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

被引：0

作者：

Choi, DongHyun ^{[1
,2
]}

Shin, Myeong Cheol ^{[1
]}

Kim, EungGyun ^{[1
]}

Shin, Dong Ryeol ^{[2
]}

机构：

[1] Kakao Enterprise, Seongnam, South Korea

[2] Sungkyunkwan Univ, Seoul, South Korea

来源：

FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, ACL-IJCNLP 2021 | 2021年

关键词：

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

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.

引用

页码：504 / 512

页数：9

共 50 条

[1] Out-of-Domain Detection for Natural Language Understanding in Dialog Systems
Zheng, Yinhe
Chen, Guanyi
Huang, Minlie
IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING, 2020, 28 : 1198 - 1209
[2] GAN-BASED OUT-OF-DOMAIN DETECTION USING BOTH IN-DOMAIN AND OUT-OF-DOMAIN SAMPLES
Liang, Chaojie
Huang, Peijie
Lai, Wenbin
Ruan, Ziheng
2021 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP 2021), 2021, : 7663 - 7667
[3] Enhancing the generalization for Intent Classification and Out-of-Domain Detection in SLU
Shen, Yilin
Hsu, Yen-Chang
Ray, Avik
Jin, Hongxia
59TH ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS AND THE 11TH INTERNATIONAL JOINT CONFERENCE ON NATURAL LANGUAGE PROCESSING, VOL 1 (ACL-IJCNLP 2021), 2021, : 2443 - 2453
[4] Intent Focused Semantic Parsing and Zero-Shot Learning for Out-of-Domain Detection in Spoken Language Understanding
Kumar, Niraj
Baghel, Bhiman Kumar
IEEE ACCESS, 2021, 9 (09): : 165786 - 165794
[5] KNN-Contrastive Learning for Out-of-Domain Intent Classification
Zhou, Yunhua
Liu, Peiju
Qiu, Xipeng
PROCEEDINGS OF THE 60TH ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2022), VOL 1: (LONG PAPERS), 2022, : 5129 - 5141
[6] Enhancing Natural Language Representation with Large-Scale Out-of-Domain Commonsense
Cui, Wanyun
Chen, Xingran
FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2022), 2022, : 1746 - 1756
[7] Using out-of-domain data to improve on-domain language models
Iyer, R
Ostendorf, M
Gish, H
IEEE SIGNAL PROCESSING LETTERS, 1997, 4 (08) : 221 - 223
[8] Handling Cross- and Out-of-Domain Samples in ThaiWord Segmentation
Limkonchotiwat, Peerat
Phatthiyaphaibun, Wannaphong
Sarwar, Raheem
Chuangsuwanich, Ekapol
Nutanong, Sarana
FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, ACL-IJCNLP 2021, 2021, : 1003 - 1016
[9] Practical and Efficient Out-of-Domain Detection with Adversarial Learning
Wang, Bo
Mine, Tsunenori
37TH ANNUAL ACM SYMPOSIUM ON APPLIED COMPUTING, 2022, : 853 - 862
[10] Efficient Out-of-Domain Detection for Sequence to Sequence Models
Vazhentsev, Artem
Tsvigun, Akim
Vashurin, Roman
Petrakov, Sergey
Vasilev, Daniil
Panov, Maxim
Panchenko, Alexander
Shelmanov, Artem
FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, ACL 2023, 2023, : 1430 - 1454

← 1 2 3 4 5 →