arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.23372cs.IRcs.AIcs.CL
收藏

普及图式RAG:线性、仅CPU的多跳问答图检索

Democratizing GraphRAG: Linear, CPU-Only Graph Retrieval for Multi-Hop QA

PingCAP, Data & AI-Innovation Lab(PingCAP数据与人工智能创新实验室)

Qizhi Wang

AI总结:

SPRIG通过使用CPU和轻量命名实体识别构建图,实现线性时间、无token的多跳问答图检索,提升召回率并降低对GPU和大语言模型的依赖。

摘要:

图式RAG系统通过建模结构来提高多跳检索,但许多方法依赖于昂贵的基于大语言模型的图构建和GPU密集型推理。我们提出了SPRIG(图中检索的种子传播),一种仅使用CPU、线性时间、无token的图式RAG流程,该流程用轻量的命名实体识别驱动的共现图替代基于大语言模型的图构建,并使用个性化页面排名(PPR)实现28%的提升,同时对Recall@10的影响微乎其微。结果表明了在何种情况下CPU友好的图检索有助于多跳召回,以及何时强词法混合(RRF)就足够,从而为在无token成本或GPU需求的情况下普及图式RAG提供了现实路径。

英文摘要:

GraphRAG systems improve multi-hop retrieval by modeling structure, but many approaches rely on expensive LLM-based graph construction and GPU-heavy inference. We present SPRIG (Seeded Propagation for Retrieval In Graphs), a CPU-only, linear-time, token-free GraphRAG pipeline that replaces LLM graph building with lightweight NER-driven co-occurrence graphs and uses Personalized PageRank (PPR) for 28% with negligible Recall@10 changes. The results characterize when CPU-friendly graph retrieval helps multi-hop recall and when strong lexical hybrids (RRF) are sufficient, outlining a realistic path to democratizing GraphRAG without token costs or GPU requirements.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑