A prompt-based approach to adversarial example generation and robustness enhancement

doi:10.1007/s11704-023-2639-2

	A prompt-based approach to adversarial example generation and robustness enhancement
	Yang, Yuting 1,2; Huang, Pei 3; Cao, Juan 1,2; Li, Jintao 1; Lin, Yun 4; Ma, Feifei 2,5
	2024-08-01
发表期刊	FRONTIERS OF COMPUTER SCIENCE
ISSN	2095-2228
卷号	18 期号:4 页码:12
摘要	Recent years have seen the wide application of natural language processing (NLP) models in crucial areas such as finance, medical treatment, and news media, raising concerns about the model robustness and vulnerabilities. We find that prompt paradigm can probe special robust defects of pre-trained language models. Malicious prompt texts are first constructed for inputs and a pre-trained language model can generate adversarial examples for victim models via maskfilling. Experimental results show that prompt paradigm can efficiently generate more diverse adversarial examples besides synonym substitution. Then, we propose a novel robust training approach based on prompt paradigm which incorporates prompt texts as the alternatives to adversarial examples and enhances robustness under a lightweight minimax-style optimization framework. Experiments on three real-world tasks and two deep neural models show that our approach can significantly improve the robustness of models to resist adversarial attacks.
关键词	robustness adversarial example prompt learning pre-trained language model
DOI	10.1007/s11704-023-2639-2
收录类别	SCI
语种	英语
资助项目	National Key R&D Program of China[2021AAA0140203] ; Zhejiang Provincial Key Research and Development Program of China[2021C01164] ; National Natural Science Foundation of China[61972384] ; National Natural Science Foundation of China[62132020] ; National Natural Science Foundation of China[62203425]
WOS研究方向	Computer Science
WOS类目	Computer Science, Information Systems ; Computer Science, Software Engineering ; Computer Science, Theory & Methods
WOS记录号	WOS:001130418100004
出版者	HIGHER EDUCATION PRESS
引用统计	被引频次：3[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/38410
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Cao, Juan; Ma, Feifei
作者单位	1.Chinese Acad Sci, Inst Comp Technol, Key Lab Intelligent Informat Proc, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Sch Comp Sci & Technol, Beijing 100049, Peoples R China 3.Stanford Univ, Dept Comp Sci, Stanford, CA 94305 USA 4.Natl Univ Singapore, Sch Comp, Singapore 119077, Singapore 5.Chinese Acad Sci, Inst Software, Lab Parallel Software & Computat Sci, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Yang, Yuting,Huang, Pei,Cao, Juan,et al. A prompt-based approach to adversarial example generation and robustness enhancement[J]. FRONTIERS OF COMPUTER SCIENCE,2024,18(4):12.
APA	Yang, Yuting,Huang, Pei,Cao, Juan,Li, Jintao,Lin, Yun,&Ma, Feifei.(2024).A prompt-based approach to adversarial example generation and robustness enhancement.FRONTIERS OF COMPUTER SCIENCE,18(4),12.
MLA	Yang, Yuting,et al."A prompt-based approach to adversarial example generation and robustness enhancement".FRONTIERS OF COMPUTER SCIENCE 18.4(2024):12.