用多智能体动态构建图谱，让小模型也能高效推理复杂问题。
Think-on-Graph 3.0: Efficient and Adaptive LLM Reasoning on Heterogeneous Graphs via Multi-Agent Dual-Evolving Context Retrieval
- 多智能体协同演化查询与子图，自适应构建专属知识图谱
- 在多个推理任务上超越基线，小模型表现接近大模型
- 适合资源有限但需精准推理的场景，如本地部署
基于图谱的检索增强生成（GraphRAG）已成为提升大语言模型外部知识能力的重要范式。然而，现有方法受限于高质量知识图谱的依赖：人工构建难扩展，自动提取又受制于大模型抽取器性能，尤其在使用小型本地部署模型时表现不佳。为此，我们提出 Think-on-Graph 3.0（ToG-3），其核心是多智能体上下文演化与检索（MACER）机制。该框架通过双演化过程，动态构建并迭代优化由块、三元组、社区组成的异构图索引，实现查询与子图的协同进化。ToG-3能根据查询动态生成目标图谱，即使在轻量级大模型下也能精准召回证据并完成推理。大量实验表明，ToG-3在深度与广度推理基准上均优于对比基线，消融实验证实了MACER各组件的有效性。源代码已公开于 https://github.com/DataArcTech/ToG-3。
原文摘要 · Abstract (English)
Graph-based Retrieval-Augmented Generation (GraphRAG) has become the important paradigm for enhancing Large Language Models (LLMs) with external knowledge. However, existing approaches are constrained by their reliance on high-quality knowledge graphs: manually built ones are not scalable, while automatically extracted ones are limited by the performance of LLM extractors, especially when using smaller, local-deployed models. To address this, we introduce Think-on-Graph 3.0 (ToG-3), a novel framework featuring a Multi-Agent Context Evolution and Retrieval (MACER) mechanism. Its core contribution is the dynamic construction and iterative refinement of a Chunk-Triplets-Community heterogeneous graph index, powered by a Dual-Evolution process that adaptively evolves both the query and the retrieved sub-graph during reasoning. ToG-3 dynamically builds a targeted graph index tailored to the query, enabling precise evidence retrieval and reasoning even with lightweight LLMs. Extensive experiments demonstrate that ToG-3 outperforms compared baselines on both deep and broad reasoning benchmarks, and ablation studies confirm the efficacy of the components of MACER framework. The source code are available in https://github.com/DataArcTech/ToG-3.
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。