Negation Understanding on CC-Neg (val)
99.7AccuracyCoN-CLIP
Evaluation Results
| Method | Links | |
|---|---|---|
| CoN-CLIPTraining Set=CC-Neg, Training Size (Images/Captions)=188K/376K2026.02 | 99.7 | |
| CLIPGlassesTraining Set=CC-Neg, Training Size (Images/Captions)=188K/376K2026.02 | 96.56 | |
| CLIPGlassesTraining Set=Neg-COCO-R, Training Size (Images/Captions)=5K/22K2026.02 | 93.36 | |
| CoN-CLIPTraining Set=Neg-COCO-R, Training Size (Images/Captions)=5K/22K2026.02 | 65.91 | |
| NegCLIPTraining Set=COCO-caption, Training Size (Images/Captions)=82K/31K2026.02 | 62.63 |