面向动态知识检索的大语言模型与知识图谱协同推理方法OA
Collaborative Reasoning Method Between LLMs and Knowledge Graphs via Dynamic Knowledge Retrieval
针对大语言模型存在知识静态性及事实幻觉,而知识图谱又面临覆盖不全与更新滞后的问题,提出一种面向动态知识检索的大语言模型与知识图谱协同推理框架.该框架通过闭环机制实现知识动态检索、协同推理与增量更新.首先,采用子图增强的检索生成策略,基于问题分解与多跳推理从知识图谱中提取相关子图,提升复杂查询下的知识召回精度;其次,通过多维度混合加权方法,从非结构化文本中检索相关段落以补充知识缺失;最后,借助大语言模型对多源知识进行融合推理,生成答案并抽取出新增三元组,实现知识图谱的动态增量更新.实验表明,该方法能有效处理复杂多跳问题,显著降低事实幻觉,并通过闭环更新保障知识库的时效性与持续演进能力.
To address the issues of static knowledge and factual hallucinations in large language mod-els(LLMs),as well as the incomplete coverage and delayed updates in knowledge graphs,a collabora-tive reasoning framework integrating LLMs and knowledge graphs is proposed via dynamic knowledge retrieval.A closed-loop mechanism is designed to implement dynamic knowledge retrieval,collabora-tive reasoning,and incremental updating.Firstly,a subgraph-enhanced retrieval strategy is adopted,which extracts relevant subgraphs from the knowledge graph through question decomposition and multi-hop reasoning to improve the recall under complex queries.Secondly,semantically relevant pas-sages are retrieved from unstructured texts through a multi-dimensional hybrid weighting method to supplement missing knowledge.Finally,LLMs are utilized to integrate multi-source knowledge for rea-soning,generating answers while extracting new triples for dynamically updating the knowledge graph.Experiments demonstrate that the proposed method can effectively handle complex multi-hop ques-tions,reduce factual hallucinations,and ensure timeliness and continuous evolution of the knowledge base through closed-loop updating.
夏毅;贺浩峰;陈静;周刚;郑嘉丽;张凯翔
信息工程大学,河南 郑州 450001信息工程大学,河南 郑州 450001信息工程大学,河南 郑州 450001信息工程大学,河南 郑州 450001信息工程大学,河南 郑州 450001信息工程大学,河南 郑州 450001
信息技术与安全科学
知识推理大语言模型知识图谱检索增强生成
knowledge reasoningLLMsknowledge graphretrieval-augmented generation
《信息工程大学学报》 2026 (4)
450-456,465,8
国家自然科学基金(42371438)河南省科技攻关项目(222102210081,222300420590)
评论