Unicoder: A Universal Language Encoder by Pre-training with Multiple Cross-lingual Tasks

Haoyang Huang, Yaobo Liang, Nan Duan, Ming Gong, Linjun Shou, Daxin Jiang, Ming Zhou


Abstract
We present Unicoder, a universal language encoder that is insensitive to different languages. Given an arbitrary NLP task, a model can be trained with Unicoder using training data in one language and directly applied to inputs of the same task in other languages. Comparing to similar efforts such as Multilingual BERT and XLM , three new cross-lingual pre-training tasks are proposed, including cross-lingual word recovery, cross-lingual paraphrase classification and cross-lingual masked language model. These tasks help Unicoder learn the mappings among different languages from more perspectives. We also find that doing fine-tuning on multiple languages together can bring further improvement. Experiments are performed on two tasks: cross-lingual natural language inference (XNLI) and cross-lingual question answering (XQA), where XLM is our baseline. On XNLI, 1.8% averaged accuracy improvement (on 15 languages) is obtained. On XQA, which is a new cross-lingual dataset built by us, 5.5% averaged accuracy improvement (on French and German) is obtained.
Anthology ID:
D19-1252
Volume:
Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)
Month:
November
Year:
2019
Address:
Hong Kong, China
Venues:
EMNLP | IJCNLP
SIG:
SIGDAT
Publisher:
Association for Computational Linguistics
Note:
Pages:
2485–2494
Language:
URL:
https://www.aclweb.org/anthology/D19-1252
DOI:
10.18653/v1/D19-1252
Bib Export formats:
BibTeX MODS XML EndNote
PDF:
http://aclanthology.lst.uni-saarland.de/D19-1252.pdf