Modeling Discriminative Representations for Out-of-Domain Detection with Supervised Contrastive Learning

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 to learn discriminative semantic features. Traditional cross-entropy loss only focuses on whether a sample is correctly classified, and does not expli...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2021-05
Main Authors: Zeng, Zhiyuan, He, Keqing, Yuanmeng Yan, Liu, Zijun, Wu, Yanan, Xu, Hong, Jiang, Huixing, Xu, Weiran
Format: Article
Language:English
Subjects:
Domains
Learning
Representations
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary: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 to learn discriminative semantic features. Traditional cross-entropy loss only focuses on whether a sample is correctly classified, and does not explicitly distinguish the margins between categories. In this paper, we propose a supervised contrastive learning objective to minimize intra-class variance by pulling together in-domain intents belonging to the same class and maximize inter-class variance by pushing apart samples from different classes. Besides, we employ an adversarial augmentation mechanism to obtain pseudo diverse views of a sample in the latent space. Experiments on two public datasets prove the effectiveness of our method capturing discriminative representations for OOD detection.
ISSN:2331-8422