让AI自省探索路径，提升知识图谱问答的准确性。
Towards Self-cognitive Exploration: Metacognitive Knowledge Graph Retrieval Augmented Generation
- 引入自我认知循环，让AI能察觉知识检索中的遗漏。
- 在五个领域数据集上，准确率显著超越现有方法。
- 适合需要精准推理的医疗、法律等专业场景。
基于知识图谱的检索增强生成（KG-RAG）通过结构化知识显著提升了大模型的推理能力。然而，现有框架多为开环系统，缺乏认知自省能力，难以识别探索过程中的缺陷，导致信息相关性漂移和证据不完整。现有针对非结构化文本的自优化方法因图谱探索具有路径依赖性，无法有效解决该问题。为此，我们提出元认知知识图谱检索增强生成（MetaKGRAG），受人类元认知机制启发，引入感知-评估-调整循环，实现路径感知的闭环优化。该机制使系统能够自评估探索质量，识别覆盖或相关性不足，并从精确的转折点进行轨迹关联修正。在医学、法律及常识推理领域的五个数据集上开展的大量实验表明，MetaKGRAG持续优于强基线的KG-RAG与自优化方法。结果验证了路径感知优化的有效性，凸显了结构化知识检索中闭环精炼的关键作用。
原文摘要 · Abstract (English)
Knowledge Graph-based Retrieval-Augmented Generation (KG-RAG) significantly enhances the reasoning capabilities of LargeLanguage Models by leveraging structured knowledge. However, existing KG-RAG frameworks typically operate as open-loop systems, suffering from cognitive blindness, an inability to recognize their exploration deficiencies. This leads to relevance drift and incomplete evidence, which existing self-refinement methods, designed for unstructured text-based RAG, cannot effectively resolve due to the path-dependent nature of graph exploration. To address this challenge, we propose Metacognitive Knowledge Graph Retrieval Augmented Generation (MetaKGRAG), a novel framework inspired by the human metacognition process, which introduces a Perceive-Evaluate-Adjust cycle to enable path-aware, closed-loop refinement. This cycle empowers the system to self-assess exploration quality, identify deficiencies in coverage or relevance, and perform trajectory-connected corrections from precise pivot points. Extensive experiments across five datasets in the medical, legal, and commonsense reasoning domains demonstrate that MetaKGRAG consistently outperforms strong KG-RAG and self-refinement baselines. Our results validate the superiority of our approach and highlight the critical need for path-aware refinement in structured knowledge retrieval.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。