KiRAG: Knowledge-Driven Iterative Retriever for Enhancing Retrieval-Augmented Generation

2025-02-25

AI Summary

In multi-hop question answering, iterative Retrieval-Augmented Generation (iRAG) faces two key bottlenecks: (1) retrieval susceptibility to irrelevant documents and erroneous reasoning, and (2) inflexibility of retrievers in adapting to evolving information needs. To address these, we propose a knowledge-triple-driven iRAG framework that explicitly integrates reasoning into retrieval. Our method models factual structure via knowledge graph triple decomposition, designs a reasoning-guided iterative retrieval mechanism to identify information gaps, and introduces dynamic information need representation coupled with a retrieval-generation co-optimization objective. Evaluated on multi-hop QA benchmarks, our approach achieves +9.40% Recall@3 and +5.14% F1 over state-of-the-art iRAG models. The core contribution is the first realization of structured-knowledge-based, factually reliable, and adaptive iterative retrieval.

Technology Category

• Knowledge Representation and Reasoning: Computational Complexity of Reasoning
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

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.

Problem

• Enhance retrieval-augmented generation accuracy
• Address irrelevant document disruption
• Adapt dynamically to evolving information needs

Innovation

• Knowledge-driven iterative retriever model
• Decomposes documents into knowledge triples
• Integrates reasoning into retrieval process