Learning Discriminative Representations for Open Relation Extraction with Instance Ranking and Label Calibration

Shusen Wang, Bin Duan, Yanan Wu, Yajing Xu


Abstract
Open relation extraction is the task to extract relational facts without pre-defined relation types from open-domain corpora. However, since there are some hard or semi-hard instances sharing similar context and entity information but belonging to different underlying relation, current OpenRE methods always cluster them into the same relation type. In this paper, we propose a novel method based on Instance Ranking and Label Calibration strategies (IRLC) to learn discriminative representations for open relation extraction. Due to lacking the original instance label, we provide three surrogate strategies to generate the positive, hard negative, and semi-hard negative instances for the original instance. Instance ranking aims to refine the relational feature space by pushing the hard and semi-hard negative instances apart from the original instance with different margins and pulling the original instance and its positive instance together. To refine the cluster probability distributions of these instances, we introduce a label calibration strategy to model the constraint relationship between instances. Experimental results on two public datasets demonstrate that our proposed method can significantly outperform the previous state-of-the-art methods.
Anthology ID:
2022.findings-naacl.186
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:
2433–2438
Language:
URL:
https://aclanthology.org/2022.findings-naacl.186
DOI:
10.18653/v1/2022.findings-naacl.186
Bibkey:
Cite (ACL):
Shusen Wang, Bin Duan, Yanan Wu, and Yajing Xu. 2022. Learning Discriminative Representations for Open Relation Extraction with Instance Ranking and Label Calibration. In Findings of the Association for Computational Linguistics: NAACL 2022, pages 2433–2438, Seattle, United States. Association for Computational Linguistics.
Cite (Informal):
Learning Discriminative Representations for Open Relation Extraction with Instance Ranking and Label Calibration (Wang et al., Findings 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.findings-naacl.186.pdf
Software:
 2022.findings-naacl.186.software.zip
Video:
 https://aclanthology.org/2022.findings-naacl.186.mp4
Code
 shusenwang/naacl2022-irlc
Data
T-REx