Image Classification on UCF-101
75.2AccuracyC-TPT
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| C-TPTBackbone=ViT-L/142026.01 | 75.2 | 11.1 | — | |
| TPTBackbone=ViT-L/142026.01 | 74.9 | 17.7 | — | |
| SoCBackbone=ViT-L/142026.01 | 74.9 | 6.5 | — | |
| O-TPTBackbone=ViT-L/142026.01 | 74.5 | 8.5 | — | |
| Zero-ShotBackbone=ViT-L/142026.01 | 73.8 | 5.6 | — | |
| SynerNetZero-shot=true2026.01 | 67.7 | — | — | |
| TPTBackbone=CLIP-ViTB/162025.03 | 67.3 | 2.54 | — | |
| TPTBackbone=CLIP-ViTB/162025.03 | 67.3 | 2.54 | — | |
| ZSNLZero-shot=true2026.01 | 67.3 | — | — | |
| C-TPTBackbone=CLIP-ViTB/162025.03 | 65.7 | 2.54 | — | |
| C-TPTBackbone=CLIP-ViTB/162025.03 | 65.7 | 2.54 | — | |
| Robust-adapt-SaLs-CTPTBackbone=CLIP-ViTB/162025.03 | 65.32 | 6.96 | — | |
| Robust-adapt-Penalty-CTPTBackbone=CLIP-ViTB/162025.03 | 65.32 | 2.29 | — | |
| Robust-adapt-SaLs-CTPTBackbone=CLIP-ViTB/162025.03 | 65.32 | 6.96 | — | |
| Robust-adapt-Penalty-CTPTBackbone=CLIP-ViTB/162025.03 | 65.32 | 2.29 | — | |
| Robust-adapt-ZS-CTPTBackbone=CLIP-ViTB/162025.03 | 65.24 | 2.17 | — | |
| Robust-adapt-ZS-CTPTBackbone=CLIP-ViTB/162025.03 | 65.24 | 2.17 | — | |
| Zero ShotBackbone=CLIP-ViTB/162025.03 | 65 | 3.59 | — | |
| Zero ShotBackbone=CLIP-ViTB/162025.03 | 65 | 3.59 | — | |
| O-TPTBackbone=CLIP-ViTB/162025.03 | 64.16 | 2.34 | — | |
| O-TPTBackbone=CLIP-ViTB/162025.03 | 64.16 | 2.34 | — | |
| C-TPTBackbone=CLIP RN502025.03 | 59.7 | 3.83 | — | |
| Robust-adapt-ZS-CTPTBackbone=CLIP RN502025.03 | 59.6 | 4.13 | — | |
| Robust-adapt-SaLs-CTPTBackbone=CLIP RN502025.03 | 59.56 | 4.2 | — | |
| TPTBackbone=CLIP RN502025.03 | 59.5 | 12.4 | — | |
| Robust-adapt-Penalty-CTPTBackbone=CLIP RN502025.03 | 59.42 | 3.93 | — | |
| O-TPTBackbone=CLIP RN502025.03 | 58.84 | 2.6 | — | |
| Zero ShotBackbone=CLIP RN502025.03 | 58.4 | 3.05 | — | |
| TransCLIPZero-shot=true2026.01 | 20.6 | — | — | |
| OpenCLIPZero-shot=true2026.01 | 19.9 | — | — | |
| CLIP+KNNShots=12024.03 | — | — | 47.8 | |
| CLIP+KNNShots=22024.03 | — | — | 58.5 | |
| CLIP+KNNShots=42024.03 | — | — | 64.6 | |
| CLIP+KNNShots=82024.03 | — | — | 68.3 | |
| CLIP+KNNShots=162024.03 | — | — | 70.6 | |
| LLaVA1.5 FinetuningShots=12024.03 | — | — | 35.4 | |
| LLaVA1.5 FinetuningShots=22024.03 | — | — | 38.7 | |
| LLaVA1.5 FinetuningShots=42024.03 | — | — | 58.9 | |
| LLaVA1.5 FinetuningShots=82024.03 | — | — | 66.5 | |
| LLaVA1.5 FinetuningShots=162024.03 | — | — | 74.7 | |
| RAR (LLaVA1.5)Shots=12024.03 | — | — | 57.4 | |
| RAR (LLaVA1.5)Shots=22024.03 | — | — | 66.5 | |
| RAR (LLaVA1.5)Shots=42024.03 | — | — | 69.6 | |
| RAR (LLaVA1.5)Shots=82024.03 | — | — | 74.3 | |
| RAR (LLaVA1.5)Shots=162024.03 | — | — | 75.9 |