LeanReasoner: Boosting Complex Logical Reasoning with Lean

arXiv:2403.13312v1 Announce Type: new
Abstract: Large language models (LLMs) often struggle with complex logical reasoning due to logical inconsistencies and the inherent difficulty of such reasoning. We use Lean, a theorem proving framework, to address these challenges. By formalizing logical reasoning problems into theorems within Lean, we can solve them by proving or disproving the corresponding theorems. This method reduces the risk of logical inconsistencies with the help of Lean's symbolic solver. It also enhances our ability to treat complex …

abstract arxiv boosting challenges cs.cl framework language language models large language large language models lean llms reasoning solve struggle them theorem type