DH-RAG: A Dynamic Historical Context-Powered Retrieval-Augmented Generation Method for Multi-Turn Dialogue

2025-02-19

AI Summary

Traditional RAG approaches for multi-turn dialogue neglect dynamic historical context, leading to suboptimal response relevance and coherence. To address this, we propose a dynamic history-aware RAG framework. Our core innovation is the construction of the first dynamic historical information database, coupled with three novel query optimization strategies: historical query clustering, hierarchical semantic matching, and chain-of-thought tracking—designed to emulate the synergistic interplay between human long-term memory and immediate contextual awareness. The method integrates history-informed query reformulation with real-time context updating, enabling history-adaptive retrieval and generation. Evaluated across multiple multi-turn dialogue benchmarks, our approach significantly outperforms baseline RAG models, achieving consistent improvements in response relevance, consistency, and overall dialogue quality.

Technology Category

• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Natural Language Processing: Question Answering
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

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

Abstract

Retrieval-Augmented Generation (RAG) systems have shown substantial benefits in applications such as question answering and multi-turn dialogue citep{lewis2020retrieval}. However, traditional RAG methods, while leveraging static knowledge bases, often overlook the potential of dynamic historical information in ongoing conversations. To bridge this gap, we introduce DH-RAG, a Dynamic Historical Context-Powered Retrieval-Augmented Generation Method for Multi-Turn Dialogue. DH-RAG is inspired by human cognitive processes that utilize both long-term memory and immediate historical context in conversational responses citep{stafford1987conversational}. DH-RAG is structured around two principal components: a History-Learning based Query Reconstruction Module, designed to generate effective queries by synthesizing current and prior interactions, and a Dynamic History Information Updating Module, which continually refreshes historical context throughout the dialogue. The center of DH-RAG is a Dynamic Historical Information database, which is further refined by three strategies within the Query Reconstruction Module: Historical Query Clustering, Hierarchical Matching, and Chain of Thought Tracking. Experimental evaluations show that DH-RAG significantly surpasses conventional models on several benchmarks, enhancing response relevance, coherence, and dialogue quality.

Problem

• Improves multi-turn dialogue with dynamic historical context.
• Enhances RAG by integrating long-term and immediate memory.
• Updates historical data continuously for better conversation quality.

Innovation

• Dynamic Historical Context-Powered
• History-Learning Query Reconstruction
• Dynamic History Information Updating