用强化学习让图谱检索自动优化，提升大模型问答准确率。
Graph-R1: Towards Agentic GraphRAG Framework via End-to-end Reinforcement Learning
- 通过强化学习构建可自适应检索的图谱代理系统
- 在多个基准上实现更高推理准确率与检索效率
- 适合需要精准知识检索的智能问答场景
检索增强生成（RAG）通过引入外部知识缓解大模型幻觉问题，但依赖基于片段的检索，缺乏结构语义。图谱RAG方法通过实体-关系图建模知识，但仍面临构建成本高、仅一次固定检索、依赖长上下文推理和提示设计的问题。为此，我们提出Graph-R1，首个基于端到端强化学习的智能图谱RAG框架。它采用轻量级知识超图构建，将检索建模为多轮代理-环境交互，并通过端到端奖励机制优化整个代理过程。在标准RAG数据集上的实验表明，Graph-R1在推理准确率、检索效率和生成质量方面均优于传统图谱RAG及强化学习增强型RAG方法。代码与数据已公开于https://github.com/LHRLAB/Graph-R1。
原文摘要 · Abstract (English)
Retrieval-Augmented Generation (RAG) mitigates hallucination in LLMs by incorporating external knowledge, but relies on chunk-based retrieval that lacks structural semantics. GraphRAG methods improve RAG by modeling knowledge as entity-relation graphs, but still face challenges in high construction cost, fixed one-time retrieval, and reliance on long-context reasoning and prompt design. To address these challenges, we propose Graph-R1, the first agentic GraphRAG framework via end-to-end reinforcement learning (RL). It introduces lightweight knowledge hypergraph construction, models retrieval as a multi-turn agent-environment interaction, and optimizes the agent process via an end-to-end reward mechanism. Experiments on standard RAG datasets show that Graph-R1 outperforms traditional GraphRAG and RL-enhanced RAG methods in reasoning accuracy, retrieval efficiency, and generation quality. Our software and data are publicly available at https://github.com/LHRLAB/Graph-R1.
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。