Proof of Thought : Neurosymbolic Program Synthesis allows Robust and Interpretable Reasoning

TL;DR

Proof of Thought enhances LLM reasoning by integrating formal logic verification with a hybrid human-interpretable representation, improving reliability and transparency in complex, novel domains.

Contribution

Introduces a neurosymbolic framework that converts LLM outputs into formal logic for verification, with a flexible, human-understandable JSON-based domain-specific language.

Findings

Improved performance on StrategyQA benchmark

Effective verification of reasoning steps

Enhanced interpretability of LLM outputs

Abstract

Large Language Models (LLMs) have revolutionized natural language processing, yet they struggle with inconsistent reasoning, particularly in novel domains and complex logical sequences. This research introduces Proof of Thought, a framework that enhances the reliability and transparency of LLM outputs. Our approach bridges LLM-generated ideas with formal logic verification, employing a custom interpreter to convert LLM outputs into First Order Logic constructs for theorem prover scrutiny. Central to our method is an intermediary JSON-based Domain-Specific Language, which by design balances precise logical structures with intuitive human concepts. This hybrid representation enables both rigorous validation and accessible human comprehension of LLM reasoning processes. Key contributions include a robust type system with sort management for enhanced logical integrity, explicit representation of rules for clear distinction between factual and inferential knowledge, and a flexible architecture that allows for easy extension to various domain-specific applications. We demonstrate Proof of Thought's effectiveness through benchmarking on StrategyQA and a novel multimodal reasoning task, showing improved performance in open-ended scenarios. By providing verifiable and interpretable results, our technique addresses critical needs for AI system accountability and sets a foundation for human-in-the-loop oversight in high-stakes domains.