license: mit
tags:
- math
- theorem-proving
Dataset Description
- Point of Contact: Sean Welleck
miniF2F+informal in Isabelle
MiniF2F is a formal mathematics benchmark (translated across multiple formal systems) consisting of exercise statements from olympiads (AMC, AIME, IMO) as well as high-school and undergraduate maths classes.
This dataset contains formal statements in Isabelle, each paired with an informal statement and an informal proof as described in Draft, Sketch, Prove [Jiang et al 2023].
This dataset is derived from the latest facebookresearch/miniF2F commit as of July 3, 2023. Please see the repository for additional information.
Licensing Information
MIT
Citation Information
This dataset contains Isabelle problem statements from the miniF2F benchmark along with informal statements and proofs.
The initial version of miniF2F is described in Zheng et al ICLR 2022:
@inproceedings{zheng2022miniff,
title={miniF2F: a cross-system benchmark for formal Olympiad-level mathematics},
author={Kunhao Zheng and Jesse Michael Han and Stanislas Polu},
booktitle={International Conference on Learning Representations},
year={2022},
url={https://openreview.net/forum?id=9ZPegFuFTFv}
}
The informal statements and proofs were curated and described in Draft, Sketch, and Prove; Jiang et al ICLR 2023, along with significant fixes and improvements to the initial version of miniF2F:
@inproceedings{jiang2023draft,
title={Draft, Sketch, and Prove: Guiding Formal Theorem Provers with Informal Proofs},
author={Albert Qiaochu Jiang and Sean Welleck and Jin Peng Zhou and Timothee Lacroix and Jiacheng Liu and Wenda Li and Mateja Jamnik and Guillaume Lample and Yuhuai Wu},
booktitle={The Eleventh International Conference on Learning Representations },
year={2023},
url={https://openreview.net/forum?id=SMa9EAovKMC}
}