Keywords: NLP, mathematics, retrieval, generation, reasoning, theorem proving
TL;DR: NaturalProofs is a multi-domain corpus of mathematical statements and proofs in natural language, along with benchmark models and tasks.
Abstract: Understanding and creating mathematics using natural mathematical language - the mixture of symbolic and natural language used by humans - is a challenging and important problem for driving progress in machine learning. As a step in this direction, we develop NaturalProofs, a multi-domain corpus of mathematical statements and their proofs, written in natural mathematical language. NaturalProofs unifies broad coverage, deep coverage, and low-resource mathematical sources, allowing for evaluating both in-distribution and zero-shot generalization. Using NaturalProofs, we benchmark strong neural methods on mathematical reference retrieval and generation tasks which test a system's ability to determine key results that appear in a proof. Large-scale sequence models show promise compared to classical information retrieval methods, yet their performance and out-of-domain generalization leave substantial room for improvement. NaturalProofs opens many avenues for research on challenging mathematical tasks.
Supplementary Material: zip
URL: https://github.com/wellecks/naturalproofs
Community Implementations: [![CatalyzeX](/images/catalyzex_icon.svg) 3 code implementations](https://www.catalyzex.com/paper/naturalproofs-mathematical-theorem-proving-in/code)
7 Replies
Loading