RSF-GLLM: Bridging the Semantic Gap in Multi-Hop Knowledge Graph QA via Recurrent Soft-Flow and Decoupled LLM Generation

2026-07-07

AI Summary

This work addresses the challenge in multi-hop knowledge graph question answering where the separation between retrieval and reading renders reasoning non-differentiable and hinders modeling of intermediate reasoning paths across semantic gaps. To overcome this, the authors propose the RSF-GLLM framework, which decouples differentiable graph-based reasoning from answer generation. It introduces a recurrent soft-flow mechanism to propagate continuous relevance scores and employs dynamic gating to traverse semantically mismatched nodes, while flow sparsity regularization ensures convergence of soft probabilities to discrete reasoning paths. The extracted paths are then used to fine-tune a large language model for factually consistent and efficient answer generation. Evaluated on WebQSP and ComplexWebQuestions (CWQ), the method achieves competitive performance, significantly outperforming high-computation pure large language model approaches while offering superior inference efficiency.

Abstract

Multi-hop Question Answering over Knowledge Graphs faces a critical challenge: traditional retrieve-then-read pipelines break differentiability, preventing the retriever from learning to bridge the semantic gap where intermediate nodes lack lexical overlap with the query. To address this, we propose RSF-GLLM, a framework decoupling differentiable graph reasoning from answer generation. Our Recurrent Soft-Flow (RSF) module employs a GRU-guided query updater to propagate continuous relevance scores, utilizing a dynamic gating mechanism to traverse semantically dissimilar bridge nodes via structural cues. We introduce flow sparsity regularization to theoretically guarantee convergence from soft probabilities to discrete reasoning paths. These paths are extracted and textualized to fine-tune a Large Language Model (LLM), ensuring generation is grounded in factual topology. Experiments on WebQSP and CWQ demonstrate that RSF-GLLM achieves competitive performance with superior inference efficiency compared to LLM based computationally expensive approaches.

Problem

• multi-hop QA
• semantic gap
• knowledge graph
• differentiability
• bridge nodes

Innovation

• Recurrent Soft-Flow
• Decoupled LLM Generation
• Multi-hop Knowledge Graph QA
• Semantic Gap Bridging
• Differentiable Reasoning