☆ 3.8 Proceedings Paper

Self-Ensembling Vision Transformer (SEViT) for Robust Medical Image Classification

MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION, MICCAI 2022, PT III (2022)

Related references

Note: Only part of the references are listed.

Proceedings Paper Computer Science, Artificial Intelligence

On the Robustness of Vision Transformers to Adversarial Examples

Kaleel Mahmood et al.

Summary: This study investigates the robustness of Vision Transformers to adversarial examples, finding that these examples do not readily transfer between CNNs and Transformers. The researchers introduce a new attack called the self-attention blended gradient attack and analyze the security of a simple ensemble defense of CNNs and Transformers.

2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021) (2021)

Add to Collection

Proceedings Paper Computer Science, Artificial Intelligence

Understanding Robustness of Transformers for Image Classification

Srinadh Bhojanapalli et al.

Summary: This study compares ViT models and ResNet baselines, finding that when pre-trained with sufficient data, ViT models are at least as robust as ResNet counterparts under a broad range of perturbations. The study also reveals that Transformers are robust to the removal of almost any single layer, showcasing their importance in classification despite high correlations among activations from later layers.

2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021) (2021)

Add to Collection

Article Computer Science, Information Systems