Enhancing Logical Reasoning in Large Language Models through Graph-based Synthetic Data
This work addresses a specific bottleneck in LLMs for complex reasoning, offering an incremental improvement in domain-specific applications.
The paper tackled the challenge of improving logical reasoning in Large Language Models (LLMs) by using graph-based synthetic data for fine-tuning, resulting in enhanced performance on inductive and spatial reasoning tasks without degrading other benchmark results.
Despite recent advances in training and prompting strategies for Large Language Models (LLMs), these models continue to face challenges with complex logical reasoning tasks that involve long reasoning chains. In this work, we explore the potential and limitations of using graph-based synthetic reasoning data as training signals to enhance LLMs' reasoning capabilities. Our extensive experiments, conducted on two established natural language reasoning tasks -- inductive reasoning and spatial reasoning -- demonstrate that supervised fine-tuning (SFT) with synthetic graph-based reasoning data effectively enhances LLMs' reasoning performance without compromising their effectiveness on other standard evaluation benchmarks.