Revisit Overconfidence for OOD Detection: Reassigned Contrastive Learning with Adaptive Class-dependent Threshold
Yanan Wu, Keqing He, Yuanmeng Yan, QiXiang Gao, Zhiyuan Zeng, Fujia Zheng, Lulu Zhao, Huixing Jiang
Abstract
Detecting Out-of-Domain (OOD) or unknown intents from user queries is essential in a task-oriented dialog system. A key challenge of OOD detection is the overconfidence of neural models. In this paper, we comprehensively analyze overconfidence and classify it into two perspectives: over-confident OOD and in-domain (IND). Then according to intrinsic reasons, we respectively propose a novel reassigned contrastive learning (RCL) to discriminate IND intents for over-confident OOD and an adaptive class-dependent local threshold mechanism to separate similar IND and OOD intents for over-confident IND. Experiments and analyses show the effectiveness of our proposed method for both aspects of overconfidence issues.
BibTeX
@inproceedings{wu-etal-2022-revisit,
title = "Revisit Overconfidence for {OOD} Detection: Reassigned Contrastive Learning with Adaptive Class-dependent Threshold",
author = "Wu, Yanan and
He, Keqing and
Yan, Yuanmeng and
Gao, QiXiang and
Zeng, Zhiyuan and
Zheng, Fujia and
Zhao, Lulu and
Jiang, Huixing and
Wu, Wei and
Xu, Weiran",
editor = "Carpuat, Marine and
de Marneffe, Marie-Catherine and
Meza Ruiz, Ivan Vladimir",
booktitle = "Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies",
month = jul,
year = "2022",
address = "Seattle, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.naacl-main.307/",
doi = "10.18653/v1/2022.naacl-main.307",
pages = "4165--4179"
}