-search-citation-reddit-surface-preflight-2026-07-31.csv", "encodingFormat": "text/csv" } ], "staticRoutes": [ "/research" ] }, { "id": "ai-search-citation-authorea-surface-preflight-2026-07-31", "title": "Authorea cited-surface eligibility preflight (2026-07-31)", "description": "A read-only audit of the exact Authorea URL cited four times in the frozen five-batch benchmark, binding run occurrences to Crossref metadata and permanent scholarly-record eligibility boundaries. It is not a manuscript, publication payload, backlink, or GEO success signal.", "url": "https://course.dragonai.tech/datasets/ai-search-citation-authorea-surface-preflight-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Content Research", "distributions": [ { "title": "Authorea cited-surface eligibility preflight CSV", "url": "https://course.dragonai.tech/datasets/ai-search-citation-authorea-surface-preflight-2026-07-31.csv", "encodingFormat": "text/csv" } ], "staticRoutes": [ "/research" ] }, { "id": "ai-search-citation-prepared-authority-host-coverage-gap-2026-07-31", "title": "AI 搜索引用来源与待审批权威渠道覆盖差距（2026-07-31）", "description": "将固定查询中实际出现的引用主机与待真人审批的权威渠道逐项对齐，公开未覆盖来源及证据边界；准备状态不代表已发布、已获外链或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-citation-prepared-authority-host-coverage-gap-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "待审批权威渠道覆盖差距 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-citation-prepared-authority-host-coverage-gap-2026-07-31.csv", "encodingFormat": "text/csv" } ], "staticRoutes": [ "/research" ] }, { "id": "ai-search-rag-baidu-serp-entry-liveness-2026-07-31", "title": "百度严格无品牌结果新进入 URL 存活性诊断（2026-07-31）", "description": "对相邻严格百度观测中新进入结果集的 URL 做匿名只读抓取，记录 HTTP 存活性与内容指纹；该诊断不改写排名，也不构成 DragonAI 搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "百度结果新进入 URL 存活性诊断 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-07-31.csv", "encodingFormat": "text/csv" } ], "staticRoutes": [ "/research" ] }, { "id": "ai-search-rag-baidu-serp-entry-liveness-2026-07-31-185000", "title": "百度严格无品牌结果新进入 URL 存活性复查（2026-07-31 18:50 UTC）", "description": "对后续相邻严格百度观测中新进入结果集的 URL 再做匿名只读抓取，保留 HTTP 存活性变化；该复查不改写排名，也不构成 DragonAI 搜索可见性或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-07-31-185000.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "百度结果新进入 URL 存活性复查 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-07-31-185000.csv", "encodingFormat": "text/csv" } ], "staticRoutes": [ "/research" ] }, { "id": "ai-search-rag-bingbot-discovery-to-target-crawl-gap-2026-07-31", "title": "Bingbot 发现面到研究目标页的抓取缺口（2026-07-31）", "description": "从当前生产发布的受控访问日志中提取经正反向 DNS 验证的 Bingbot 请求，公开研究 sitemap 与 Atom feed 已抓取而唯一目标页仍未抓取的最小化路径级证据；不公开 IP 或原始日志，也不把发现面抓取当作收录、排名、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bingbot-discovery-to-target-crawl-gap-2026-07-31.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "Bingbot 发现面到目标页抓取缺口 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bingbot-discovery-to-target-crawl-gap-2026-07-31.csv", "encodingFormat": "text/csv" } ], "staticRoutes": [ "/research" ] }, { "id": "ai-search-rag-baidu-serp-entry-liveness-2026-08-11t232351z", "title": "百度严格无品牌结果新进入 URL 存活性诊断（2026-08-11 23:23 UTC）", "description": "对相邻严格百度观测中唯一新进入结果 URL 做匿名只读抓取，记录其 HTTP 521 状态及退出 URL；该诊断不改写排名，也不构成 DragonAI 搜索可见性、来源质量、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-11t232351z.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI 内容研究组", "distributions": [ { "title": "百度结果新进入 URL 存活性诊断 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-11t232351z.csv", "encodingFormat": "text/csv" } ], "staticRoutes": [ "/research" ] }, { "id": "ai-search-rag-baidu-serp-entry-liveness-2026-08-12-143823", "title": "Baidu strict unbranded result-entry liveness diagnostic (2026-08-12 14:38 UTC)", "description": "Anonymous read-only fetch evidence for the single URL entering the current strict Baidu result set. The observed destination returned a Baidu verification page; this does not establish source quality, DragonAI search visibility, AI citation or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-12-143823.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Baidu result-entry liveness diagnostic CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-serp-entry-liveness-2026-08-12-143823.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-current-release-adjacent-churn-2026-08-12", "title": "Baidu strict unbranded adjacent-result churn for the current evidence release", "description": "Adjacent anonymous Baidu organic-result comparison across three fixed unbranded queries. The source-comparison query replaced six of seven URLs, while the target and brand domain remained absent from every unbranded result set.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-release-adjacent-churn-2026-08-12.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Baidu current-release adjacent-result churn CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-release-adjacent-churn-2026-08-12.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-strict-query-answer-map-v1", "title": "RAG、微调与 AI 引用来源三条严格无品牌查询答案地图", "description": "将三条固定严格无品牌查询映射到当前模型引用、百度竞争结果、Bing 十二次观察和可下载证据；DragonAI 仅在 Bing 的来源对比查询出现。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-strict-query-answer-map-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "三条严格无品牌查询答案地图 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-strict-query-answer-map-v1.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-current-baidu-ai-overlap-2026-08-14-r2", "title": "当前百度严格结果与 2026-08-14 三平台引用交集 r2", "description": "把当前百度三条严格无品牌查询的 22 个不同结果 URL 与 2026-08-14 千问、Kimi、Gemini 九次复测的 50 个不同引用 URL 做主机与精确 URL 全外联结；两侧有 2 个精确 URL 和 3 个规范化主机重合，DragonAI 在两侧均未出现。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-overlap-2026-08-14-r2.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "当前百度与三平台引用主机交集 r2 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-overlap-2026-08-14-r2.csv", "encodingFormat": "text/csv" }, { "title": "当前百度与三平台引用逐 URL 全外联结 r2 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-current-baidu-ai-overlap-2026-08-14-r2-urls.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-current-top3-delivery-features-2026-08-12", "title": "Current Baidu top-three delivery features with DragonAI target reference", "description": "Read-only delivery-feature capture for nine current top-three observations across three fixed unbranded Baidu queries, plus the DragonAI target as an unranked reference. Most competitor pages blocked direct capture, so the dataset preserves failures and does not infer ranking causality, source quality, search visibility or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-top3-delivery-features-2026-08-12.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Current Baidu top-three delivery features CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-top3-delivery-features-2026-08-12.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-baidu-current-top3-delivery-features-2026-08-14-r2", "title": "Current-release Baidu top-three delivery features with DragonAI target reference, revision 2", "description": "Read-only delivery-feature capture for the current production release and the nine top-three observations across three fixed unbranded Baidu queries, plus the DragonAI target as an unranked reference. Failed competitor fetches remain explicit; feature co-occurrence is not ranking causality, source quality, visibility or GEO success.", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-top3-delivery-features-2026-08-14-r2.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Current-release Baidu top-three delivery features revision 2 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-baidu-current-top3-delivery-features-2026-08-14-r2.csv", "encodingFormat": "text/csv" } ], "staticRoutes": [ "/research" ] }, { "id": "rag-fine-tuning-selection-query-authority-answer-card-v1", "title": "RAG 与微调选型 AI 引用来源：查询级权威答案卡 v1", "description": "把固定无品牌选型查询的直接答案、六份一手技术与治理来源、逐项验证动作、不可推断边界、2026-08-13 22:00 UTC 百度当前八个自然结果，以及当前目标与八份竞品文档的 16 指标交付对照绑定在同一对象中；仅三份竞品返回可用 HTML，DragonAI 仍为零命中。", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-query-authority-answer-card-v1.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "RAG 与微调选型查询级权威答案卡（CSV）", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-query-authority-answer-card-v1.csv", "encodingFormat": "text/csv" }, { "title": "RAG 与微调选型查询级权威答案图（JSON-LD）", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-query-authority-answer-card-v1.jsonld", "encodingFormat": "application/ld+json" }, { "title": "RAG 与微调选型查询级权威答案卡（CSL-JSON）", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-query-authority-answer-card-v1.csl.json", "encodingFormat": "application/vnd.citationstyles.csl+json" }, { "title": "RAG 与微调选型查询级权威答案卡（CFF）", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-query-authority-answer-card-v1.cff", "encodingFormat": "text/yaml" }, { "title": "RAG 与微调选型查询级权威答案卡（BibTeX）", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-query-authority-answer-card-v1.bib", "encodingFormat": "application/x-bibtex" }, { "title": "RAG 与微调选型查询级权威答案卡（RIS）", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-query-authority-answer-card-v1.ris", "encodingFormat": "application/x-research-info-systems" }, { "title": "RAG 与微调选型查询级权威答案卡（RFC 9264 Linkset）", "url": "https://course.dragonai.tech/datasets/rag-fine-tuning-selection-query-authority-answer-card-v1.linkset.json", "encodingFormat": "application/linkset+json" } ] }, { "id": "baidu-selection-answer-source-layer-gap-2026-08-13", "title": "Baidu selection-answer named, organic and concrete-citation source gap", "description": "A seven-row join of the sources named in one anonymous Baidu generated answer, the eight current organic results, and the one exposed reference-document link. Six source classes were named, four also appeared organically, one exposed a concrete citation, and DragonAI was absent from all three layers. These point-in-time layers are not interchangeable and do not prove ranking causality or GEO success.", "url": "https://course.dragonai.tech/datasets/baidu-selection-answer-source-layer-gap-2026-08-13.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Baidu selection-answer source-layer gap CSV", "url": "https://course.dragonai.tech/datasets/baidu-selection-answer-source-layer-gap-2026-08-13.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-51cto-cross-engine-channel-candidate-2026-08-14", "title": "千问 Kimi Gemini RAG 微调引用来源：51CTO 跨百度与 Bing 证据", "description": "固定无品牌查询在百度与 Bing 中命中 blog.51cto.com 的 20 条可复算记录；两侧页面不同，DragonAI 目标与品牌域均未出现，因此它只支持渠道候选判断，不构成排名因果、推荐或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-51cto-cross-engine-channel-candidate-2026-08-14.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "51CTO 跨引擎逐观察证据 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-51cto-cross-engine-channel-candidate-2026-08-14.csv", "encodingFormat": "text/csv" }, { "title": "51CTO 跨引擎证据复算说明", "url": "https://course.dragonai.tech/datasets/ai-search-rag-51cto-cross-engine-channel-candidate-2026-08-14.md", "encodingFormat": "text/markdown" } ] }, { "id": "ai-search-rag-bing-current-query-delta-2026-08-14-r2", "title": "Bing 中国无品牌三查询同发布变化 r2", "description": "2026-08-14 13:22 UTC 对同一生产发布下三条固定 Bing 中国匿名无品牌查询复核：30 个查询级 URL 出现保留 27、进入 3、退出 3，16 个留存 URL 位次变化；全部进入和退出来自第一条查询。DragonAI 目标页与品牌域均为 0，结果变化不构成排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r2.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询同发布变化 r2（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r2.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-bing-current-query-delta-2026-08-14-r3", "title": "Bing 中国无品牌三查询目标页首次点时出现 r3", "description": "2026-08-14 22:08 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在无品牌查询“AI 搜索 RAG 微调 引用来源 对比”位于第 1，另外两条查询未出现。该单次点时结果计作搜索可见性，不证明持续排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r3.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询目标页首次点时出现 r3（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r3.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-bing-current-query-delta-2026-08-14-r4", "title": "Bing 中国无品牌三查询目标页连续第二次位列第 1 r4", "description": "2026-08-14 22:44 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第二次位于第 1，另外两条仍未出现。相邻快照保留 26 个查询级 URL、进入 4、退出 4，12 个留存 URL 换位；两个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r4.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询目标页连续第二次位列第 1 r4（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r4.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-bing-current-query-delta-2026-08-14-r5", "title": "Bing 中国无品牌三查询目标页连续第三次位列第 1 r5", "description": "2026-08-14 23:10 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第三次位于第 1，另外两条仍未出现。相邻快照的 30 个查询级 URL 全部留存、没有进入或退出，12 个 URL 换位；三个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r5.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询目标页连续第三次位列第 1 r5（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r5.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-bing-current-query-delta-2026-08-14-r6", "title": "Bing 中国无品牌三查询目标页连续第四次位列第 1 r6", "description": "2026-08-14 23:41 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第四次位于第 1，另外两条仍未出现。相邻快照保留 29 个查询级 URL、进入 1、退出 1，7 个 URL 换位；四个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r6.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询目标页连续第四次位列第 1 r6（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-14-r6.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-bing-current-query-delta-2026-08-15-r7", "title": "Bing 中国无品牌三查询目标页连续第五次位列第 1 r7", "description": "2026-08-15 00:17 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第五次位于第 1，另外两条仍未出现。相邻快照保留 27 个查询级 URL、进入 3、退出 3，12 个 URL 换位；五个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r7.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询目标页连续第五次位列第 1 r7（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r7.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-bing-five-observation-persistence-2026-08-15", "title": "Bing 中国无品牌三查询五次观察 URL 与主机持久性", "description": "合并 2026-08-14 22:08 至 2026-08-15 00:17 UTC 的五次匿名 Bing 中国区只读观察：15 个固定查询快照、150 个自然结果出现、36 条查询—URL 行与 26 条查询—主机行。DragonAI 目标研究页只在一条无品牌查询中五次均位于第 1；这计作已观察搜索可见性，不证明长期排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-five-observation-persistence-2026-08-15.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询五次观察 URL 持久性（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-five-observation-persistence-2026-08-15.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-bing-current-query-delta-2026-08-15-r8", "title": "Bing 中国无品牌三查询目标页连续第六次位列第 1 r8", "description": "2026-08-15 01:06 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第六次位于第 1，另外两条仍未出现。相邻快照保留 29 个查询级 URL、进入 1、退出 1，9 个 URL 换位；六个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r8.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询目标页连续第六次位列第 1 r8（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r8.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-bing-six-observation-persistence-2026-08-15", "title": "Bing 中国无品牌三查询六次观察 URL 与主机持久性", "description": "合并 2026-08-14 22:08 至 2026-08-15 01:06 UTC 的六次匿名 Bing 中国区只读观察：18 个固定查询快照、180 个自然结果出现、36 条查询—URL 行与 26 条查询—主机行。DragonAI 目标研究页只在一条无品牌查询中六次均位于第 1；这计作已观察搜索可见性，不证明长期排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-six-observation-persistence-2026-08-15.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询六次观察 URL 持久性（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-six-observation-persistence-2026-08-15.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-bing-current-query-delta-2026-08-15-r9", "title": "Bing 中国无品牌三查询目标页连续第七次位列第 1 r9", "description": "2026-08-15 01:42 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第七次位于第 1，另外两条仍未出现。相邻快照保留 29 个查询级 URL、进入 1、退出 1，7 个 URL 换位；七个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r9.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询目标页连续第七次位列第 1 r9（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r9.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-bing-seven-observation-persistence-2026-08-15", "title": "Bing 中国无品牌三查询七次观察 URL 与主机持久性", "description": "合并 2026-08-14 22:08 至 2026-08-15 01:42 UTC 的七次匿名 Bing 中国区只读观察：21 个固定查询快照、210 个自然结果出现、36 条查询—URL 行与 26 条查询—主机行。DragonAI 目标研究页只在一条无品牌查询中七次均位于第 1；这计作已观察搜索可见性，不证明长期排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-seven-observation-persistence-2026-08-15.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询七次观察 URL 持久性（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-seven-observation-persistence-2026-08-15.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-bing-current-query-delta-2026-08-15-r10", "title": "Bing 中国无品牌三查询目标页连续第八次位列第 1 r10", "description": "2026-08-15 02:17 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第八次位于第 1，另外两条仍未出现。相邻快照 30 个查询级 URL 全部保留，4 个 URL 换位；八个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r10.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询目标页连续第八次位列第 1 r10（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r10.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-bing-eight-observation-persistence-2026-08-15", "title": "Bing 中国无品牌三查询八次观察 URL 与主机持久性", "description": "合并 2026-08-14 22:08 至 2026-08-15 02:17 UTC 的八次匿名 Bing 中国区只读观察：24 个固定查询快照、240 个自然结果出现、36 条查询—URL 行与 26 条查询—主机行。DragonAI 目标研究页只在一条无品牌查询中八次均位于第 1；这计作已观察搜索可见性，不证明长期排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-eight-observation-persistence-2026-08-15.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询八次观察 URL 持久性（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-eight-observation-persistence-2026-08-15.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-bing-current-query-delta-2026-08-15-r11", "title": "Bing 中国无品牌三查询目标页连续第九次位列第 1 r11", "description": "2026-08-15 03:52 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第九次位于第 1，另外两条仍未出现。相邻快照保留 29 个查询级 URL、进入 1、退出 1，10 个留存 URL 换位；九个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r11.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询目标页连续第九次位列第 1 r11（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r11.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-bing-nine-observation-persistence-2026-08-15", "title": "Bing 中国无品牌三查询九次观察 URL 与主机持久性", "description": "合并 2026-08-14 22:08 至 2026-08-15 03:52 UTC 的九次匿名 Bing 中国区只读观察：27 个固定查询快照、270 个自然结果出现、36 条查询—URL 行与 26 条查询—主机行。DragonAI 目标研究页只在一条无品牌查询中九次均位于第 1；这计作已观察搜索可见性，不证明长期排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-nine-observation-persistence-2026-08-15.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询九次观察 URL 持久性（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-nine-observation-persistence-2026-08-15.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-bing-current-query-delta-2026-08-15-r12", "title": "Bing 中国无品牌三查询目标页连续第十次位列第 1 r12", "description": "2026-08-15 04:58 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第十次位于第 1，另外两条仍未出现。相邻快照保留 28 个查询级 URL、进入 2、退出 2，3 个留存 URL 换位；十个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r12.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询目标页连续第十次位列第 1 r12（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r12.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-bing-ten-observation-persistence-2026-08-15", "title": "Bing 中国无品牌三查询十次观察 URL 与主机持久性", "description": "合并 2026-08-14 22:08 至 2026-08-15 04:58 UTC 的十次匿名 Bing 中国区只读观察：30 个固定查询快照、300 个自然结果出现、36 条查询—URL 行与 26 条查询—主机行。DragonAI 目标研究页只在一条无品牌查询中十次均位于第 1；这计作已观察搜索可见性，不证明长期排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-ten-observation-persistence-2026-08-15.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询十次观察 URL 持久性（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-ten-observation-persistence-2026-08-15.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-bing-current-query-delta-2026-08-15-r13", "title": "Bing 中国无品牌三查询目标页连续第十一次位列第 1 r13", "description": "2026-08-15 05:55 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第十一次位于第 1，另外两条仍未出现。相邻快照保留 26 个查询级 URL、进入 4、退出 4，12 个留存 URL 换位；十一个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r13.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询目标页连续第十一次位列第 1 r13（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r13.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-bing-eleven-observation-persistence-2026-08-15", "title": "Bing 中国无品牌三查询十一次观察 URL 与主机持久性", "description": "合并 2026-08-14 22:08 至 2026-08-15 05:55 UTC 的十一次匿名 Bing 中国区只读观察：33 个固定查询快照、330 个自然结果出现、38 条查询—URL 行与 27 条查询—主机行。DragonAI 目标研究页只在一条无品牌查询中十一次均位于第 1；这计作已观察搜索可见性，不证明长期排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-eleven-observation-persistence-2026-08-15.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询十一次观察 URL 持久性（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-eleven-observation-persistence-2026-08-15.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-bing-current-query-delta-2026-08-15-r14", "title": "Bing 中国无品牌三查询目标页连续第十二次位列第 1 r14", "description": "2026-08-15 09:02 UTC 匿名 Bing 中国区复核中，DragonAI 目标研究页在同一无品牌查询中连续第十二次位于第 1，另外两条仍未出现。相邻快照保留 27 个查询级 URL、进入 3、退出 3，7 个留存 URL 换位；十二个点时观察不证明长期稳定排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r14.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询目标页连续第十二次位列第 1 r14（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-current-query-delta-2026-08-15-r14.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-bing-twelve-observation-persistence-2026-08-15", "title": "Bing 中国无品牌三查询十二次观察 URL 与主机持久性", "description": "合并 2026-08-14 22:08 至 2026-08-15 09:02 UTC 的十二次匿名 Bing 中国区只读观察：36 个固定查询快照、360 个自然结果出现、38 条查询—URL 行与 27 条查询—主机行。DragonAI 目标研究页只在一条无品牌查询中十二次均位于第 1；这计作已观察搜索可见性，不证明长期排名、排名因果、AI 引用或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-twelve-observation-persistence-2026-08-15.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "Bing 中国无品牌三查询十二次观察 URL 持久性（CSV）", "url": "https://course.dragonai.tech/datasets/ai-search-rag-bing-twelve-observation-persistence-2026-08-15.csv", "encodingFormat": "text/csv" } ] }, { "id": "ai-search-rag-csdn-baidu-qwen-current-overlap-2026-08-14", "title": "CSDN 当前百度自然结果与合格千问引用逐出现投影", "description": "blog.csdn.net 在当前百度三问中有 4 个自然结果位次、4 个不同 URL；最近合格千问批次 3 次运行中有 11 次引用 URL 出现、4 个不同 URL。两侧精确 URL 交集为 0，绑定不同生产发布，DragonAI 目标与品牌证据均为 0，因此不构成同发布效应、共享索引、来源质量、渠道推荐或 GEO 成功。", "url": "https://course.dragonai.tech/datasets/ai-search-rag-csdn-baidu-qwen-current-overlap-2026-08-14.json", "license": "https://creativecommons.org/licenses/by/4.0/", "creator": "DragonAI Editorial Research Team", "distributions": [ { "title": "CSDN 当前百度与合格千问逐出现投影 CSV", "url": "https://course.dragonai.tech/datasets/ai-search-rag-csdn-baidu-qwen-current-overlap-2026-08-14.csv", "encodingFormat": "text/csv" } ] } ] }