Navigating the Metrics Maze: Reconciling Score Magnitudes and Accuracies

Tom Kocmi, Vilém Zouhar, Christian Federmann, Matt Post


Abstract
Ten years ago a single metric, BLEU, governed progress in machine translation research. For better or worse, there is no such consensus today, and consequently it is difficult for researchers to develop and retain intuitions about metric deltas that drove earlier research and deployment decisions. This paper investigates the “dynamic range” of a number of modern metrics in an effort to provide a collective understanding of the meaning of differences in scores both within and among metrics; in other words, we ask “what point difference x in metric y is required between two systems for humans to notice?”. We conduct our evaluation on a new large dataset, ToShip23, using it to discover deltas at which metrics achieve system-level differences that are meaningful to humans, which we measure by pairwise system accuracy. We additionally show that this method of establishing delta-accuracy is more stable than the standard use of statistical p-values in regards to testset size. Where data size permits, we also explore the effect of metric deltas and accuracy across finer-grained features such as translation direction, domain, and system closeness.
Anthology ID:
2024.acl-long.110
Volume:
Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
August
Year:
2024
Address:
Bangkok, Thailand
Editors:
Lun-Wei Ku, Andre Martins, Vivek Srikumar
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1999–2014
Language:
URL:
https://aclanthology.org/2024.acl-long.110
DOI:
10.18653/v1/2024.acl-long.110
Bibkey:
Cite (ACL):
Tom Kocmi, Vilém Zouhar, Christian Federmann, and Matt Post. 2024. Navigating the Metrics Maze: Reconciling Score Magnitudes and Accuracies. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1999–2014, Bangkok, Thailand. Association for Computational Linguistics.
Cite (Informal):
Navigating the Metrics Maze: Reconciling Score Magnitudes and Accuracies (Kocmi et al., ACL 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.acl-long.110.pdf