GeAR: Graph-enhanced Agent for Retrieval-augmented Generation
This addresses the bottleneck of multi-hop retrieval for RAG systems, offering a significant performance gain in question answering tasks.
The paper tackles the problem of multi-hop retrieval in Retrieval-augmented Generation (RAG) by introducing GeAR, a system that enhances retrieval performance with a graph expansion mechanism and agent framework, achieving state-of-the-art results with over 10% improvement on the MuSiQue dataset while using fewer tokens and iterations.
Retrieval-augmented Generation (RAG) relies on effective retrieval capabilities, yet traditional sparse and dense retrievers inherently struggle with multi-hop retrieval scenarios. In this paper, we introduce GeAR, a system that advances RAG performance through two key innovations: (i) an efficient graph expansion mechanism that augments any conventional base retriever, such as BM25, and (ii) an agent framework that incorporates the resulting graph-based retrieval into a multi-step retrieval framework. Our evaluation demonstrates GeAR's superior retrieval capabilities across three multi-hop question answering datasets. Notably, our system achieves state-of-the-art results with improvements exceeding 10% on the challenging MuSiQue dataset, while consuming fewer tokens and requiring fewer iterations than existing multi-step retrieval systems. The project page is available at https://gear-rag.github.io.