Abstract
Active learning aims to deliver maximum benefit when resources are scarce. We use COMET-QE, a reference-free evaluation metric, to select sentences for low-resource neural machine translation. Using Swahili, Kinyarwanda and Spanish for our experiments, we show that COMET-QE significantly outperforms two variants of Round Trip Translation Likelihood (RTTL) and random sentence selection by up to 5 BLEU points for 20k sentences selected by Active Learning on a 30k baseline. This suggests that COMET-QE is a powerful tool for sentence selection in the very low-resource limit.- Anthology ID:
- 2022.findings-emnlp.348
- Volume:
- Findings of the Association for Computational Linguistics: EMNLP 2022
- Month:
- December
- Year:
- 2022
- Address:
- Abu Dhabi, United Arab Emirates
- Editors:
- Yoav Goldberg, Zornitsa Kozareva, Yue Zhang
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 4735–4740
- Language:
- URL:
- https://aclanthology.org/2022.findings-emnlp.348
- DOI:
- 10.18653/v1/2022.findings-emnlp.348
- Cite (ACL):
- Everlyn Chimoto and Bruce Bassett. 2022. COMET-QE and Active Learning for Low-Resource Machine Translation. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 4735–4740, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
- Cite (Informal):
- COMET-QE and Active Learning for Low-Resource Machine Translation (Chimoto & Bassett, Findings 2022)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-4/2022.findings-emnlp.348.pdf