Image Classification on ImageNet base-to-novel
77.9Base AccuracyMMRL
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| MMRL2026.05 | 77.9 | 71.3 | 74.45 | |
| MMRL + NeRPRefining Method=NeRP2026.05 | 77.9 | 71.53 | 74.58 | |
| SkipT2026.05 | 77.73 | 70.4 | 73.89 | |
| CoPrompt2026.05 | 77.67 | 71.27 | 74.33 | |
| PromptSRC2026.05 | 77.6 | 70.73 | 74.01 | |
| PromptSRC + NeRPRefining Method=NeRP2026.05 | 77.6 | 71.33 | 74.33 | |
| MMA2026.05 | 77.31 | 71 | 74.02 | |
| MMA + NeRPRefining Method=NeRP2026.05 | 77.31 | 71.73 | 74.42 | |
| MaPLeBackbone=CLIP-ViT-B/16, Training Shots=162026.05 | 77.07 | 70.43 | 73.6 | |
| BiMaPLeBackbone=CLIP-ViT-B/16, Training Shots=162026.05 | 76.73 | 71 | 73.76 | |
| MaPLe2026.05 | 76.66 | 70.54 | 73.47 | |
| CoOp2026.05 | 76.47 | 67.88 | 71.92 | |
| AdaptiveBiMaPLeBackbone=CLIP-ViT-B/16, Training Shots=16, Gate Learning Rate Multiplier=1x2026.05 | 76.23 | 70.2 | 73.09 | |
| CoCoOp2026.05 | 75.98 | 70.43 | 73.1 | |
| CoCoOp + NeRPRefining Method=NeRP2026.05 | 75.98 | 71.37 | 73.6 | |
| AdaptiveBiMaPLeBackbone=CLIP-ViT-B/16, Training Shots=16, Gate Learning Rate Multiplier=50x2026.05 | 75.86 | 69.8 | 72.7 |