A Unified Framework for Context-Aware and Relation-Aware Graph Retrieval-Augmented Generation

  • 类型:arxiv
  • 标识:2606.18075
  • 链接:http://arxiv.org/abs/2606.18075v1
  • 主分类:rag
  • 形态:method
  • 被引:0
  • 被引来源:Semantic Scholar + OpenAlex
  • S2被引:0
  • OpenAlex被引:0
  • 影响力被引:0
  • TLDR:HyGRAG is proposed, a hierarchical graph RAG framework that transcends source documents by addressing three core challenges: constructing summaries that genuinely integrate contextual and relational information, leveraging these synthesized representations to access emergent knowledge during retrieval, and efficiently updating hierarchical structures for dynamic corpora.
  • OpenAlex ID:W7165025267
  • OpenAlex DOI:10.48550/arxiv.2606.18075
  • DOI:10.48550/arxiv.2606.18075
  • DOI来源:OpenAlex
  • 开放获取:green
  • 开放获取链接:https://doi.org/10.48550/arxiv.2606.18075
  • OpenAlex更新:2026-07-19
  • 待LLM分类:否
  • 标题中文:面向上下文感知与关系感知图 RAG 的统一框架
  • TLDR中文:提出 HyGRAG,一种分层图 RAG 框架,通过解决三个核心挑战超越源文档限制:构建真正融合上下文与关系信息的摘要、利用这些综合表示在检索阶段访问涌现知识、以及为动态语料高效更新分层结构。
  • 来源文件
  • /inbox/tom/_candidates/2026-06-17-agent-rag-longcontext-candidates.json
  • [S2 enrich]
  • [OpenAlex backfill]