| Joint Token Pruning and Squeezing Towards More Aggressive Compression of Vision Transformers | ✓ Link | 72.9 | 0.8 | dTPS | 2023-04-21 |
| Multi-criteria Token Fusion with One-step-ahead Attention for Efficient Vision Transformers | ✓ Link | 72.9 | 1.0 | MCTF ($r=8$) | 2024-03-15 |
| Multi-criteria Token Fusion with One-step-ahead Attention for Efficient Vision Transformers | ✓ Link | 72.7 | 0.7 | MCTF ($r=16$) | 2024-03-15 |
| Beyond Attentive Tokens: Incorporating Token Importance and Diversity for Efficient Vision Transformers | ✓ Link | 72.3 | 0.8 | BAT | 2022-11-21 |
| Joint Token Pruning and Squeezing Towards More Aggressive Compression of Vision Transformers | ✓ Link | 72.3 | 0.8 | eTPS | 2023-04-21 |
| SPViT: Enabling Faster Vision Transformers via Soft Token Pruning | ✓ Link | 72.2 | 1.0 | SPViT (1.0G) | 2021-12-27 |
| Training data-efficient image transformers & distillation through attention | ✓ Link | 72.2 | 1.2 | Base (DeiT-T) | 2020-12-23 |
| Patch Slimming for Efficient Vision Transformers | | 72.1 | 0.6 | DPS-ViT | 2021-06-05 |
| PPT: Token Pruning and Pooling for Efficient Vision Transformers | ✓ Link | 72.1 | 0.8 | PPT | 2023-10-03 |
| SPViT: Enabling Faster Vision Transformers via Soft Token Pruning | ✓ Link | 72.1 | 0.9 | SPViT (0.9G) | 2021-12-27 |
| Patch Slimming for Efficient Vision Transformers | | 72.0 | 0.7 | PS-ViT | 2021-06-05 |
| Evo-ViT: Slow-Fast Token Evolution for Dynamic Vision Transformer | ✓ Link | 72.0 | 0.8 | EvoViT | 2021-08-03 |
| Learned Thresholds Token Merging and Pruning for Vision Transformers | ✓ Link | 72.0 | 1.0 | LTMP (80%) | 2023-07-20 |
| Token Merging: Your ViT But Faster | ✓ Link | 71.7 | 0.9 | ToMe ($r=8$) | 2022-10-17 |
| Learned Thresholds Token Merging and Pruning for Vision Transformers | ✓ Link | 71.5 | 0.8 | LTMP (60%) | 2023-07-20 |
| Multi-criteria Token Fusion with One-step-ahead Attention for Efficient Vision Transformers | ✓ Link | 71.4 | 0.6 | MCTF ($r=20$) | 2024-03-15 |
| Token Merging: Your ViT But Faster | ✓ Link | 71.4 | 0.8 | ToMe ($r=12$) | 2022-10-17 |
| Token Merging: Your ViT But Faster | ✓ Link | 70.7 | 0.6 | ToMe ($r=16$) | 2022-10-17 |
| Pruning Self-attentions into Convolutional Layers in Single Path | ✓ Link | 70.7 | 1.0 | SPViT | 2021-11-23 |
| Chasing Sparsity in Vision Transformers: An End-to-End Exploration | ✓ Link | 70.1 | 0.9 | S$^2$ViTE | 2021-06-08 |
| Learned Thresholds Token Merging and Pruning for Vision Transformers | ✓ Link | 69.8 | 0.7 | LTMP (45%) | 2023-07-20 |
| Scalable Vision Transformers with Hierarchical Pooling | ✓ Link | 69.6 | 0.6 | HVT-Ti-1 | 2021-03-19 |