Image Reasoning on ZeroBench sub
24.4AccuracyCoPD
Evaluation Results
| Method | Links | |
|---|---|---|
| CoPDTraining Strategy=Co-Evolving Policy Distillation2026.04 | 24.4 | |
| CoPDTraining Strategy=CoPD2026.04 | 23.35 | |
| MOPDTraining Strategy=MOPD2026.04 | 22.75 | |
| Mixed RLVRTraining Strategy=Mixed RLVR2026.04 | 22.6 | |
| Text-ExpertTraining Strategy=Text-Specific Expert2026.04 | 22.01 | |
| Text-ExpertTraining Strategy=Text-Expert2026.04 | 22.01 | |
| Video-ExpertTraining Strategy=Video-Expert2026.04 | 22.01 | |
| Mixed RLVRTraining Strategy=Joint optimization on combined budget2026.04 | 21.86 | |
| OPD T→VTraining Strategy=Static On-policy Policy Distillation (Text to Image)2026.04 | 21.71 | |
| BaseTraining Strategy=Base Model2026.04 | 21.41 | |
| Image-ExpertTraining Strategy=Image-Specific Expert2026.04 | 21.41 | |
| OPD V→TTraining Strategy=Static On-policy Policy Distillation (Image to Text)2026.04 | 21.41 | |
| BaseTraining Strategy=Base2026.04 | 21.41 | |
| Image-ExpertTraining Strategy=Image-Expert2026.04 | 21.41 |