Cog-RAG: Cognitive-Inspired Dual-Hypergraph with Theme Alignment Retrieval-Augmented Generation

TL;DR

Cog-RAG introduces a novel dual-hypergraph framework inspired by human cognition, enhancing retrieval-augmented generation by modeling global themes and high-order entity relations for improved response quality.

Contribution

It proposes a theme-aligned dual-hypergraph structure and a two-stage retrieval strategy, addressing limitations of previous hypergraph-based RAG models in capturing global thematic organization.

Findings

Significantly outperforms existing RAG models in experiments.

Effectively captures inter-chunk thematic structures and high-order semantic relations.

Enhances response relevance and consistency through global and local semantic alignment.

Abstract

Retrieval-Augmented Generation (RAG) enhances the response quality and domain-specific performance of large language models (LLMs) by incorporating external knowledge to combat hallucinations. In recent research, graph structures have been integrated into RAG to enhance the capture of semantic relations between entities. However, it primarily focuses on low-order pairwise entity relations, limiting the high-order associations among multiple entities. Hypergraph-enhanced approaches address this limitation by modeling multi-entity interactions via hyperedges, but they are typically constrained to inter-chunk entity-level representations, overlooking the global thematic organization and alignment across chunks. Drawing inspiration from the top-down cognitive process of human reasoning, we propose a theme-aligned dual-hypergraph RAG framework (Cog-RAG) that uses a theme hypergraph to capture inter-chunk thematic structure and an entity hypergraph to model high-order semantic relations. Furthermore, we design a cognitive-inspired two-stage retrieval strategy that first activates query-relevant thematic content from the theme hypergraph, and then guides fine-grained recall and diffusion in the entity hypergraph, achieving semantic alignment and consistent generation from global themes to local details. Our extensive experiments demonstrate that Cog-RAG significantly outperforms existing state-of-the-art baseline approaches.