When Vision Transformers Outperform Resnets without Pre-Training or Strong data Augmentations

Where
Lab
Keywords
Vision