Measuring Mathematical Problem Solving With the MATH Dataset
Introduces the MATH benchmark of 12,500 competition-level math problems with step-by-step solutions, spanning algebra to number theory at high-school olympiad difficulty.
- Publisher
- University of California, Berkeley
- Year
- 2021
- Venue
- NeurIPS
- Authors
- 8
- Hosting
- External sourcelicense unknown
Cite
Notes
Only stored in your browser.
Introduces 2 artifacts - 2 evals
TL;DR
Semantic Scholar
This work introduces MATH, a new dataset of 12,500 challenging competition mathematics problems which can be used to teach models to generate answer derivations and explanations and shows that accuracy remains relatively low, even with enormous Transformer models.