TY - GEN
T1 - Multilingual Agreement for Multilingual Neural Machine Translation
AU - Yang, Jian
AU - Yin, Yuwei
AU - Ma, Shuming
AU - Huang, Haoyang
AU - Zhang, Dongdong
AU - Li, Zhoujun
AU - Wei, Furu
N1 - Publisher Copyright:
© 2021 Association for Computational Linguistics.
PY - 2021
Y1 - 2021
N2 - Although multilingual neural machine translation (MNMT) enables multiple language translations, the training process is based on independent multilingual objectives. Most multilingual models can not explicitly exploit different language pairs to assist each other, ignoring the relationships among them. In this work, we propose a novel agreement-based method to encourage multilingual agreement among different translation directions, which minimizes the differences among them. We combine the multilingual training objectives with the agreement term by randomly substituting some fragments of the source language with their counterpart translations of auxiliary languages. To examine the effectiveness of our method, we conduct experiments on the multilingual translation task of 10 language pairs. Experimental results show that our method achieves significant improvements over the previous multilingual baselines.
AB - Although multilingual neural machine translation (MNMT) enables multiple language translations, the training process is based on independent multilingual objectives. Most multilingual models can not explicitly exploit different language pairs to assist each other, ignoring the relationships among them. In this work, we propose a novel agreement-based method to encourage multilingual agreement among different translation directions, which minimizes the differences among them. We combine the multilingual training objectives with the agreement term by randomly substituting some fragments of the source language with their counterpart translations of auxiliary languages. To examine the effectiveness of our method, we conduct experiments on the multilingual translation task of 10 language pairs. Experimental results show that our method achieves significant improvements over the previous multilingual baselines.
UR - https://www.scopus.com/pages/publications/85122220134
U2 - 10.18653/v1/2021.acl-short.31
DO - 10.18653/v1/2021.acl-short.31
M3 - 会议稿件
AN - SCOPUS:85122220134
T3 - ACL-IJCNLP 2021 - 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, Proceedings of the Conference
SP - 233
EP - 239
BT - ACL-IJCNLP 2021 - 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, Proceedings of the Conference
PB - Association for Computational Linguistics (ACL)
T2 - Joint Conference of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, ACL-IJCNLP 2021
Y2 - 1 August 2021 through 6 August 2021
ER -