Tohoku-AIP-NTT at WMT 2020 News Translation Task

Shun Kiyono, Takumi Ito, Ryuto Konno, Makoto Morishita, Jun Suzuki


Abstract
In this paper, we describe the submission of Tohoku-AIP-NTT to the WMT’20 news translation task. We participated in this task in two language pairs and four language directions: English <–> German and English <–> Japanese. Our system consists of techniques such as back-translation and fine-tuning, which are already widely adopted in translation tasks. We attempted to develop new methods for both synthetic data filtering and reranking. However, the methods turned out to be ineffective, and they provided us with no significant improvement over the baseline. We analyze these negative results to provide insights for future studies.
Anthology ID:
2020.wmt-1.12
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:
145–155
Language:
URL:
https://aclanthology.org/2020.wmt-1.12
DOI:
Bibkey:
Cite (ACL):
Shun Kiyono, Takumi Ito, Ryuto Konno, Makoto Morishita, and Jun Suzuki. 2020. Tohoku-AIP-NTT at WMT 2020 News Translation Task. In Proceedings of the Fifth Conference on Machine Translation, pages 145–155, Online. Association for Computational Linguistics.
Cite (Informal):
Tohoku-AIP-NTT at WMT 2020 News Translation Task (Kiyono et al., WMT 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.wmt-1.12.pdf
Video:
 https://slideslive.com/38939545