APE-Bench: Evaluating Automated Proof Engineering for Formal Math Libraries

Huajian Xin, Zheng Yuan, Jacques D. Fleuriot, Wenda Li
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:139953-139965, 2026.

Abstract

While frontier formal mathematics systems now routinely develop repository-scale proof engineering artifacts requiring multi-file coordination and semantic correctness beyond compilation, existing evaluation benchmarks remain focused on isolated theorem proving. We introduce Automated Proof Engineering (APE), the first systematic framework for evaluating repository-scale proof engineering through dual verification that validates both syntactic compilation and semantic requirement satisfaction in pinned library environments. We present a complete infrastructure comprising APE-Bench, which automatically extracts proof engineering tasks from real library commit histories, and APE-Harness, a unified execution framework based on task contract abstraction. This contract-based design enables standardized evaluation across diverse formal mathematics tasks and fair systematic comparison of different agent implementations (including our APE-Agent reference scaffold alongside Claude Code and Codex CLI) on identical task specifications. We demonstrate the framework’s effectiveness through comprehensive evaluation. All code, benchmarks, and infrastructure will be open-sourced.

Cite this Paper


BibTeX
@InProceedings{pmlr-v306-xin26c, title = {{APE}-Bench: Evaluating Automated Proof Engineering for Formal Math Libraries}, author = {Xin, Huajian and Yuan, Zheng and Fleuriot, Jacques D. and Li, Wenda}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {139953--139965}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/xin26c/xin26c.pdf}, url = {https://proceedings.mlr.press/v306/xin26c.html}, abstract = {While frontier formal mathematics systems now routinely develop repository-scale proof engineering artifacts requiring multi-file coordination and semantic correctness beyond compilation, existing evaluation benchmarks remain focused on isolated theorem proving. We introduce Automated Proof Engineering (APE), the first systematic framework for evaluating repository-scale proof engineering through dual verification that validates both syntactic compilation and semantic requirement satisfaction in pinned library environments. We present a complete infrastructure comprising APE-Bench, which automatically extracts proof engineering tasks from real library commit histories, and APE-Harness, a unified execution framework based on task contract abstraction. This contract-based design enables standardized evaluation across diverse formal mathematics tasks and fair systematic comparison of different agent implementations (including our APE-Agent reference scaffold alongside Claude Code and Codex CLI) on identical task specifications. We demonstrate the framework’s effectiveness through comprehensive evaluation. All code, benchmarks, and infrastructure will be open-sourced.} }
Endnote
%0 Conference Paper %T APE-Bench: Evaluating Automated Proof Engineering for Formal Math Libraries %A Huajian Xin %A Zheng Yuan %A Jacques D. Fleuriot %A Wenda Li %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-xin26c %I PMLR %P 139953--139965 %U https://proceedings.mlr.press/v306/xin26c.html %V 306 %X While frontier formal mathematics systems now routinely develop repository-scale proof engineering artifacts requiring multi-file coordination and semantic correctness beyond compilation, existing evaluation benchmarks remain focused on isolated theorem proving. We introduce Automated Proof Engineering (APE), the first systematic framework for evaluating repository-scale proof engineering through dual verification that validates both syntactic compilation and semantic requirement satisfaction in pinned library environments. We present a complete infrastructure comprising APE-Bench, which automatically extracts proof engineering tasks from real library commit histories, and APE-Harness, a unified execution framework based on task contract abstraction. This contract-based design enables standardized evaluation across diverse formal mathematics tasks and fair systematic comparison of different agent implementations (including our APE-Agent reference scaffold alongside Claude Code and Codex CLI) on identical task specifications. We demonstrate the framework’s effectiveness through comprehensive evaluation. All code, benchmarks, and infrastructure will be open-sourced.
APA
Xin, H., Yuan, Z., Fleuriot, J.D. & Li, W.. (2026). APE-Bench: Evaluating Automated Proof Engineering for Formal Math Libraries. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:139953-139965 Available from https://proceedings.mlr.press/v306/xin26c.html.

Related Material