Abstract
Recent work has shown the importance of adaptation of broad-coverage contextualised embedding models on the domain of the target task of interest. Current self-supervised adaptation methods are simplistic, as the training signal comes from a small percentage of randomly masked-out tokens. In this paper, we show that careful masking strategies can bridge the knowledge gap of masked language models (MLMs) about the domains more effectively by allocating self-supervision where it is needed. Furthermore, we propose an effective training strategy by adversarially masking out those tokens which are harder to reconstruct by the underlying MLM. The adversarial objective leads to a challenging combinatorial optimisation problem over subsets of tokens, which we tackle efficiently through relaxation to a variational lower-bound and dynamic programming. On six unsupervised domain adaptation tasks involving named entity recognition, our method strongly outperforms the random masking strategy and achieves up to +1.64 F1 score improvements.
Original language | English |
---|---|
Title of host publication | EMNLP 2020, 2020 Conference on Empirical Methods in Natural Language Processing |
Subtitle of host publication | Proceedings of the Conference |
Editors | Trevor Cohn, Yulan He, Yang Liu |
Place of Publication | Stroudsburg PA USA |
Publisher | Association for Computational Linguistics (ACL) |
Pages | 6163-6173 |
Number of pages | 11 |
ISBN (Electronic) | 9781952148606 |
DOIs | |
Publication status | Published - 2020 |
Event | Empirical Methods in Natural Language Processing 2020 - Virtual, Punta Cana, Dominican Republic Duration: 16 Nov 2020 → 20 Nov 2020 https://2020.emnlp.org/ (Website) http://www.aclweb.org/anthology/volumes/2020.emnlp-main/ (Proceedings) https://aclanthology.org/volumes/2020.findings-emnlp/ |
Conference
Conference | Empirical Methods in Natural Language Processing 2020 |
---|---|
Abbreviated title | EMNLP 2020 |
Country/Territory | Dominican Republic |
City | Punta Cana |
Period | 16/11/20 → 20/11/20 |
Internet address |