Improving Generalizability and Undetectability for Targeted Adversarial Attacks on Multimodal Pre-trained Models

Ähnliche Einträge

• Defending Multimodal Backdoored Models by Repulsive Visual Prompt Tuning
  von: Zhang, Zhifang, et al.
  Veröffentlicht: (2024)
• Candidate Pseudolabel Learning: Enhancing Vision-Language Models by Prompt Tuning with Unlabeled Data
  von: Zhang, Jiahan, et al.
  Veröffentlicht: (2024)
• Unleashing the Power of Pre-trained Encoders for Universal Adversarial Attack Detection
  von: Zhang, Yinghe, et al.
  Veröffentlicht: (2025)
• VLATTACK: Multimodal Adversarial Attacks on Vision-Language Tasks via Pre-trained Models
  von: Yin, Ziyi, et al.
  Veröffentlicht: (2023)
• Enhancing Adversarial Attacks: The Similar Target Method
  von: Zhang, Shuo, et al.
  Veröffentlicht: (2023)