The DeepMind Chinese–English Document Translation System at WMT2020

Lei Yu, Laurent Sartran, Po-Sen Huang, Wojciech Stokowiec, Domenic Donato, Srivatsan Srinivasan, Alek Andreev, Wang Ling, Sona Mokra, Agustin Dal Lago, Yotam Doron, Susannah Young, Phil Blunsom, Chris Dyer


Abstract
This paper describes the DeepMind submission to the ChineseEnglish constrained data track of the WMT2020 Shared Task on News Translation. The submission employs a noisy channel factorization as the backbone of a document translation system. This approach allows the flexible combination of a number of independent component models which are further augmented with back-translation, distillation, fine-tuning with in-domain data, Monte-Carlo Tree Search decoding, and improved uncertainty estimation. In order to address persistent issues with the premature truncation of long sequences we included specialized length models and sentence segmentation techniques. Our final system provides a 9.9 BLEU points improvement over a baseline Transformer on our test set (newstest 2019).
Anthology ID:
2020.wmt-1.36
Volume:
Proceedings of the Fifth Conference on Machine Translation
Month:
November
Year:
2020
Address:
Online
Venue:
WMT
SIG:
SIGMT
Publisher:
Association for Computational Linguistics
Note:
Pages:
326–337
Language:
URL:
https://aclanthology.org/2020.wmt-1.36
DOI:
Bibkey:
Cite (ACL):
Lei Yu, Laurent Sartran, Po-Sen Huang, Wojciech Stokowiec, Domenic Donato, Srivatsan Srinivasan, Alek Andreev, Wang Ling, Sona Mokra, Agustin Dal Lago, Yotam Doron, Susannah Young, Phil Blunsom, and Chris Dyer. 2020. The DeepMind Chinese–English Document Translation System at WMT2020. In Proceedings of the Fifth Conference on Machine Translation, pages 326–337, Online. Association for Computational Linguistics.
Cite (Informal):
The DeepMind Chinese–English Document Translation System at WMT2020 (Yu et al., WMT 2020)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-script-update/2020.wmt-1.36.pdf
Video:
 https://slideslive.com/38939586