2023/04/27 by Jiazhao Li, Li, Jiazhao, Yijin Yang +7 · 4 citations
Computer Science · #Advanced Malware Detection Techniques #Adversarial Robustness in Machine Learning #Cryptography and Security (cs.CR) #FOS: Computer and information sciences #Hate Speech and Cyberbullying Detection #Machine Learning (cs.LG)
paper · pdf · doi:10.48550/arxiv.2304.14475
openalex publication_date 2023/04/27 · openalex created_date 2023/05/02 · openalex updated_date 2026/07/28
Textual backdoor attacks pose a practical threat to existing systems, as they can compromise the model by inserting imperceptible triggers into inputs and manipulating labels in the training dataset. With cutting-edge generative models such as GPT-4 pushing rewriting to extraordinary levels, such attacks are becoming even harder to detect. We conduct a comprehensive investigation of the role of black-box generative models as a backdoor attack tool, highlighting the importance of researching relative defense strategies. In this paper, we reveal that the proposed generative model-based attack, BGMAttack, could effectively deceive textual classifiers. Compared with the traditional attack methods, BGMAttack makes the backdoor trigger less conspicuous by leveraging state-of-the-art generative models. Our extensive evaluation of attack effectiveness across five datasets, complemented by three distinct human cognition assessments, reveals that Figure 4 achieves comparable attack performance while maintaining superior stealthiness relative to baseline methods.