Learning Cross-lingual Representations for Event Coreference Resolution with Multi-view Alignment and Optimal Transport

Duy Phung, Hieu Minh Tran, Minh Van Nguyen, Thien Huu Nguyen


Abstract
We study a new problem of cross-lingual transfer learning for event coreference resolution (ECR) where models trained on data from a source language are adapted for evaluations in different target languages. We introduce the first baseline model for this task based on XLM-RoBERTa, a state-of-the-art multilingual pre-trained language model. We also explore language adversarial neural networks (LANN) that present language discriminators to distinguish texts from the source and target languages to improve the language generalization for ECR. In addition, we introduce two novel mechanisms to further enhance the general representation learning of LANN, featuring: (i) multi-view alignment to penalize cross coreference-label alignment of examples in the source and target languages, and (ii) optimal transport to select close examples in the source and target languages to provide better training signals for the language discriminators. Finally, we perform extensive experiments for cross-lingual ECR from English to Spanish and Chinese to demonstrate the effectiveness of the proposed methods.
Anthology ID:
2021.mrl-1.6
Volume:
Proceedings of the 1st Workshop on Multilingual Representation Learning
Month:
November
Year:
2021
Address:
Punta Cana, Dominican Republic
Editors:
Duygu Ataman, Alexandra Birch, Alexis Conneau, Orhan Firat, Sebastian Ruder, Gozde Gul Sahin
Venue:
MRL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
62–73
Language:
URL:
https://aclanthology.org/2021.mrl-1.6
DOI:
10.18653/v1/2021.mrl-1.6
Bibkey:
Cite (ACL):
Duy Phung, Hieu Minh Tran, Minh Van Nguyen, and Thien Huu Nguyen. 2021. Learning Cross-lingual Representations for Event Coreference Resolution with Multi-view Alignment and Optimal Transport. In Proceedings of the 1st Workshop on Multilingual Representation Learning, pages 62–73, Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
Learning Cross-lingual Representations for Event Coreference Resolution with Multi-view Alignment and Optimal Transport (Phung et al., MRL 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.mrl-1.6.pdf
Video:
 https://aclanthology.org/2021.mrl-1.6.mp4