Can Domains Be Transferred across Languages in Multi-Domain Multilingual Neural Machine Translation?

Thuy-trang Vu, Shahram Khadivi, Xuanli He, Dinh Phung, Gholamreza Haffari

Abstract
Previous works mostly focus on either multilingual or multi-domain aspects of neural machine translation (NMT). This paper investigates whether the domain information can be transferred across languages on the composition of multi-domain and multilingual NMT, particularly for the incomplete data condition where in-domain bitext is missing for some language pairs. Our results in the curated leave-one-domain-out experiments show that multi-domain multilingual (MDML) NMT can boost zero-shot translation performance up to +10 gains on BLEU, as well as aid the generalisation of multi-domain NMT to the missing domain. We also explore strategies for effective integration of multilingual and multi-domain NMT, including language and domain tag combination and auxiliary task training. We find that learning domain-aware representations and adding target-language tags to the encoder leads to effective MDML-NMT.
Anthology ID:
2022.wmt-1.34
Volume:
Proceedings of the Seventh Conference on Machine Translation (WMT)
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates (Hybrid)
Venue:
WMT
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
381–396
Language:
URL:
https://aclanthology.org/2022.wmt-1.34
DOI:
Bibkey:
Cite (ACL):
Thuy-trang Vu, Shahram Khadivi, Xuanli He, Dinh Phung, and Gholamreza Haffari. 2022. Can Domains Be Transferred across Languages in Multi-Domain Multilingual Neural Machine Translation?. In Proceedings of the Seventh Conference on Machine Translation (WMT), pages 381–396, Abu Dhabi, United Arab Emirates (Hybrid). Association for Computational Linguistics.
Cite (Informal):
Can Domains Be Transferred across Languages in Multi-Domain Multilingual Neural Machine Translation? (Vu et al., WMT 2022)
Copy Citation:
PDF:
https://preview.aclanthology.org/nodalida-main-page/2022.wmt-1.34.pdf
PDF Search