Image-to-Text Retrieval on ShareGPT4V 10k
95.5R@1CAFT++
Evaluation results
| Method | Links | |
|---|---|---|
| CAFT++Training Data Scale=30M, Training Strategy=Trained on Long-Captions from Scratch, Vision Encoder=ViT-B/16, Zero-shot=true2026.02 | 95.5 | |
| CAFTTraining Data Scale=30M, Training Strategy=Trained on Long-Captions from Scratch, Vision Encoder=ViT-B/16, Zero-shot=true2026.02 | 95.1 | |
| FLAIRTraining Data Scale=30M, Training Strategy=Trained on Long-Captions from Scratch, Vision Encoder=ViT-B/16, Zero-shot=true2026.02 | 90.3 | |
| LoTLIPTraining Data Scale=100M, Training Strategy=Trained on Long-Captions from Scratch, Vision Encoder=ViT-B/16, Zero-shot=true2026.02 | 86.8 | |
| SigLIPTraining Data Scale=10B, Training Strategy=Trained on Short-Captions Only, Vision Encoder=ViT-B/16, Zero-shot=true2026.02 | 83.4 | |
| Long-CLIPTraining Data Scale=400M to 1M, Training Strategy=Finetuned on Long-Captions, Vision Encoder=ViT-B/16, Zero-shot=true2026.02 | 73.1 | |
| OpenCLIPTraining Data Scale=2B, Training Strategy=Trained on Short-Captions Only, Vision Encoder=ViT-B/16, Zero-shot=true2026.02 | 69.6 | |
| ALIGNTraining Data Scale=700M, Training Strategy=Trained on Short-Captions Only, Vision Encoder=ViT-B/16, Zero-shot=true2026.02 | 65.1 | |
| LiTTraining Data Scale=100M, Training Strategy=Trained on Short-Captions Only, Vision Encoder=ViT-B/16, Zero-shot=true2026.02 | 61.4 |