Classification on CLIP Evaluation Suite v1 (test)
46.5Food101 AccuracyCLIP-PGS_0.3
Evaluation Results
| Method | Links | ||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CLIP-PGS_0.3Relative Training Time (R.T.T.)=0.6x, Protocol=zero-shot2025.03 | 46.5 | 73.5 | 37.3 | 47.5 | 19.9 | 55.1 | 3.1 | 19.8 | 58.1 | 72.7 | 30.7 | 88.2 | 22.8 | 30.4 | 10.9 | 4.5 | 50.8 | 39.5 | |
| CLIP-PGS_0.5Relative Training Time (R.T.T.)=0.5x, Protocol=zero-shot2025.03 | 42.8 | 62.5 | 35.5 | 45.5 | 17.3 | 50 | 1.9 | 17.4 | 55.7 | 71.8 | 33.2 | 88.2 | 20.5 | 31.8 | 10.1 | 4.7 | 50 | 37.6 | |
| CLIPRelative Training Time (R.T.T.)=1.0x, Protocol=zero-shot2025.03 | 42.3 | 57.7 | 25 | 44.1 | 17 | 50.5 | 1.7 | 16.5 | 53.9 | 73.5 | 26 | 82 | 18.7 | 26.5 | 9.4 | 4.5 | 48 | 35.1 | |
| E-CLIPRelative Training Time (R.T.T.)=0.6x, Protocol=zero-shot2025.03 | 42.1 | 70.7 | 32 | 43.9 | 15.1 | 43.6 | 2.2 | 17 | 55.4 | 73.7 | 28.4 | 85.6 | 22.9 | 30 | 9.6 | 4.7 | 50 | 36.9 | |
| A-CLIPRelative Training Time (R.T.T.)=1.1x, Protocol=zero-shot2025.03 | 41.8 | 61.6 | 27.1 | 46.6 | 16 | 51.1 | 1.3 | 17.1 | 51.2 | 73.5 | 25.7 | 85.8 | 20.5 | 29.1 | 8 | 4.2 | 50.1 | 35.9 | |
| FLIPRelative Training Time (R.T.T.)=0.5x, Protocol=zero-shot2025.03 | 39.9 | 52.8 | 24.5 | 42.8 | 15.9 | 46.6 | 1.4 | 15.9 | 46 | 70.4 | 25.3 | 80.2 | 17 | 25.8 | 5.6 | 4 | 47.1 | 33 |