Can Monolingual Pretrained Models Help Cross-Lingual Classification?

Zewen Chi, Li Dong, Furu Wei, Xianling Mao, Heyan Huang


Abstract
Multilingual pretrained language models (such as multilingual BERT) have achieved impressive results for cross-lingual transfer. However, due to the constant model capacity, multilingual pre-training usually lags behind the monolingual competitors. In this work, we present two approaches to improve zero-shot cross-lingual classification, by transferring the knowledge from monolingual pretrained models to multilingual ones. Experimental results on two cross-lingual classification benchmarks show that our methods outperform vanilla multilingual fine-tuning.
Anthology ID:
2020.aacl-main.2
Volume:
Proceedings of the 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing
Month:
December
Year:
2020
Address:
Suzhou, China
Venue:
AACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
12–17
Language:
URL:
https://aclanthology.org/2020.aacl-main.2
DOI:
Bibkey:
Cite (ACL):
Zewen Chi, Li Dong, Furu Wei, Xianling Mao, and Heyan Huang. 2020. Can Monolingual Pretrained Models Help Cross-Lingual Classification?. In Proceedings of the 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing, pages 12–17, Suzhou, China. Association for Computational Linguistics.
Cite (Informal):
Can Monolingual Pretrained Models Help Cross-Lingual Classification? (Chi et al., AACL 2020)
Copy Citation:
PDF:
https://preview.aclanthology.org/auto-file-uploads/2020.aacl-main.2.pdf