How Does Context Matter? On the Robustness of Event Detection with Context-Selective Mask Generalization

Jian Liu, Yubo Chen, Kang Liu, Yantao Jia, Zhicheng Sheng

Abstract
Event detection (ED) aims to identify and classify event triggers in texts, which is a crucial subtask of event extraction (EE). Despite many advances in ED, the existing studies are typically centered on improving the overall performance of an ED model, which rarely consider the robustness of an ED model. This paper aims to fill this research gap by stressing the importance of robustness modeling in ED models. We first pinpoint three stark cases demonstrating the brittleness of the existing ED models. After analyzing the underlying reason, we propose a new training mechanism, called context-selective mask generalization for ED, which can effectively mine context-specific patterns for learning and robustify an ED model. The experimental results have confirmed the effectiveness of our model regarding defending against adversarial attacks, exploring unseen predicates, and tackling ambiguity cases. Moreover, a deeper analysis suggests that our approach can learn a complementary predictive bias with most ED models that use full context for feature learning.
Anthology ID:
2020.findings-emnlp.229
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2020
Month:
November
Year:
2020
Address:
Online
Editors:
Trevor Cohn, Yulan He, Yang Liu
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2523–2532
Language:
URL:
https://aclanthology.org/2020.findings-emnlp.229
DOI:
10.18653/v1/2020.findings-emnlp.229
Bibkey:
Cite (ACL):
Jian Liu, Yubo Chen, Kang Liu, Yantao Jia, and Zhicheng Sheng. 2020. How Does Context Matter? On the Robustness of Event Detection with Context-Selective Mask Generalization. In Findings of the Association for Computational Linguistics: EMNLP 2020, pages 2523–2532, Online. Association for Computational Linguistics.
Cite (Informal):
How Does Context Matter? On the Robustness of Event Detection with Context-Selective Mask Generalization (Liu et al., Findings 2020)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-5/2020.findings-emnlp.229.pdf
PDF Search