Skip to content
conftrace_
2020 EMNLP EMNLP 2020

FJWU participation for the WMT20 Biomedical Translation Task

Abstract

AbstractThis paper reports system descriptions for FJWU-NRPU team for participation in the WMT20 Biomedical shared translation task. We focused our submission on exploring the effects of adding in-domain corpora extracted from various out-of-domain sources. Systems were built for French to English using in-domain corpora through fine tuning and selective data training. We further explored BERT based models specifically with focus on effect of domain adaptive subword units.

Authors

Sumbal Naz, Sadaf Abdul Rauf, Noor-e- Hira, Sami Ul Haq

Topics

Machine Learning > Application Areas > Domain Adaptation

Keywords

domain adaptation neural machine translation subword unit biomedical translation bert-based model in-domain corpus
Download PDF

More from EMNLP 2020

All EMNLP 2020 papers →
Pretrained Language Models and Backtranslation for English-Basque Biomedical Neural Machine Translation
Lite Training Strategies for Portuguese-English and English-Portuguese Translation
The ADAPT’s Submissions to the WMT20 Biomedical Translation Task
Huawei’s Submissions to the WMT20 Biomedical Translation Task
Addressing Exposure Bias With Document Minimum Risk Training: Cambridge at the WMT20 Biomedical Translation Task
UoS Participation in the WMT20 Translation of Biomedical Abstracts