Do Vision Transformers See Like Convolutional Neural Networks? - 42Papers