Classification on STL10
98.3Top-1 AccuracyITO
Evaluation results
| Method | Links | |
|---|---|---|
| ITOPre-training Dataset=DataComp-1B, Backbone=ViT-B/16, Training Epochs=10 epochs, Zero-shot protocol=EVA-CLIP2026.03 | 98.3 | |
| ITO sub2Pre-training Dataset=DataComp-1B, Backbone=ViT-B/16, Training Epochs=10 epochs, Zero-shot protocol=EVA-CLIP2026.03 | 98.1 | |
| CLIPPre-training Dataset=DataComp-1B, Backbone=ViT-B/16, Training Epochs=10 epochs, Zero-shot protocol=EVA-CLIP2026.03 | 97.7 | |
| ITO sub2Backbone=ViT-B/16, Pre-training Dataset=CC3M-recap, Evaluation Protocol=Zero-shot2026.03 | 94.4 | |
| ITO sub3Backbone=ViT-B/16, Pre-training Dataset=CC3M-recap, Evaluation Protocol=Zero-shot2026.03 | 94 | |
| FLAIRBackbone=ViT-B/16, Pre-training Dataset=CC3M-recap, Evaluation Protocol=Zero-shot2026.03 | 92.9 | |
| CLIP-PGSImage Encoder=ViT-S/16, p=0.3, Protocol=Zero-shot2025.03 | 88.3 | |
| CLIP-PGSImage Encoder=ViT-B/16, p=0.5, Protocol=Zero-shot2025.03 | 88.2 | |
| CLIP-PGSImage Encoder=ViT-B/16, p=0.3, Protocol=Zero-shot2025.03 | 88.2 | |
| CLIP-PGSImage Encoder=ViT-S/16, p=0.5, Protocol=Zero-shot2025.03 | 86.4 | |
| A-CLIPImage Encoder=ViT-B/16, Protocol=Zero-shot2025.03 | 85.8 | |
| E-CLIPImage Encoder=ViT-B/16, Protocol=Zero-shot2025.03 | 85.6 | |
| CLIPImage Encoder=ViT-B/16, Protocol=Zero-shot2025.03 | 82 | |
| FLIPImage Encoder=ViT-B/16, Protocol=Zero-shot2025.03 | 80.2 |