Image Classification on StanfordCars base-to-new
0.8551Base Accuracy2SFS
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| 2SFSVisual Backbone=ViT-L/14, Shots (k)=82025.03 | 0.8551 | 0.8497 | 0.8524 | |
| SkipT2026.05 | 0.8293 | 0.725 | 0.7737 | |
| PromptKDBackbone=ViT-B/162024.03 | 0.828 | 0.8337 | 0.8313 | |
| MMAVisual Backbone=ViT-L/14, Shots (k)=82025.03 | 0.8263 | 0.842 | 0.8341 | |
| MMRL2026.05 | 0.813 | 0.7507 | 0.7806 | |
| MMRL + NeRPRefining Method=NeRP2026.05 | 0.813 | 0.753 | 0.7819 | |
| SCPTShots=16-shots2026.07 | 0.8117 | 0.7566 | 0.7832 | |
| SCPTShots=16-shots2026.07 | 0.8117 | 0.7566 | 0.7832 | |
| TCPShots=16-shots2026.07 | 0.808 | 0.7413 | 0.7732 | |
| TCPShots=16-shots2026.07 | 0.808 | 0.7413 | 0.7732 | |
| MMA2026.05 | 0.785 | 0.731 | 0.757 | |
| MMA + NeRPRefining Method=NeRP2026.05 | 0.785 | 0.7397 | 0.7617 | |
| PromptSRCBackbone=ViT-B/162024.03 | 0.7827 | 0.7497 | 0.7658 | |
| PromptSRC2026.05 | 0.7827 | 0.7497 | 0.7658 | |
| PromptSRC + NeRPRefining Method=NeRP2026.05 | 0.7827 | 0.757 | 0.7696 | |
| CoOpSetup=Prompt-tuning2022.12 | 0.7812 | 0.604 | 0.6813 | |
| CoOpShots=162024.02 | 0.7812 | 0.604 | 0.6813 | |
| CoOpBackbone=ViT-B/162024.03 | 0.7812 | 0.604 | 0.6813 | |
| CoOp2026.05 | 0.7812 | 0.604 | 0.6813 | |
| ProGradBackbone=ViT-B/16, Shots=16, Context Length (M)=42023.03 | 0.7768 | 0.6863 | 0.7288 | |
| ProGradBackbone=ViT-B/16, Shots=16, Context Length (M)=42023.03 | 0.7768 | 0.6863 | 0.7288 | |
| ATPromptShots=16-shots2026.07 | 0.7727 | 0.6654 | 0.715 | |
| ATPromptShots=16-shots2026.07 | 0.7727 | 0.6654 | 0.715 | |
| CoPrompt2026.05 | 0.7697 | 0.744 | 0.7566 | |
| MetaPromptSetup=Meta-prompting2022.12 | 0.7634 | 0.7501 | 0.7548 | |
| CoOpBackbone=ViT-B/16, Shots=16, Context Length (M)=42023.03 | 0.762 | 0.6914 | 0.7249 | |
| CoOpBackbone=ViT-B/16, Shots=16, Context Length (M)=42023.03 | 0.762 | 0.6914 | 0.7249 | |
| CoOpShots=16-shots2026.07 | 0.762 | 0.6914 | 0.7249 | |
| CoOpShots=16-shots2026.07 | 0.762 | 0.6914 | 0.7249 | |
| CLIPVisual Backbone=ViT-L/14, Shots (k)=82025.03 | 0.7456 | 0.8465 | 0.7929 | |
| BPLShots=162024.02 | 0.7323 | — | — | |
| Any-shift PromptingShots=162024.02 | 0.7305 | 0.7583 | 0.7441 | |
| MaPLeSetup=Multi-modal prompt-tuning2022.12 | 0.7294 | 0.74 | 0.7347 | |
| MaPLeShots=162024.02 | 0.7294 | 0.74 | 0.7347 | |
| MaPLeBackbone=ViT-B/162024.03 | 0.7294 | 0.74 | 0.7347 | |
| MaPLe2026.05 | 0.7294 | 0.74 | 0.7347 | |
| KgCoOpBackbone=ViT-B/16, Shots=16, Context Length (M)=42023.03 | 0.7176 | 0.7504 | 0.7336 | |
| KgCoOpBackbone=ViT-B/16, Shots=16, Context Length (M)=42023.03 | 0.7176 | 0.7504 | 0.7336 | |
| KgCoOpShots=16-shots2026.07 | 0.7176 | 0.7504 | 0.7336 | |
| KgCoOpShots=16-shots2026.07 | 0.7176 | 0.7504 | 0.7336 | |
| CoCoOpBackbone=ViT-B/16, Shots=16, Context Length (M)=42023.03 | 0.7049 | 0.7359 | 0.7201 | |
| CoCoOpBackbone=ViT-B/16, Shots=16, Context Length (M)=42023.03 | 0.7049 | 0.7359 | 0.7201 | |
| CoCoOpSetup=Prompt-tuning2022.12 | 0.7049 | 0.7359 | 0.7201 | |
| CoCoOpShots=162024.02 | 0.7049 | 0.7359 | 0.7201 | |
| CoCoOpBackbone=ViT-B/162024.03 | 0.7049 | 0.7359 | 0.7201 | |
| CoCoOp2026.05 | 0.7049 | 0.7359 | 0.7201 | |
| CoCoOp + NeRPRefining Method=NeRP2026.05 | 0.7049 | 0.7413 | 0.7226 | |
| CoCoOpShots=16-shots2026.07 | 0.7049 | 0.7359 | 0.7201 | |
| CoCoOpShots=16-shots2026.07 | 0.7049 | 0.7359 | 0.7201 | |
| ProTextShots=16-shots2026.07 | 0.6454 | 0.7608 | 0.6984 | |
| ProTextShots=16-shots2026.07 | 0.6454 | 0.7608 | 0.6984 | |
| CLIPBackbone=ViT-B/162023.03 | 0.6337 | 0.7489 | 0.6865 | |
| CLIPBackbone=ViT-B/16, Shots=16, Context Length (M)=42023.03 | 0.6337 | 0.7489 | 0.6865 | |
| CLIPSetup=Zero-shot2022.12 | 0.6337 | 0.7489 | 0.6865 | |
| CLIPShots=162024.02 | 0.6337 | 0.7489 | 0.6865 | |
| CLIPBackbone=ViT-B/162024.03 | 0.6337 | 0.7489 | 0.6865 | |
| CLIPShots=16-shots2026.07 | 0.6337 | 0.7489 | 0.6865 | |
| CLIPShots=16-shots2026.07 | 0.6337 | 0.7489 | 0.6865 |