跳到主要内容Skip to main content

GraphRAGGraphRAG

GraphRAG

  • 数据与检索Data & retrieval
  • 第三站 · 用深Station 3 · Going Deeper
  • 引用核验 2026-11-19refs checked 2026-11-19
解决什么问题What problem it solves

从文档里抽实体和关系,建成图,再用图结构回答「跨文档、全局性」的问题。Extract entities and relations into a graph so the system can answer cross-document and global questions.

这是什么What it is

索引阶段用模型抽取实体、关系与社区摘要,形成知识图谱。

查询阶段先定位相关实体/社区,再沿边扩展上下文。

它补的是向量 RAG 不擅长的「多跳关系」和「整个语料在讲什么」。

Indexing extracts entities, relations and community summaries into a knowledge graph.

Querying locates relevant entities/communities and expands along edges.

It covers multi-hop relations and corpus-level themes that vector RAG misses.

为什么需要知道Why it matters

当问题不是找某一段原文,而是「这些文档之间谁和谁有关」时,纯向量检索会漏。When the question is “how are these documents related” rather than “find this passage”, pure vector search falls short.

出处Sources
  • GraphRAG 论文(arXiv)From Local to Global: A Graph RAG Approach to Query-Focused Summarization (arXiv 2404.16130) ↗