PRADA: Practical Black-box Adversarial Attacks against Neural Ranking Models

doi:10.1145/3576923

	PRADA: Practical Black-box Adversarial Attacks against Neural Ranking Models
	Wu, Chen 1,2; Zhang, Ruqing 1,2; Guo, Jiafeng 1,2; De Rijke, Maarten 3; Fan, Yixing 2,4; Cheng, Xueqi 2,4
	2023-10-01
发表期刊	ACM TRANSACTIONS ON INFORMATION SYSTEMS
ISSN	1046-8188
卷号	41 期号:4 页码:27
摘要	Neural ranking models (NRMs) have shown remarkable success in recent years, especially with pre-trained language models. However, deep neural models are notorious for their vulnerability to adversarial examples. Adversarial attacks may become a new type of web spamming technique given our increased reliance on neural information retrieval models. Therefore, it is important to study potential adversarial attacks to identify vulnerabilities of NRMs before they are deployed. In this article, we introduce the Word Substitution Ranking Attack (WSRA) task against NRMs, which aims at promoting a target document in rankings by adding adversarial perturbations to its text. We focus on the decision-based black-box attack setting, where the attackers cannot directly get access to the model information, but can only query the target model to obtain the rank positions of the partial retrieved list. This attack setting is realistic in real-world search engines. We propose a novel Pseudo Relevance-based ADversarial ranking Attack method (PRADA) that learns a surrogate model based on Pseudo Relevance Feedback (PRF) to generate gradients for finding the adversarial perturbations. Experiments on two web search benchmark datasets show that PRADA can outperform existing attack strategies and successfully fool the NRM with small indiscernible perturbations of text.
关键词	Adversarial attack decision-based black-box attack setting neural ranking models
DOI	10.1145/3576923
收录类别	SCI
语种	英语
资助项目	National Natural Science Foundation of China (NSFC)[62006218] ; National Natural Science Foundation of China (NSFC)[61902381] ; Youth Innovation Promotion Association CAS[20144310] ; Youth Innovation Promotion Association CAS[2021100] ; Young Elite Scientist Sponsorship Program by CAST[YESS20200121] ; Lenovo-CAS Joint Lab Youth Scientist Project ; Hybrid Intelligence Center, a 10-year program - Dutch Ministry of Education, Culture and Science through the Netherlands Organisation for Scientific Research
WOS研究方向	Computer Science
WOS类目	Computer Science, Information Systems
WOS记录号	WOS:001068685300008
出版者	ASSOC COMPUTING MACHINERY
引用统计	被引频次：4[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/2XEOYT63/21124
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Guo, Jiafeng
作者单位	1.Inst Comp Technol Acad Sci, Beijing, Peoples R China 2.Univ Chinese Acad Sci, 6 Kexueyuan South Rd, Beijing 100190, Peoples R China 3.Univ Amsterdam, NL-1012WX Amsterdam, Netherlands 4.Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Wu, Chen,Zhang, Ruqing,Guo, Jiafeng,et al. PRADA: Practical Black-box Adversarial Attacks against Neural Ranking Models[J]. ACM TRANSACTIONS ON INFORMATION SYSTEMS,2023,41(4):27.
APA	Wu, Chen,Zhang, Ruqing,Guo, Jiafeng,De Rijke, Maarten,Fan, Yixing,&Cheng, Xueqi.(2023).PRADA: Practical Black-box Adversarial Attacks against Neural Ranking Models.ACM TRANSACTIONS ON INFORMATION SYSTEMS,41(4),27.
MLA	Wu, Chen,et al."PRADA: Practical Black-box Adversarial Attacks against Neural Ranking Models".ACM TRANSACTIONS ON INFORMATION SYSTEMS 41.4(2023):27.