paper-with-me

홈 › Papers

OutFlip: Generating Out-of-Domain Samples for Unknown Intent Detection with Natural Language Attack

2021-05-12 · DongHyun Choi, Myeong Cheol Shin, EungGyun Kim, Dong Ryeol Shin

Out-of-domain (OOD) input detection is vital in a task-oriented dialogue system since the acceptance of unsupported inputs could lead to an incorrect response of the system. This paper proposes OutFlip, a method to generate out-of-domain samples using only in-domain training dataset automatically. A white-box natural language attack method HotFlip is revised to generate out-of-domain samples instead of adversarial examples. Our evaluation results showed that integrating OutFlip-generated out-of-domain samples into the training dataset could significantly improve an intent classification model's out-of-domain detection performance.

📄 PDF Abstract BibTeX arXiv:2105.05601

Code (1)

kakaoenterprise/OutFlip tf

Tasks

intent-classificationIntent ClassificationIntent Detection

Similar Papers 제목 키워드 기반

OutFlip: Generating Examples for Unknown Intent Detection with Natural Language Attack

2021-08-01 · Findings (ACL) 2021 8 · DongHyun Choi, Myeong Cheol Shin, EungGyun Kim, Dong Ryeol Shin
Intent Detection

Unknown Intent Detection Using Multi-Objective Optimization on Deep Learning Classifiers

2021-09-01 · RANLP 2021 9 · Prerna Prem, Zishan Ahmad, Asif Ekbal, Shubhashis Sengupta 외

Modelling and understanding dialogues in a conversation depends on identifying the user intent from the given text. Unknown or new intent detection is a critical task, as in a realistic scenario a user intent may frequen…

Intent DetectionIntent Discovery

Intent Detection and Discovery from User Logs via Deep Semi-Supervised Contrastive Clustering

2022-07-01 · NAACL 2022 7 · Rajat Kumar, Mayur Patidar, Vaibhav Varshney, Lovekesh Vig 외

Intent Detection is a crucial component of Dialogue Systems wherein the objective is to classify a user utterance into one of multiple pre-defined intents. A pre-requisite for developing an effective intent identifier is…

ClusteringIntent DetectionIntent DiscoveryOpen Intent Discovery+2

Disentangling Confidence Score Distribution for Out-of-Domain Intent Detection with Energy-Based Learning

2022-10-17 · Yanan Wu, Zhiyuan Zeng, Keqing He, Yutao Mou 외

Detecting Out-of-Domain (OOD) or unknown intents from user queries is essential in a task-oriented dialog system. Traditional softmax-based confidence scores are susceptible to the overconfidence issue. In this paper, we…

Intent DetectionOut of Distribution (OOD) Detection

Learning to Classify Open Intent via Soft Labeling and Manifold Mixup

2022-04-16 · Zifeng Cheng, Zhiwei Jiang, Yafeng Yin, Cong Wang 외

Open intent classification is a practical yet challenging task in dialogue systems. Its objective is to accurately classify samples of known intents while at the same time detecting those of open (unknown) intents. Exist…

intent-classificationIntent ClassificationOutlier Detection