Dynamic Sparse Attention and Mixture-of-Experts for Iterative Visual Classification

Open paper
Benchmark Model Rank Results
image-classification-on-cifar-10DSA-MoE–Percentage correct: 92.3
image-classification-on-cifar-100DSA-MoE–Percentage correct: 69.2
image-classification-on-fashion-mnistDSA-MoE–Accuracy: 92.3
image-classification-on-mnistDSA-MoE–Accuracy: 98.7

Rank counts only results with a code link.