Relink:为GraphRAG构建查询驱动的证据图谱
Relink: Constructing Query-Driven Evidence Graph On-the-Fly for GraphRAG
AI总结:
Relink通过动态构建查询驱动的证据图谱,解决GraphRAG中知识图谱不完整和干扰事实的问题,提升问答性能。
摘要:
基于图谱的检索增强生成(GraphRAG)通过将大型语言模型(LLMs)与结构化知识相结合,缓解了幻觉问题。然而,当前的GraphRAG方法受限于普遍存在的'构建后再推理'范式,这种范式依赖于静态的、预先构建的知识图谱(KG)。这种范式面临两个关键挑战。首先,KG的固有不完整性常常破坏推理路径。其次,图谱的低信噪比引入了误导性的事实,呈现了与查询相关但具有误导性的知识,破坏了推理过程。为了解决这些挑战,我们主张采用'推理后再构建'的范式,并提出Relink,一种动态构建查询特定证据图谱的框架。为解决不完整性问题,Relink从原始文本语料中衍生出的潜在关系池中实例化所需事实,实时修复断裂的路径。为处理误导或干扰事实,Relink采用统一的、查询感知的评估策略,同时考虑KG和潜在关系中的候选者,选择最有利于回答查询的事实,而不是依赖其预先存在。这使Relink能够主动剔除干扰事实,并为每个查询构建最忠实和精确的证据路径。在五个开放领域问答基准上的广泛实验表明,Relink在EM和F1指标上分别比领先的GraphRAG基线提升了5.4%和5.2%,证明了我们提出框架的优越性。
英文摘要:
Graph-based Retrieval-Augmented Generation (GraphRAG) mitigates hallucinations in Large Language Models (LLMs) by grounding them in structured knowledge. However, current GraphRAG methods are constrained by a prevailing \textit{build-then-reason} paradigm, which relies on a static, pre-constructed Knowledge Graph (KG). This paradigm faces two critical challenges. First, the KG's inherent incompleteness often breaks reasoning paths. Second, the graph's low signal-to-noise ratio introduces distractor facts, presenting query-relevant but misleading knowledge that disrupts the reasoning process. To address these challenges, we argue for a \textit{reason-and-construct} paradigm and propose Relink, a framework that dynamically builds a query-specific evidence graph. To tackle incompleteness, \textbf{Relink} instantiates required facts from a latent relation pool derived from the original text corpus, repairing broken paths on the fly. To handle misleading or distractor facts, Relink employs a unified, query-aware evaluation strategy that jointly considers candidates from both the KG and latent relations, selecting those most useful for answering the query rather than relying on their pre-existence. This empowers Relink to actively discard distractor facts and construct the most faithful and precise evidence path for each query. Extensive experiments on five Open-Domain Question Answering benchmarks show that Relink achieves significant average improvements of 5.4\% in EM and 5.2\% in F1 over leading GraphRAG baselines, demonstrating the superiority of our proposed framework.
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。