Image Classification on ImageNet Base-to-New
75.2H ScoreManiPT
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| ManiPT2026.02 | 75.2 | 78.67 | 72.01 | |
| TAC2026.02 | 74.61 | 78.57 | 71.03 | |
| LLaMP2026.02 | 74.48 | 77.99 | 71.27 | |
| CoPromptfine-tuned=true2023.06 | 74.33 | 77.67 | 71.27 | |
| CoPrompt2026.02 | 74.33 | 77.67 | 71.27 | |
| PromptSRCfine-tuned=true2023.06 | 74.01 | 77.6 | 70.73 | |
| PromptSRC2026.02 | 74.01 | 77.6 | 70.73 | |
| TAP2026.02 | 73.99 | 77.97 | 70.4 | |
| Any-shift PromptingShots=162024.02 | 73.88 | 76.63 | 71.33 | |
| MaPLeShots=162024.02 | 73.47 | 76.66 | 70.54 | |
| MaPLefine-tuned=true2023.06 | 73.47 | 76.66 | 70.54 | |
| MaPLe2026.02 | 73.47 | 76.66 | 70.54 | |
| CoCoOpBackbone=ViT-B/16, Shots=16, Context Length=4, Prompting Strategy=Dynamic (Conditional)2022.03 | 73.1 | 75.98 | 70.43 | |
| CoCoOpBackbone=ViT-B/16, Context length (M)=4, Shots=162023.03 | 73.1 | 75.98 | 70.43 | |
| CoCoOpShots=162024.02 | 73.1 | 75.98 | 70.43 | |
| Co-CoOpfine-tuned=true2023.06 | 73.1 | 75.98 | 70.43 | |
| CoCoOp2026.02 | 73.1 | 75.98 | 70.43 | |
| CoOp# params=8,1922026.06 | 72.98 | 76.65 | 69.65 | |
| KgCoOpBackbone=ViT-B/16, Context length (M)=4, Shots=162023.03 | 72.78 | 75.83 | 69.96 | |
| KgCoOpfine-tuned=true2023.06 | 72.78 | 75.83 | 69.96 | |
| SubT†# params=256,0002026.06 | 72.56 | 76.47 | 69.03 | |
| TEnmodels=(BLIP, BLIP2, CLIP-ViT-B/32)2023.11 | 72.45 | — | — | |
| CoOpBackbone=ViT-B/16, Shots=16, Context Length=4, Prompting Strategy=Static2022.03 | 71.92 | 76.47 | 67.88 | |
| CoOpShots=162024.02 | 71.92 | 76.47 | 67.88 | |
| CoOpfine-tuned=true2023.06 | 71.92 | 76.47 | 67.88 | |
| CoOp2026.02 | 71.92 | 76.47 | 67.88 | |
| CLIP-Adapter# params=8,1922026.06 | 71.55 | 74.53 | 68.81 | |
| ProGradBackbone=ViT-B/16, Context length (M)=4, Shots=162023.03 | 71.46 | 77.02 | 66.66 | |
| ProGradfine-tuned=true2023.06 | 71.46 | 77.02 | 66.66 | |
| CoOpBackbone=ViT-B/16, Context length (M)=4, Shots=162023.03 | 71.02 | 76.46 | 66.31 | |
| TFEnmodels=(BLIP, BLIP2, CLIP-ViT-B/32)2023.11 | 70.79 | — | — | |
| CLIPBackbone=ViT-B/16, Prompting Strategy=Manual2022.03 | 70.22 | 72.43 | 68.14 | |
| CLIPBackbone=ViT-B/162023.03 | 70.22 | 72.43 | 68.14 | |
| CLIPShots=02024.02 | 70.22 | 72.43 | 68.14 | |
| CLIPzero-shot=true2023.06 | 70.22 | 72.43 | 68.14 | |
| CLIP2026.02 | 70.22 | 72.43 | 68.14 | |
| Zero-shot# params=02026.06 | 70.22 | 72.43 | 68.14 | |
| Mean Ensemblemodels=(BLIP, BLIP2, CLIP-ViT-B/32)2023.11 | 69.43 | — | — | |
| CLIP-ViT-B/32mode=Zero-shot2023.11 | 65.69 | — | — | |
| TEnmodels=(BLIP, BLIP2)2023.11 | 60.18 | — | — | |
| TFEnmodels=(BLIP, BLIP2)2023.11 | 57.46 | — | — | |
| Mean Ensemblemodels=(BLIP, BLIP2)2023.11 | 55.38 | — | — | |
| BLIPmode=Zero-shot2023.11 | 53.02 | — | — | |
| BLIP2mode=Zero-shot2023.11 | 37.57 | — | — | |
| BPLShots=162024.02 | — | 70.93 | — |