UCAM Biomedical translation at WMT19: Transfer learning multi-domain ensembles

06/13/2019

∙

The 2019 WMT Biomedical translation task involved translating Medline abstracts. We approached this using transfer learning to obtain a series of strong neural models on distinct domains, and combining them into multi-domain ensembles. We further experiment with an adaptive language-model ensemble weighting scheme. Our submission achieved the best submitted results on both directions of English-Spanish.

READ FULL TEXT

UCAM Biomedical translation at WMT19: Transfer learning multi-domain ensembles

UFRGS Participation on the WMT Biomedical Translation Shared Task

To Stay or Not to Stay in the Pre-train Basin: Insights on Ensembling in Transfer Learning

English-Catalan Neural Machine Translation in the Biomedical Domain through the cascade approach

Machine Learning on Biomedical Images: Interactive Learning, Transfer Learning, Class Imbalance, and Beyond

Addressing Exposure Bias With Document Minimum Risk Training: Cambridge at the WMT20 Biomedical Translation Task

Ensemble Transfer Learning for Multilingual Coreference Resolution

BactInt: A domain driven transfer learning approach and a corpus for extracting inter-bacterial interactions from biomedical text

UCAM Biomedical translation at WMT19: Transfer learning multi-domain ensembles

Related Research

UFRGS Participation on the WMT Biomedical Translation Shared Task

To Stay or Not to Stay in the Pre-train Basin: Insights on Ensembling in Transfer Learning

English-Catalan Neural Machine Translation in the Biomedical Domain through the cascade approach

Machine Learning on Biomedical Images: Interactive Learning, Transfer Learning, Class Imbalance, and Beyond

Addressing Exposure Bias With Document Minimum Risk Training: Cambridge at the WMT20 Biomedical Translation Task

Ensemble Transfer Learning for Multilingual Coreference Resolution

BactInt: A domain driven transfer learning approach and a corpus for extracting inter-bacterial interactions from biomedical text