Image Classification on UCM
98.2AccuracyViT-g/14
Evaluation results
| Method | Links | |
|---|---|---|
| ViT-g/14N-shot=10, Probe=linear regression2023.05 | 98.2 | |
| SoViT-400m/14N-shot=10, Probe=linear regression2023.05 | 97.7 | |
| ViT-L/16N-shot=10, Probe=linear regression2023.05 | 96.3 | |
| S-CLIPModel=ViT-B/162023.05 | 88.9 | |
| S-CLIPModel=ViT-B/322023.05 | 88.6 | |
| CLIPModel=ViT-B/16, Training Mode=fine-tune2023.05 | 88.5 | |
| CLIPModel=ViT-B/32, Training Mode=fine-tune2023.05 | 87.3 | |
| S-CLIPModel=ResNet-502023.05 | 77.7 | |
| CLIPModel=ResNet-50, Training Mode=fine-tune2023.05 | 76.5 | |
| Domain-Aware Reinforcement Fine-TuningBase Model=QwenVL, Fine-tuning Strategy=Reinforcement Fine-Tuning2026.01 | 28.8 | |
| QwenVLBase Model=QwenVL2026.01 | 8.91 | |
| QwenVL + Domain PromptBase Model=QwenVL, Injection Strategy=Domain Prompt2026.01 | 8.89 | |
| QwenVL + Caption(BLIP)Base Model=QwenVL, Injection Strategy=Caption (BLIP)2026.01 | 8.62 | |
| QwenVL + Caption(MLLM)Base Model=QwenVL, Injection Strategy=Caption (MLLM)2026.01 | 6.42 |