What Makes VLMs Robust? Towards Reconciling Robustness and Accuracy in Vision-Language Models

Documenti analoghi

• Assimilation Matters: Model-level Backdoor Detection in Vision-Language Pretrained Models
  di: Wang, Zhongqi, et al.
  Pubblicazione: (2025)
• Contrastive Spectral Rectification: Test-Time Defense towards Zero-shot Adversarial Robustness of CLIP
  di: Nie, Sen, et al.
  Pubblicazione: (2026)
• Trigger without Trace: Towards Stealthy Backdoor Attack on Text-to-Image Diffusion Models
  di: Zhang, Jie, et al.
  Pubblicazione: (2025)
• Dynamic Attention Analysis for Backdoor Detection in Text-to-Image Diffusion Models
  di: Wang, Zhongqi, et al.
  Pubblicazione: (2025)
• T2IShield: Defending Against Backdoors on Text-to-Image Diffusion Models
  di: Wang, Zhongqi, et al.
  Pubblicazione: (2024)