多智能体协作优化生成，自训练提升复杂问答表现
CIIR@LiveRAG 2025: Optimizing Multi-Agent Retrieval Augmented Generation through Self-Training
- 分角色智能体协同完成规划、搜索、推理与协调任务
- 自训练结合奖励引导采样，在竞赛数据集上超越传统RAG
- 适合需要多步推理的复杂真实场景问答系统
本文提出mRAG，一种由专门负责规划、搜索、推理和协调等子任务的智能体组成的多智能体检索增强生成框架。系统采用自训练范式，结合奖励引导的轨迹采样，优化智能体间协作并提升生成质量。在SIGIR 2025 LiveRAG竞赛中，基于DataMorgana数据集的评估显示，mRAG优于传统RAG基线模型。我们进一步分析了竞赛结果，并通过案例研究展示该框架在复杂真实RAG任务中的有效性。
原文摘要 · Abstract (English)
This paper presents mRAG, a multi-agent retrieval-augmented generation (RAG) framework composed of specialized agents for subtasks such as planning, searching, reasoning, and coordination. Our system uses a self-training paradigm with reward-guided trajectory sampling to optimize inter-agent collaboration and enhance response generation. Evaluated on DataMorgana-derived datasets during the SIGIR 2025 LiveRAG competition, mRAG outperforms conventional RAG baselines. We further analyze competition outcomes and showcase the framework's strengths with case studies, demonstrating its efficacy for complex, real-world RAG tasks.
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。