%0 Conference Proceedings
%A Xu, Ningyu
%A Gui, Tao
%A Ma, Ruotian
%A Zhang, Qi
%A Ye, Jingting
%A Zhang, Menghan
%A Huang, Xuanjing
%+ Department of Linguistic and Cultural Evolution, Max Planck Institute for Evolutionary Anthropology, Max Planck Society
%T Cross-linguistic syntactic difference in multilingual BERT: how good is it and how does it affect transfer? : 
%G eng
%U https://hdl.handle.net/21.11116/0000-000D-0F03-0
%R 10.48550/arXiv.2212.10879
%D 2022
%B The 2022 Conference on Empirical Methods in Natural Language Processing
%Z date of event: 2022-12-07 - 2022-12-11
%C Abu Dhabi
%X Multilingual BERT (mBERT) has demonstrated considerable cross-lingual syntactic ability, whereby it enables effective zero-shot cross-lingual transfer of syntactic knowledge. The transfer is more successful between some languages, but it is not well understood what leads to this variation and whether it fairly reflects difference between languages. In this work, we investigate the distributions of grammatical relations induced from mBERT in the context of 24 typologically different languages. We demonstrate that the distance between the distributions of different languages is highly consistent with the syntactic difference in terms of linguistic formalisms. Such difference learnt via self-supervision plays a crucial role in the zero-shot transfer performance and can be predicted by variation in morphosyntactic properties between languages. These results suggest that mBERT properly encodes languages in a way consistent with linguistic diversity and provide insights into the mechanism of cross-lingual transfer.
%B Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
%P 8073 - 8092
%I Association for Computational Linguistics
%U https://aclanthology.org/2022.emnlp-main.552