NRC Systems for Low Resource German-Upper Sorbian Machine Translation 2020: Transfer Learning with Lexical Modifications

Rebecca Knowles, Samuel Larkin, Darlene Stewart, Patrick Littell


Abstract
We describe the National Research Council of Canada (NRC) neural machine translation systems for the German-Upper Sorbian supervised track of the 2020 shared task on Unsupervised MT and Very Low Resource Supervised MT. Our models are ensembles of Transformer models, built using combinations of BPE-dropout, lexical modifications, and backtranslation.
Anthology ID:
2020.wmt-1.132
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:
1112–1122
Language:
URL:
https://aclanthology.org/2020.wmt-1.132
DOI:
Bibkey:
Cite (ACL):
Rebecca Knowles, Samuel Larkin, Darlene Stewart, and Patrick Littell. 2020. NRC Systems for Low Resource German-Upper Sorbian Machine Translation 2020: Transfer Learning with Lexical Modifications. In Proceedings of the Fifth Conference on Machine Translation, pages 1112–1122, Online. Association for Computational Linguistics.
Cite (Informal):
NRC Systems for Low Resource German-Upper Sorbian Machine Translation 2020: Transfer Learning with Lexical Modifications (Knowles et al., WMT 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.wmt-1.132.pdf
Video:
 https://slideslive.com/38939634