Abstract
Based on massive amounts of data, recent pretrained contextual representation models have made significant strides in advancing a number of different English NLP tasks. However, for other languages, relevant training data may be lacking, while state-of-the-art deep learning methods are known to be data-hungry. In this paper, we present an elegantly simple robust self-learning framework to include unlabeled non-English samples in the fine-tuning process of pretrained multilingual representation models. We leverage a multilingual model’s own predictions on unlabeled non-English data in order to obtain additional information that can be used during further fine-tuning. Compared with original multilingual models and other cross-lingual classification models, we observe significant gains in effectiveness on document and sentiment classification for a range of diverse languages.- Anthology ID:
- D19-1658
- 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
- Editors:
- Kentaro Inui, Jing Jiang, Vincent Ng, Xiaojun Wan
- Venues:
- EMNLP | IJCNLP
- SIG:
- SIGDAT
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 6306–6310
- Language:
- URL:
- https://aclanthology.org/D19-1658
- DOI:
- 10.18653/v1/D19-1658
- Cite (ACL):
- Xin Dong and Gerard de Melo. 2019. A Robust Self-Learning Framework for Cross-Lingual Text Classification. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 6306–6310, Hong Kong, China. Association for Computational Linguistics.
- Cite (Informal):
- A Robust Self-Learning Framework for Cross-Lingual Text Classification (Dong & de Melo, EMNLP-IJCNLP 2019)
- PDF:
- https://preview.aclanthology.org/proper-vol2-ingestion/D19-1658.pdf