This paper considers the development of an AI-based provably-correct mathematical proof tutor. While Large Language Models (LLMs) allow seamless communication in natural language, they are error prone. Theorem provers such as Lean allow for provable-correctness, but these are hard for students to learn. We present a proof-of-concept system (LeanTutor) by combining the complementary strengths of LLMs and theorem provers. 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 the 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.
@article{arxiv.2506.08321,
title = {LeanTutor: Towards a Verified AI Mathematical Proof Tutor},
author = {Manooshree Patel and Rayna Bhattacharyya and Thomas Lu and Arnav Mehta and Niels Voss and Narges Norouzi and Gireeja Ranade},
journal= {arXiv preprint arXiv:2506.08321},
year = {2026}
}
Comments
Comments: Previously this version appeared as arXiv:2601.17473 which was submitted as a new work by accident