Sparse MLP for Image Recognition: Is Self-Attention Really Necessary?

Open paper
Benchmark Model Rank Results
image-classification-on-imagenetsMLPNet-B (ImageNet-1k)#457Top 1 Accuracy: 83.4%Number of params: 65.9M
image-classification-on-imagenetsMLPNet-S (ImageNet-1k)#496Top 1 Accuracy: 83.1%Number of params: 48.6M
image-classification-on-imagenetsMLPNet-T (ImageNet-1k)#631Top 1 Accuracy: 81.9%Number of params: 24.1M

Rank counts only results with a code link.