Extending Multilingual BERT to Low-Resource Languages
Zihan Wang, K Karthikeyan, Stephen Mayhew, Dan Roth
Abstract
Multilingual BERT (M-BERT) has been a huge success in both supervised and zero-shot cross-lingual transfer learning. However, this success is focused only on the top 104 languages in Wikipedia it was trained on. In this paper, we propose a simple but effective approach to extend M-BERT (E-MBERT) so it can benefit any new language, and show that our approach aids languages that are already in M-BERT as well. We perform an extensive set of experiments with Named Entity Recognition (NER) on 27 languages, only 16 of which are in M-BERT, and show an average increase of about 6% F 1 on M-BERT languages and 23% F 1 increase on new languages. We release models and code at 1 .
Topics & Concepts
Computer scienceNatural language processingCode (set theory)Artificial intelligenceSet (abstract data type)Resource (disambiguation)Second-generation programming languageZero (linguistics)Programming languageLinguisticsFifth-generation programming languagePhilosophyComputer networkProgramming paradigmTopic ModelingNatural Language Processing TechniquesText and Document Classification Technologies