Knowledge Quiz
Test your understanding of this article
1.According to the article, what is a key finding regarding Vision Transformers (ViTs) and distribution shift robustness compared to ConvNets?
2.What is the primary reason given for Vision Transformers' (ViTs) higher adversarial robustness when successfully adversarially trained, compared to ConvNets?
3.What is PixMix described as in the article?
4.If ConvNets use GELU activation functions, how does their adversarial robustness compare to that of Vision Transformers (ViTs) that are successfully adversarially trained?
