Abstract
We benchmark the performance of segment-level metrics submitted to WMT 2023 using the ACES Challenge Set (Amrhein et al., 2022). The challenge set consists of 36K examples representing challenges from 68 phenomena and covering 146 language pairs. The phenomena range from simple perturbations at the word/character level to more complex errors based on discourse and real-world knowledge. For each metric, we provide a detailed profile of performance over a range of error categories as well as an overall ACES-Score for quick comparison. We also measure the incremental performance of the metrics submitted to both WMT 2023 and 2022. We find that 1) there is no clear winner among the metrics submitted to WMT 2023, and 2) performance change between the 2023 and 2022 versions of the metrics is highly variable. Our recommendations are similar to those from WMT 2022. Metric developers should focus on: building ensembles of metrics from different design families, developing metrics that pay more attention to the source and rely less on surface-level overlap, and carefully determining the influence of multilingual embeddings on MT evaluation.- Anthology ID:
- 2023.wmt-1.57
- Volume:
- Proceedings of the Eighth Conference on Machine Translation
- Month:
- December
- Year:
- 2023
- Address:
- Singapore
- Editors:
- Philipp Koehn, Barry Haddow, Tom Kocmi, Christof Monz
- Venue:
- WMT
- SIG:
- SIGMT
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 695–712
- Language:
- URL:
- https://aclanthology.org/2023.wmt-1.57
- DOI:
- 10.18653/v1/2023.wmt-1.57
- Cite (ACL):
- Chantal Amrhein, Nikita Moghe, and Liane Guillou. 2023. ACES: Translation Accuracy Challenge Sets at WMT 2023. In Proceedings of the Eighth Conference on Machine Translation, pages 695–712, Singapore. Association for Computational Linguistics.
- Cite (Informal):
- ACES: Translation Accuracy Challenge Sets at WMT 2023 (Amrhein et al., WMT 2023)
- PDF:
- https://preview.aclanthology.org/emnlp22-frontmatter/2023.wmt-1.57.pdf