MCTF ($r=20$)

Reported on 4 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision4 results

Image ClassificationonImageNet-1K (with DeiT-S)
GFLOPs· 2024-03-15
2.2
best: 4.6 (Base (DeiT-S))
Multi-criteria Token Fusion with One-step-ahead Attention for Efficient Vision Transformers arXiv:2403.10030
Image ClassificationonImageNet-1K (with DeiT-S)
Top 1 Accuracy· 2024-03-15
79.5
best: 80.1 (MCTF ($r=16$))
Multi-criteria Token Fusion with One-step-ahead Attention for Efficient Vision Transformers arXiv:2403.10030
Image ClassificationonImageNet-1K (with DeiT-T)
GFLOPs· 2024-03-15
0.6
best: 1.2 (Base (DeiT-T))
Multi-criteria Token Fusion with One-step-ahead Attention for Efficient Vision Transformers arXiv:2403.10030
Image ClassificationonImageNet-1K (with DeiT-T)
Top 1 Accuracy· 2024-03-15
71.4
best: 72.9 (dTPS)
Multi-criteria Token Fusion with One-step-ahead Attention for Efficient Vision Transformers arXiv:2403.10030