Enregistré dans:
Détails bibliographiques
Auteurs principaux: Fang, Jinyuan, Meng, Zaiqiao, Macdonald, Craig
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2502.18397
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912245736275968
author Fang, Jinyuan
Meng, Zaiqiao
Macdonald, Craig
author_facet Fang, Jinyuan
Meng, Zaiqiao
Macdonald, Craig
contents Iterative retrieval-augmented generation (iRAG) models offer an effective approach for multi-hop question answering (QA). However, their retrieval process faces two key challenges: (1) it can be disrupted by irrelevant documents or factually inaccurate chain-of-thoughts; (2) their retrievers are not designed to dynamically adapt to the evolving information needs in multi-step reasoning, making it difficult to identify and retrieve the missing information required at each iterative step. Therefore, we propose KiRAG, which uses a knowledge-driven iterative retriever model to enhance the retrieval process of iRAG. Specifically, KiRAG decomposes documents into knowledge triples and performs iterative retrieval with these triples to enable a factually reliable retrieval process. Moreover, KiRAG integrates reasoning into the retrieval process to dynamically identify and retrieve knowledge that bridges information gaps, effectively adapting to the evolving information needs. Empirical results show that KiRAG significantly outperforms existing iRAG models, with an average improvement of 9.40% in R@3 and 5.14% in F1 on multi-hop QA.
format Preprint
id arxiv_https___arxiv_org_abs_2502_18397
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle KiRAG: Knowledge-Driven Iterative Retriever for Enhancing Retrieval-Augmented Generation
Fang, Jinyuan
Meng, Zaiqiao
Macdonald, Craig
Computation and Language
Iterative retrieval-augmented generation (iRAG) models offer an effective approach for multi-hop question answering (QA). However, their retrieval process faces two key challenges: (1) it can be disrupted by irrelevant documents or factually inaccurate chain-of-thoughts; (2) their retrievers are not designed to dynamically adapt to the evolving information needs in multi-step reasoning, making it difficult to identify and retrieve the missing information required at each iterative step. Therefore, we propose KiRAG, which uses a knowledge-driven iterative retriever model to enhance the retrieval process of iRAG. Specifically, KiRAG decomposes documents into knowledge triples and performs iterative retrieval with these triples to enable a factually reliable retrieval process. Moreover, KiRAG integrates reasoning into the retrieval process to dynamically identify and retrieve knowledge that bridges information gaps, effectively adapting to the evolving information needs. Empirical results show that KiRAG significantly outperforms existing iRAG models, with an average improvement of 9.40% in R@3 and 5.14% in F1 on multi-hop QA.
title KiRAG: Knowledge-Driven Iterative Retriever for Enhancing Retrieval-Augmented Generation
topic Computation and Language
url https://arxiv.org/abs/2502.18397