English

What makes multilingual BERT multilingual?

Computation and Language 2020-10-22 v1 Machine Learning

Abstract

Recently, multilingual BERT works remarkably well on cross-lingual transfer tasks, superior to static non-contextualized word embeddings. In this work, we provide an in-depth experimental study to supplement the existing literature of cross-lingual ability. We compare the cross-lingual ability of non-contextualized and contextualized representation model with the same data. We found that datasize and context window size are crucial factors to the transferability.

Keywords

Cite

@article{arxiv.2010.10938,
  title  = {What makes multilingual BERT multilingual?},
  author = {Chi-Liang Liu and Tsung-Yuan Hsu and Yung-Sung Chuang and Hung-yi Lee},
  journal= {arXiv preprint arXiv:2010.10938},
  year   = {2020}
}

Comments

arXiv admin note: substantial text overlap with arXiv:2004.09205

R2 v1 2026-06-23T19:31:13.381Z