Impart: An Imperceptible and Effective Label-Specific Backdoor Attack

arXiv:2403.13017v1 Announce Type: new
Abstract: Backdoor attacks have been shown to impose severe threats to real security-critical scenarios. Although previous works can achieve high attack success rates, they either require access to victim models which may significantly reduce their threats in practice, or perform visually noticeable in stealthiness. Besides, there is still room to improve the attack success rates in the scenario that different poisoned samples may have different target labels (a.k.a., the all-to-all setting). In this study, we propose …

access arxiv attack attacks backdoor backdoor attack backdoor attacks can critical cs.ai cs.cr high may practice real security threats victim