DiffRate : Differentiable Compression Rate for Efficient Vision Transformers
Open paper
Benchmark
Model
Rank
Results
efficient-vits-on-imagenet-1k-with-deit-s
DiffRate
#10
Top 1 Accuracy: 79.8
GFLOPs: 2.9
efficient-vits-on-imagenet-1k-with-lv-vit-s
DiffRate
#15
Top 1 Accuracy: 82.6
GFLOPs: 3.9
Rank counts only results with a code link.