AI RESEARCH
LEAP: Supercharging LLMs for Formal Mathematics with Agentic Frameworks
arXiv CS.AI
•
ArXi:2606.03303v1 Announce Type: new Large Language Models (LLMs) exhibit strong informal mathematical reasoning but struggle to generate mechanically verifiable proofs in formal languages like Lean. We present LEAP, an agentic framework that enables general-purpose foundation models to achieve state-of-the-art performance on automated formal theorem proving. LEAP leverages foundation model capabilities, such as informal reasoning, instruction following, and iterative self-refinement.