BadCLIP: Dual-Embedding Guided Backdoor Attack on Multimodal Contrastive Learning

Documenti analoghi

• BadCLIP++: Stealthy and Persistent Backdoors in Multimodal Contrastive Learning
  di: Liang, Siyuan, et al.
  Pubblicazione: (2026)
• BadCLIP: Trigger-Aware Prompt Learning for Backdoor Attacks on CLIP
  di: Bai, Jiawang, et al.
  Pubblicazione: (2023)
• VL-Trojan: Multimodal Instruction Backdoor Attacks against Autoregressive Visual Language Models
  di: Liang, Jiawei, et al.
  Pubblicazione: (2024)
• Breaking the False Sense of Security in Backdoor Defense through Re-Activation Attack
  di: Zhu, Mingli, et al.
  Pubblicazione: (2024)
• Unlearning Backdoor Threats: Enhancing Backdoor Defense in Multimodal Contrastive Learning via Local Token Unlearning
  di: Liang, Siyuan, et al.
  Pubblicazione: (2024)