AI RESEARCH
Stepwise Reasoning Enhancement for LLMs via External Subgraph Generation
arXiv CS.CL
•
ArXi:2606.04454v1 Announce Type: new Large language models have shown strong performance in natural language generation and downstream reasoning tasks, but they still struggle with logical consistency, factual grounding, and interpretability in complex multi-step reasoning. To address these limitations, this paper proposes SGR, a stepwise reasoning enhancement framework that integrates large language models with external knowledge graphs through query-relevant subgraph generation.