Nothing Special   »   [go: up one dir, main page]

FERMAT: An Alternative to Accuracy for Numerical Reasoning

Jasivan Sivakumar, Nafise Sadat Moosavi


Abstract
While pre-trained language models achieve impressive performance on various NLP benchmarks, they still struggle with tasks that require numerical reasoning. Recent advances in improving numerical reasoning are mostly achieved using very large language models that contain billions of parameters and are not accessible to everyone. In addition, numerical reasoning is measured using a single score on existing datasets. As a result, we do not have a clear understanding of the strengths and shortcomings of existing models on different numerical reasoning aspects and therefore, potential ways to improve them apart from scaling them up. Inspired by CheckList (Ribeiro et al., 2020), we introduce a multi-view evaluation set for numerical reasoning in English, called FERMAT. Instead of reporting a single score on a whole dataset, FERMAT evaluates models on various key numerical reasoning aspects such as number understanding, mathematical operations, and training dependency. Apart from providing a comprehensive evaluation of models on different numerical reasoning aspects, FERMAT enables a systematic and automated generation of an arbitrarily large training or evaluation set for each aspect. The datasets and codes are publicly available to generate further multi-view data for ulterior tasks and languages.
Anthology ID:
2023.acl-long.838
Volume:
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
July
Year:
2023
Address:
Toronto, Canada
Editors:
Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
15026–15043
Language:
URL:
https://aclanthology.org/2023.acl-long.838
DOI:
10.18653/v1/2023.acl-long.838
Bibkey:
Cite (ACL):
Jasivan Sivakumar and Nafise Sadat Moosavi. 2023. FERMAT: An Alternative to Accuracy for Numerical Reasoning. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 15026–15043, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
FERMAT: An Alternative to Accuracy for Numerical Reasoning (Sivakumar & Moosavi, ACL 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.acl-long.838.pdf
Video:
 https://aclanthology.org/2023.acl-long.838.mp4