Comparing Multilingual NMT Models and Pivoting

Celia Soler Uguet, Fred Bane, Anna Zaretskaya, Tània Blanch Miró


Abstract
Following recent advancements in multilingual machine translation at scale, our team carried out tests to compare the performance of multilingual models (M2M from Facebook and multilingual models from Helsinki-NLP) with a two-step translation process using English as a pivot language. Direct assessment by linguists rated translations produced by pivoting as consistently better than those obtained from multilingual models of similar size, while automated evaluation with COMET suggested relative performance was strongly impacted by domain and language family.
Anthology ID:
2022.eamt-1.26
Volume:
Proceedings of the 23rd Annual Conference of the European Association for Machine Translation
Month:
June
Year:
2022
Address:
Ghent, Belgium
Venue:
EAMT
SIG:
Publisher:
European Association for Machine Translation
Note:
Pages:
231–239
Language:
URL:
https://aclanthology.org/2022.eamt-1.26
DOI:
Bibkey:
Cite (ACL):
Celia Soler Uguet, Fred Bane, Anna Zaretskaya, and Tània Blanch Miró. 2022. Comparing Multilingual NMT Models and Pivoting. In Proceedings of the 23rd Annual Conference of the European Association for Machine Translation, pages 231–239, Ghent, Belgium. European Association for Machine Translation.
Cite (Informal):
Comparing Multilingual NMT Models and Pivoting (Uguet et al., EAMT 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.eamt-1.26.pdf