H. Chefer

Latest

• Discriminative Class Tokens for Text-to-Image Diffusion Models
• Optimizing Relevance Maps of Vision Transformers Improves Robustness