用大模型引导知识图谱遍历，低成本提升RAG性能
ReMindRAG: Low-Cost LLM-Guided Knowledge Graph Traversal for Efficient RAG
- 大模型指导图谱遍历，融合探索、利用与记忆回放
- 通过边嵌入存储遍历经验，减少重复计算
- 无需训练，适配多种大模型和数据集
知识图谱（KG）凭借其结构化表示能力，为增强检索增强生成（RAG）系统提供了潜力，催生了KG-RAG系统。然而，现有方法难以在系统效果与成本效率之间取得平衡，导致性能不理想或大模型提示词过多、推理时间过长。为此，本文提出ReMindRAG，采用大模型引导的图谱遍历机制，包含节点探索、节点利用及最具特色的能力——记忆回放，以同时提升系统效能与成本效益。具体而言，ReMindRAG将遍历经验存储于知识图谱边嵌入中，类似大模型在参数中‘记忆’世界知识，但无需训练。我们从理论和实验上验证了ReMindRAG的有效性，在多个基准数据集和大模型基座上均优于现有基线。代码已开源：https://github.com/kilgrims/ReMindRAG。
原文摘要 · Abstract (English)
Knowledge graphs (KGs), with their structured representation capabilities, offer promising avenue for enhancing Retrieval Augmented Generation (RAG) systems, leading to the development of KG-RAG systems. Nevertheless, existing methods often struggle to achieve effective synergy between system effectiveness and cost efficiency, leading to neither unsatisfying performance nor excessive LLM prompt tokens and inference time. To this end, this paper proposes REMINDRAG, which employs an LLM-guided graph traversal featuring node exploration, node exploitation, and, most notably, memory replay, to improve both system effectiveness and cost efficiency. Specifically, REMINDRAG memorizes traversal experience within KG edge embeddings, mirroring the way LLMs "memorize" world knowledge within their parameters, but in a train-free manner. We theoretically and experimentally confirm the effectiveness of REMINDRAG, demonstrating its superiority over existing baselines across various benchmark datasets and LLM backbones. Our code is available at https://github.com/kilgrims/ReMindRAG.
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。