A Stepwise-Enhanced Reasoning Framework for Large Language Models Based on External Subgraph Generation

TL;DR

This paper introduces SGR, a framework that enhances large language models' reasoning by dynamically generating and utilizing external knowledge subgraphs, leading to more accurate and consistent outputs in complex tasks.

Contribution

The paper proposes a novel stepwise reasoning framework using external subgraph generation to improve LLMs' reasoning accuracy and reduce noise influence.

Findings

SGR outperforms baseline models on multiple benchmarks.

External subgraph guidance improves reasoning accuracy.

Stepwise reasoning reduces noise and irrelevant information influence.

Abstract

Large Language Models (LLMs) have achieved strong performance across a wide range of natural language processing tasks in recent years, including machine translation, text generation, and question answering. As their applications extend to increasingly complex scenarios, however, LLMs continue to face challenges in tasks that require deep reasoning and logical inference. In particular, models trained on large scale textual corpora may incorporate noisy or irrelevant information during generation, which can lead to incorrect predictions or outputs that are inconsistent with factual knowledge. To address this limitation, we propose a stepwise reasoning enhancement framework for LLMs based on external subgraph generation, termed SGR. The proposed framework dynamically constructs query relevant subgraphs from external knowledge bases and leverages their semantic structure to guide the reasoning process. By performing reasoning in a step by step manner over structured subgraphs, SGR reduces the influence of noisy information and improves reasoning accuracy. Specifically, the framework first generates an external subgraph tailored to the input query, then guides the model to conduct multi step reasoning grounded in the subgraph, and finally integrates multiple reasoning paths to produce the final answer. Experimental results on multiple benchmark datasets demonstrate that SGR consistently outperforms strong baselines, indicating its effectiveness in enhancing the reasoning capabilities of LLMs.