Effective and Efficient Adversarial Detection for Vision-Language Models via A Single Vector

Documenti analoghi

• Boosting Adversarial Transferability with Spatial Adversarial Alignment
  di: Chen, Zhaoyu, et al.
  Pubblicazione: (2025)
• Improving Adversarial Transferability of Vision-Language Pre-training Models through Collaborative Multimodal Interaction
  di: Fu, Jiyuan, et al.
  Pubblicazione: (2024)
• AdaSteer: Your Aligned LLM is Inherently an Adaptive Jailbreak Defender
  di: Zhao, Weixiang, et al.
  Pubblicazione: (2025)
• RTD-Guard: A Black-Box Textual Adversarial Detection Framework via Replacement Token Detection
  di: Zhu, He, et al.
  Pubblicazione: (2026)
• Query-Efficient Video Adversarial Attack with Stylized Logo
  di: Tang, Duoxun, et al.
  Pubblicazione: (2024)