🎆 Optimizing Relevance Maps of Vision Transformers Improves Robustness This code allows to finetune the explainability maps of Vision Transformers to enhance robustness<a target='_blank' rel='noreferrer nofollow' href='https://sun9-41.userapi.com/s/v1/if2/qB0CnGjazK7Vag_tRzyJbc8GXlhZwXdcCrzzX6Ma-Wenwy57_TMc746ATnYXhT45s4Ezf0Dd5eyhKbwe1Cbum50I.jpg?size=1148x830&quality=96&type=album'>. Github: https://github.com/hila-chefer/robustvit Colab: colab.research.google.com/github/…iT.ipynb Paper: https://arxiv.org/abs/2206.01161 Dataset: github.com/Unsuper…ageNet-S @ai_machinelearning_big_data