arXiv 2505.225712025-05-30cs.CLcs.AIcs.DBcs.IR

Agent-UniRAG：用于统一检索增强生成系统的可训练开源LLM智能体框架

Agent-UniRAG: A Trainable Open-Source LLM Agent Framework for Unified Retrieval-Augmented Generation Systems

Viettel Group（越通集团）；Hanoi University of Science and Technology（河内科学技术大学）

Hoang Pham, Thuy-Duong Nguyen, Khac-Hoai Nam Bui

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出可训练开源框架Agent-UniRAG，通过LLM智能体根据输入复杂度以端到端方式逐步处理单跳与多跳查询，并引入SynAgent-RAG数据集使小型开源LLM在多RAG基准上达到与闭源大模型相当的性能。

摘要：

本文提出了一种利用近期兴起的大语言模型（LLM）智能体概念来构建统一检索增强生成（RAG）系统的新方法。具体而言，以LLM作为基础控制器的Agent LLM已成为提升RAG任务可解释性的一种有前景的方法，尤其适用于复杂推理问答系统（例如多跳查询）。尽管如此，先前的研究主要集中在分别使用单跳或多跳方法来解决RAG系统问题，这限制了这些方法在实际应用中的适用性。在本研究中，我们提出了一个名为Agent-UniRAG的可训练智能体框架，用于统一检索增强LLM系统，该框架增强了RAG系统的有效性和可解释性。其主要思想是设计一个LLM智能体框架，根据输入的复杂程度逐步解决RAG任务，同时以端到端的方式处理单跳和多跳查询。此外，我们引入了SynAgent-RAG，这是一个合成数据集，旨在使所提出的智能体框架能够适用于小型开源LLM（例如Llama-3-8B）。结果表明，在各种RAG基准测试中，其性能与闭源和更大的开源LLM相当。我们的源代码和数据集已公开，供进一步研究使用。

英文摘要：

This paper presents a novel approach for unified retrieval-augmented generation (RAG) systems using the recent emerging large language model (LLM) agent concept. Specifically, Agent LLM, which utilizes LLM as fundamental controllers, has become a promising approach to enable the interpretability of RAG tasks, especially for complex reasoning question-answering systems (e.g., multi-hop queries). Nonetheless, previous works mainly focus on solving RAG systems with either single-hop or multi-hop approaches separately, which limits the application of those approaches to real-world applications. In this study, we propose a trainable agent framework called Agent-UniRAG for unified retrieval-augmented LLM systems, which enhances the effectiveness and interpretability of RAG systems. The main idea is to design an LLM agent framework to solve RAG tasks step-by-step based on the complexity of the inputs, simultaneously including single-hop and multi-hop queries in an end-to-end manner. Furthermore, we introduce SynAgent-RAG, a synthetic dataset to enable the proposed agent framework for small open-source LLMs (e.g., Llama-3-8B). The results show comparable performances with closed-source and larger open-source LLMs across various RAG benchmarks. Our source code and dataset are publicly available for further exploitation.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。