Image Reasoning Accuracy on MMMU-Pro
55.12AccuracyCoPD
Evaluation Results
| Method | Links | |
|---|---|---|
| CoPDTraining Strategy=CoPD2026.04 | 55.12 | |
| CoPDTraining Strategy=Co-Evolving Policy Distillation2026.04 | 55.1 | |
| OPD T→VTraining Strategy=Static On-policy Policy Distillation (Text to Image)2026.04 | 54.71 | |
| MOPDTraining Strategy=MOPD2026.04 | 54.64 | |
| Mixed RLVRTraining Strategy=Joint optimization on combined budget2026.04 | 54.28 | |
| OPD V→TTraining Strategy=Static On-policy Policy Distillation (Image to Text)2026.04 | 54.09 | |
| Mixed RLVRTraining Strategy=Mixed RLVR2026.04 | 54.06 | |
| Image-ExpertTraining Strategy=Image-Specific Expert2026.04 | 53.89 | |
| Image-ExpertTraining Strategy=Image-Expert2026.04 | 53.89 | |
| Video-ExpertTraining Strategy=Video-Expert2026.04 | 53.08 | |
| Text-ExpertTraining Strategy=Text-Specific Expert2026.04 | 53.05 | |
| Text-ExpertTraining Strategy=Text-Expert2026.04 | 53.05 | |
| BaseTraining Strategy=Base Model2026.04 | 53.03 | |
| BaseTraining Strategy=Base2026.04 | 53.03 |