Abstract
Event detection is a classic natural language processing task. However, the constantly emerging new events make supervised methods not applicable to unseen types. Previous zero-shot event detection methods either require predefined event types as heuristic rules or resort to external semantic analyzing tools. To overcome this weakness, we propose an end-to-end framework named Zero-Shot Event Detection Based on Ordered Contrastive Learning and Prompt-Based Prediction (ZEOP). By creatively introducing multiple contrastive samples with ordered similarities, the encoder can learn event representations from both instance-level and class-level, which makes the distinctions between different unseen types more significant. Meanwhile, we utilize the prompt-based prediction to identify trigger words without relying on external resources. Experiments demonstrate that our model detects events more effectively and accurately than state-of-the-art methods.- Anthology ID:
- 2022.findings-naacl.196
- Volume:
- Findings of the Association for Computational Linguistics: NAACL 2022
- Month:
- July
- Year:
- 2022
- Address:
- Seattle, United States
- Editors:
- Marine Carpuat, Marie-Catherine de Marneffe, Ivan Vladimir Meza Ruiz
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 2572–2580
- Language:
- URL:
- https://aclanthology.org/2022.findings-naacl.196
- DOI:
- 10.18653/v1/2022.findings-naacl.196
- Cite (ACL):
- Senhui Zhang, Tao Ji, Wendi Ji, and Xiaoling Wang. 2022. Zero-Shot Event Detection Based on Ordered Contrastive Learning and Prompt-Based Prediction. In Findings of the Association for Computational Linguistics: NAACL 2022, pages 2572–2580, Seattle, United States. Association for Computational Linguistics.
- Cite (Informal):
- Zero-Shot Event Detection Based on Ordered Contrastive Learning and Prompt-Based Prediction (Zhang et al., Findings 2022)
- PDF:
- https://preview.aclanthology.org/proper-vol2-ingestion/2022.findings-naacl.196.pdf
- Code
- kindroach/naacl-zeop