IA-RED$^2$: Interpretability-Aware Redundancy Reduction for Vision Transformers
Open paper
Benchmark
Model
Rank
Results
efficient-vits-on-imagenet-1k-with-deit-s
IA-RED$^2$
–
Top 1 Accuracy: 79.1
GFLOPs: 3.2
Rank counts only results with a code link.