Hierarchical Retrieval-Augmented Generation Model with Rethink for Multi-hop Question Answering

TL;DR

This paper introduces HiRAG, a hierarchical retrieval-augmented generation framework for multi-hop question answering that combines sparse and dense retrieval strategies, addressing knowledge update issues and improving performance.

Contribution

The paper proposes a novel hierarchical retrieval strategy and a unified retrieval method, along with new corpora, to enhance multi-hop QA systems' accuracy and knowledge freshness.

Findings

HiRAG outperforms state-of-the-art models on four datasets.

Hierarchical retrieval effectively combines document and chunk-level information.

New corpora improve knowledge update and retrieval effectiveness.

Abstract

Multi-hop Question Answering (QA) necessitates complex reasoning by integrating multiple pieces of information to resolve intricate questions. However, existing QA systems encounter challenges such as outdated information, context window length limitations, and an accuracy-quantity trade-off. To address these issues, we propose a novel framework, the Hierarchical Retrieval-Augmented Generation Model with Rethink (HiRAG), comprising Decomposer, Definer, Retriever, Filter, and Summarizer five key modules. We introduce a new hierarchical retrieval strategy that incorporates both sparse retrieval at the document level and dense retrieval at the chunk level, effectively integrating their strengths. Additionally, we propose a single-candidate retrieval method to mitigate the limitations of multi-candidate retrieval. We also construct two new corpora, Indexed Wikicorpus and Profile Wikicorpus, to address the issues of outdated and insufficient knowledge. Our experimental results on four datasets demonstrate that HiRAG outperforms state-of-the-art models across most metrics, and our Indexed Wikicorpus is effective. The code for HiRAG is available at https://github.com/2282588541a/HiRAG