NRC Systems for Low Resource German-Upper Sorbian Machine Translation 2020: Transfer Learning with Lexical Modifications
2020-11-01WMT (EMNLP) 2020Unverified0· sign in to hype
Rebecca Knowles, Samuel Larkin, Darlene Stewart, Patrick Littell
Unverified — Be the first to reproduce this paper.
ReproduceAbstract
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.