COMET-QE and Active Learning for Low-Resource Machine Translation
Computation and Language
2022-10-31 v1
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.
Keywords
Cite
@article{arxiv.2210.15696,
title = {COMET-QE and Active Learning for Low-Resource Machine Translation},
author = {Everlyn Asiko Chimoto and Bruce A. Bassett},
journal= {arXiv preprint arXiv:2210.15696},
year = {2022}
}
Comments
Accepted to Findings of EMNLP 2022