LeanTutor: A Lean-Verified Tutor for Mathematical Proofs

Manooshree Patel, Rayna Bhattacharyya, Thomas Lu, Arnav Mehta, Niels Voss, Narges Norouzi, Gireeja Ranade

Keywords: AI for Math, AI Tutor, autoformalization, Lean

TL;DR: We present LeanTutor, a Lean-based AI tutoring system that formally verifies student-written math proofs, generates correct next steps, and provides pedagogically motivated guidance to the student user.

Abstract: We present LeanTutor, a Large Language Model (LLM)-based tutoring system for math proofs. LeanTutor interacts with the student in natural language, formally verifies student-written math proofs in Lean, generates correct next steps, and provides the appropriate instructional guidance. LeanTutor is composed of three modules: (i) an autoformalizer/proof-checker, (ii) a next-step generator, and (iii) a natural language feedback generator. To evaluate our system, we introduce PeanoBench, a dataset of 371 Peano Arithmetic proofs in human-written natural language and formal language, derived from the Natural Numbers Game. Each natural language proof step is paired with the corresponding logically equivalent tactic in Lean. The autoformalizer correctly formalizes 57% of tactics in correct proofs and accurately identifies the incorrect step in 30% of incorrect proofs. In generating natural language hints for erroneous proofs, LeanTutor outperforms a simple baseline on accuracy and relevance metrics.

Submission Number: 109