RLMEval: Evaluating Research-Level Neural Theorem Proving

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Poiroux, Auguste, Bosselut, Antoine, Kunčak, Viktor
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866909876142211072
author Poiroux, Auguste
Bosselut, Antoine
Kunčak, Viktor
author_facet Poiroux, Auguste
Bosselut, Antoine
Kunčak, Viktor
contents Despite impressive results on curated benchmarks, the practical impact of large language models (LLMs) on research-level neural theorem proving and proof autoformalization is still limited. We introduce RLMEval, an evaluation suite for these tasks, focusing on research-level mathematics from real-world Lean formalization projects. RLMEval targets the evaluation of neural theorem proving and proof autoformalization on challenging research-level theorems by leveraging real Lean Blueprint formalization projects. Our evaluation of state-of-the-art models on RLMEval, comprising 613 theorems from 6 Lean projects, reveals a significant gap: progress on existing benchmarks does not readily translate to these more realistic settings, with the best model achieving only a 10.3 % pass rate. RLMEval provides a new, challenging benchmark designed to guide and accelerate progress in automated reasoning for formal mathematics.
format Preprint
id arxiv_https___arxiv_org_abs_2510_25427
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle RLMEval: Evaluating Research-Level Neural Theorem Proving
Poiroux, Auguste
Bosselut, Antoine
Kunčak, Viktor
Computation and Language
Artificial Intelligence
Despite impressive results on curated benchmarks, the practical impact of large language models (LLMs) on research-level neural theorem proving and proof autoformalization is still limited. We introduce RLMEval, an evaluation suite for these tasks, focusing on research-level mathematics from real-world Lean formalization projects. RLMEval targets the evaluation of neural theorem proving and proof autoformalization on challenging research-level theorems by leveraging real Lean Blueprint formalization projects. Our evaluation of state-of-the-art models on RLMEval, comprising 613 theorems from 6 Lean projects, reveals a significant gap: progress on existing benchmarks does not readily translate to these more realistic settings, with the best model achieving only a 10.3 % pass rate. RLMEval provides a new, challenging benchmark designed to guide and accelerate progress in automated reasoning for formal mathematics.
title RLMEval: Evaluating Research-Level Neural Theorem Proving
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2510.25427