<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title><![CDATA[GraphRAG 适合公司知识库吗？]]></title><description><![CDATA[<p dir="auto">最近看到 GraphRAG，感觉比普通 RAG 高级。公司知识库是不是应该直接上？</p>
]]></description><link>https://localaihub.com/topic/76/graphrag-适合公司知识库吗</link><generator>RSS for Node</generator><lastBuildDate>Wed, 03 Jun 2026 19:19:54 GMT</lastBuildDate><atom:link href="https://localaihub.com/topic/76.rss" rel="self" type="application/rss+xml"/><pubDate>Mon, 04 May 2026 11:02:00 GMT</pubDate><ttl>60</ttl><item><title><![CDATA[Reply to GraphRAG 适合公司知识库吗？ on Tue, 05 May 2026 09:47:00 GMT]]></title><description><![CDATA[<p dir="auto">先定义问题，再选检索形态。别反过来。</p>
]]></description><link>https://localaihub.com/post/423</link><guid isPermaLink="true">https://localaihub.com/post/423</guid><dc:creator><![CDATA[阿航]]></dc:creator><pubDate>Tue, 05 May 2026 09:47:00 GMT</pubDate></item><item><title><![CDATA[Reply to GraphRAG 适合公司知识库吗？ on Tue, 05 May 2026 09:07:00 GMT]]></title><description><![CDATA[<p dir="auto">普通 RAG、结构化查询、GraphRAG 是工具箱，不是升级路线。</p>
]]></description><link>https://localaihub.com/post/422</link><guid isPermaLink="true">https://localaihub.com/post/422</guid><dc:creator><![CDATA[MingK]]></dc:creator><pubDate>Tue, 05 May 2026 09:07:00 GMT</pubDate></item><item><title><![CDATA[Reply to GraphRAG 适合公司知识库吗？ on Tue, 05 May 2026 08:51:00 GMT]]></title><description><![CDATA[<p dir="auto">可以从“跨文档分析”场景试点，不要从“员工手册问答”开始。</p>
]]></description><link>https://localaihub.com/post/421</link><guid isPermaLink="true">https://localaihub.com/post/421</guid><dc:creator><![CDATA[林小北]]></dc:creator><pubDate>Tue, 05 May 2026 08:51:00 GMT</pubDate></item><item><title><![CDATA[Reply to GraphRAG 适合公司知识库吗？ on Tue, 05 May 2026 06:35:00 GMT]]></title><description><![CDATA[<p dir="auto">我原来想拿 GraphRAG 解决所有召回问题，现在看不现实。</p>
]]></description><link>https://localaihub.com/post/420</link><guid isPermaLink="true">https://localaihub.com/post/420</guid><dc:creator><![CDATA[青菜]]></dc:creator><pubDate>Tue, 05 May 2026 06:35:00 GMT</pubDate></item><item><title><![CDATA[Reply to GraphRAG 适合公司知识库吗？ on Tue, 05 May 2026 03:48:00 GMT]]></title><description><![CDATA[<p dir="auto">所以每条实体关系最好能追溯到原文片段，不然图谱也会变成另一种幻觉。</p>
]]></description><link>https://localaihub.com/post/419</link><guid isPermaLink="true">https://localaihub.com/post/419</guid><dc:creator><![CDATA[nora]]></dc:creator><pubDate>Tue, 05 May 2026 03:48:00 GMT</pubDate></item><item><title><![CDATA[Reply to GraphRAG 适合公司知识库吗？ on Tue, 05 May 2026 00:58:00 GMT]]></title><description><![CDATA[<p dir="auto">还有可解释。图上边是模型抽的，用户会问这条关系从哪来。</p>
]]></description><link>https://localaihub.com/post/418</link><guid isPermaLink="true">https://localaihub.com/post/418</guid><dc:creator><![CDATA[阿白]]></dc:creator><pubDate>Tue, 05 May 2026 00:58:00 GMT</pubDate></item><item><title><![CDATA[Reply to GraphRAG 适合公司知识库吗？ on Tue, 05 May 2026 00:19:00 GMT]]></title><description><![CDATA[<p dir="auto">我们试过在项目复盘文档上做实体关系，找人和系统关系挺好。但查具体配置不如向量。</p>
]]></description><link>https://localaihub.com/post/417</link><guid isPermaLink="true">https://localaihub.com/post/417</guid><dc:creator><![CDATA[小满满]]></dc:creator><pubDate>Tue, 05 May 2026 00:19:00 GMT</pubDate></item><item><title><![CDATA[Reply to GraphRAG 适合公司知识库吗？ on Mon, 04 May 2026 23:01:00 GMT]]></title><description><![CDATA[<p dir="auto">图谱最大的坑是更新。文档每天变，关系和摘要怎么增量更新？</p>
]]></description><link>https://localaihub.com/post/416</link><guid isPermaLink="true">https://localaihub.com/post/416</guid><dc:creator><![CDATA[rootless]]></dc:creator><pubDate>Mon, 04 May 2026 23:01:00 GMT</pubDate></item><item><title><![CDATA[Reply to GraphRAG 适合公司知识库吗？ on Mon, 04 May 2026 21:00:00 GMT]]></title><description><![CDATA[<p dir="auto">不是。小团队也能试，但别一上来把制度问答改成图谱工程。</p>
]]></description><link>https://localaihub.com/post/415</link><guid isPermaLink="true">https://localaihub.com/post/415</guid><dc:creator><![CDATA[小路灯]]></dc:creator><pubDate>Mon, 04 May 2026 21:00:00 GMT</pubDate></item><item><title><![CDATA[Reply to GraphRAG 适合公司知识库吗？ on Mon, 04 May 2026 19:39:00 GMT]]></title><description><![CDATA[<p dir="auto">那是不是只有大公司能用？</p>
]]></description><link>https://localaihub.com/post/414</link><guid isPermaLink="true">https://localaihub.com/post/414</guid><dc:creator><![CDATA[小树]]></dc:creator><pubDate>Mon, 04 May 2026 19:39:00 GMT</pubDate></item><item><title><![CDATA[Reply to GraphRAG 适合公司知识库吗？ on Mon, 04 May 2026 18:41:00 GMT]]></title><description><![CDATA[<p dir="auto">GraphRAG 的构建成本不低。实体抽取、关系抽取、社区摘要、更新都要维护。</p>
]]></description><link>https://localaihub.com/post/413</link><guid isPermaLink="true">https://localaihub.com/post/413</guid><dc:creator><![CDATA[林小北]]></dc:creator><pubDate>Mon, 04 May 2026 18:41:00 GMT</pubDate></item><item><title><![CDATA[Reply to GraphRAG 适合公司知识库吗？ on Mon, 04 May 2026 15:48:00 GMT]]></title><description><![CDATA[<p dir="auto">如果问“过去半年客户投诉主要集中在哪些产品线”，图和社区摘要可能有用。</p>
]]></description><link>https://localaihub.com/post/412</link><guid isPermaLink="true">https://localaihub.com/post/412</guid><dc:creator><![CDATA[MingK]]></dc:creator><pubDate>Mon, 04 May 2026 15:48:00 GMT</pubDate></item><item><title><![CDATA[Reply to GraphRAG 适合公司知识库吗？ on Mon, 04 May 2026 15:35:00 GMT]]></title><description><![CDATA[<p dir="auto">如果用户问“某个报销标准是多少”，普通检索更直接。</p>
]]></description><link>https://localaihub.com/post/411</link><guid isPermaLink="true">https://localaihub.com/post/411</guid><dc:creator><![CDATA[阿航]]></dc:creator><pubDate>Mon, 04 May 2026 15:35:00 GMT</pubDate></item><item><title><![CDATA[Reply to GraphRAG 适合公司知识库吗？ on Mon, 04 May 2026 13:16:00 GMT]]></title><description><![CDATA[<p dir="auto">先别被名字带跑。GraphRAG 更适合全局主题、实体关系、跨文档综合，不是替代所有 RAG。</p>
]]></description><link>https://localaihub.com/post/410</link><guid isPermaLink="true">https://localaihub.com/post/410</guid><dc:creator><![CDATA[nora]]></dc:creator><pubDate>Mon, 04 May 2026 13:16:00 GMT</pubDate></item></channel></rss>