GitHub Repository Introduces ProofOfThought: LLM-Based Reasoning System Using Z3 Theorem Proving for Robust, Interpretable AI Reasoning

GitHub - DebarghaG/proofofthought: "Proof of thought: Neurosymbolic program synthesis allows robust and interpretable reasoning" published Sys2Reasoning Workshop NeurIPS 2024

ProofOfThought LLM-based reasoning using Z3 theorem proving. Quick Start from openai import OpenAI from z3dsl.reasoning import ProofOfThought client = OpenAI(api_key="...") pot = ProofOfThought(llm_client=client) result = pot.query("Would Nancy Pelosi publicly denounce abortion?") print(result.answer) # False Batch Evaluation from z3dsl.reasoning import EvaluationPipeline evaluator = EvaluationPipeline(pot, output_dir="results/") result = evaluator.evaluate( dataset="strategyqa_train.json", max...