Text-Graph Synergy: A Bidirectional Verification and Completion Framework for RAG

TL;DR

TGS-RAG introduces a bidirectional Text-Graph synergy framework that enhances retrieval accuracy and reasoning in RAG by refining evidence and resurrecting pruned reasoning paths.

Contribution

It proposes a novel bidirectional mechanism with graph-to-text and text-to-graph channels, addressing the 'Information Island' problem in hybrid RAG systems.

Findings

Outperforms state-of-the-art baselines on multiple reasoning benchmarks.

Achieves better balance between retrieval precision and efficiency.

Effectively filters semantic noise and resurrects pruned reasoning paths.

Abstract

Retrieval-Augmented Generation (RAG) has become a core paradigm for enhancing factual grounding and multi-hop reasoning in Large Language Models (LLMs). Traditional text-based RAG often retrieves logically irrelevant pseudo-evidence, while graph-based RAG is frequently hindered by search-time pruning, which may discard potentially valid reasoning paths. Existing hybrid approaches primarily adopt simple evidence concatenation or unidirectional enhancement, which fails to address the fundamental "Information Island" problem caused by asymmetric reasoning flows between unstructured text and structured graphs. We propose \textbf{TGS-RAG}, a unified framework for \textbf{T}ext-\textbf{G}raph \textbf{S}ynergistic enhancement. TGS-RAG introduces a bidirectional mechanism: (i) a \textbf{Graph-to-Text} channel that employs a Global Voting strategy from visited graph nodes to re-rank and refine textual evidence, filtering out semantic noise; and (ii) a \textbf{Text-to-Graph} channel that utilizes the \textbf{Memory-based Orphan Entity Bridging} algorithm. This algorithm utilizes textual cues to proactively resurrect valid but previously pruned reasoning paths from the search history without additional database overhead. Experimental results on multiple multi-hop reasoning benchmarks demonstrate that TGS-RAG significantly outperforms state-of-the-art baselines, achieving a superior balance between retrieval precision and computational efficiency.