Are all languages created equal in multilingual BERT?

Shijie Wu, Mark Dredze

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Multilingual BERT (mBERT) (Devlin, 2018) trained on 104 languages has shown surprisingly good cross-lingual performance on several NLP tasks, even without explicit crosslingual signals (Wu and Dredze, 2019; Pires et al., 2019). However, these evaluations have focused on cross-lingual transfer with high-resource languages, covering only a third of the languages covered by mBERT. We explore how mBERT performs on a much wider set of languages, focusing on the quality of representation for low-resource languages, measured by within-language performance. We consider three tasks: Named Entity Recognition (99 languages), Part-of-speech Tagging, and Dependency Parsing (54 languages each). mBERT does better than or comparable to baselines on high resource languages but does much worse for low resource languages. Furthermore, monolingual BERT models for these languages do even worse. Paired with similar languages, the performance gap between monolingual BERT and mBERT can be narrowed. We find that better models for low resource languages require more efficient pretraining techniques or more data.

Original languageEnglish (US)
Title of host publicationACL 2020 - 5th Workshop on Representation Learning for NLP, RepL4NLP 2020, Proceedings of the Workshop
PublisherAssociation for Computational Linguistics (ACL)
Pages120-130
Number of pages11
ISBN (Electronic)9781952148156
StatePublished - 2020
Event5th Workshop on Representation Learning for NLP, RepL4NLP 2020 at the 58th Annual Meeting of the Association for Computational Linguistics, ACL 2020 - Virtual, Online, United States
Duration: Jul 9 2020 → …

Publication series

NameProceedings of the Annual Meeting of the Association for Computational Linguistics
ISSN (Print)0736-587X

Conference

Conference5th Workshop on Representation Learning for NLP, RepL4NLP 2020 at the 58th Annual Meeting of the Association for Computational Linguistics, ACL 2020
Country/TerritoryUnited States
CityVirtual, Online
Period7/9/20 → …

ASJC Scopus subject areas

  • Computer Science Applications
  • Linguistics and Language
  • Language and Linguistics

Fingerprint

Dive into the research topics of 'Are all languages created equal in multilingual BERT?'. Together they form a unique fingerprint.

Cite this