Spot the Difference on CGD (val)
159CIDErONION
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| ONIONModel=Qwen-VL2026.06 | 159 | 46.1 | 50.1 | |
| SFTModel=Qwen-VL2026.06 | 158.3 | 46.1 | 99.5 | |
| BYEModel=Qwen-VL2026.06 | 155.8 | 46.3 | 99 | |
| BYORn-FModel=Qwen-VL2026.06 | 149.4 | 45.4 | 0.8 | |
| BYORnModel=Qwen-VL2026.06 | 148.5 | 45.6 | 0.8 | |
| ONIONModel=LLaVA2026.06 | 146.2 | 45.3 | 50.4 | |
| SFTModel=LLaVA2026.06 | 144.9 | 45.4 | 100 | |
| BYEModel=LLaVA2026.06 | 144.1 | 45 | 100 | |
| BYORnModel=LLaVA2026.06 | 134.9 | 44.3 | 1.1 | |
| BYORn-FModel=LLaVA2026.06 | 129.8 | 43.8 | 1 |