TencentFmRD Neural Machine Translation for WMT18

Bojie Hu, Ambyer Han, Shen Huang


Abstract
This paper describes the Neural Machine Translation (NMT) system of TencentFmRD for Chinese↔English news translation tasks of WMT 2018. Our systems are neural machine translation systems trained with our original system TenTrans. TenTrans is an improved NMT system based on Transformer self-attention mechanism. In addition to the basic settings of Transformer training, TenTrans uses multi-model fusion techniques, multiple features reranking, different segmentation models and joint learning. Finally, we adopt some data selection strategies to fine-tune the trained system and achieve a stable performance improvement. Our Chinese→English system achieved the second best BLEU scores and fourth best cased BLEU scores among all WMT18 submitted systems.
Anthology ID:
W18-6413
Volume:
Proceedings of the Third Conference on Machine Translation: Shared Task Papers
Month:
October
Year:
2018
Address:
Belgium, Brussels
Venues:
EMNLP | WMT | WS
SIG:
SIGMT
Publisher:
Association for Computational Linguistics
Note:
Pages:
410–417
Language:
URL:
https://aclanthology.org/W18-6413
DOI:
10.18653/v1/W18-6413
Bibkey:
Cite (ACL):
Bojie Hu, Ambyer Han, and Shen Huang. 2018. TencentFmRD Neural Machine Translation for WMT18. In Proceedings of the Third Conference on Machine Translation: Shared Task Papers, pages 410–417, Belgium, Brussels. Association for Computational Linguistics.
Cite (Informal):
TencentFmRD Neural Machine Translation for WMT18 (Hu et al., 2018)
Copy Citation:
PDF:
https://preview.aclanthology.org/update-css-js/W18-6413.pdf