WeChat Neural Machine Translation Systems for WMT21

Xianfeng Zeng, Yijin Liu, Ernan Li, Qiu Ran, Fandong Meng, Peng Li, Jinan Xu, Jie Zhou


Abstract
This paper introduces WeChat AI’s participation in WMT 2021 shared news translation task on English->Chinese, English->Japanese, Japanese->English and English->German. Our systems are based on the Transformer (Vaswani et al., 2017) with several novel and effective variants. In our experiments, we employ data filtering, large-scale synthetic data generation (i.e., back-translation, knowledge distillation, forward-translation, iterative in-domain knowledge transfer), advanced finetuning approaches, and boosted Self-BLEU based model ensemble. Our constrained systems achieve 36.9, 46.9, 27.8 and 31.3 case-sensitive BLEU scores on English->Chinese, English->Japanese, Japanese->English and English->German, respectively. The BLEU scores of English->Chinese, English->Japanese and Japanese->English are the highest among all submissions, and that of English->German is the highest among all constrained submissions.
Anthology ID:
2021.wmt-1.23
Volume:
Proceedings of the Sixth Conference on Machine Translation
Month:
November
Year:
2021
Address:
Online
Venue:
WMT
SIG:
SIGMT
Publisher:
Association for Computational Linguistics
Note:
Pages:
243–254
Language:
URL:
https://aclanthology.org/2021.wmt-1.23
DOI:
Bibkey:
Cite (ACL):
Xianfeng Zeng, Yijin Liu, Ernan Li, Qiu Ran, Fandong Meng, Peng Li, Jinan Xu, and Jie Zhou. 2021. WeChat Neural Machine Translation Systems for WMT21. In Proceedings of the Sixth Conference on Machine Translation, pages 243–254, Online. Association for Computational Linguistics.
Cite (Informal):
WeChat Neural Machine Translation Systems for WMT21 (Zeng et al., WMT 2021)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-script-update/2021.wmt-1.23.pdf