Image-Text Reasoning on MMMU-P
58.78AccuracyGRPO + ReMind
Evaluation Results
| Method | Links | |
|---|---|---|
| GRPO + ReMindBase Model=Qwen3-VL-8B-Instruct2026.06 | 58.78 | |
| DAPO + ReMindBase Model=Qwen3-VL-8B-Instruct2026.06 | 57.89 | |
| DAPOBase Model=Qwen3-VL-8B-Instruct2026.06 | 56.85 | |
| GRPOBase Model=Qwen3-VL-8B-Instruct2026.06 | 55.78 | |
| ExGRPOBase Model=Qwen3-VL-8B-Instruct2026.06 | 55.49 | |
| RLEPBase Model=Qwen3-VL-8B-Instruct2026.06 | 55.38 | |
| RePOBase Model=Qwen3-VL-8B-Instruct2026.06 | 54.95 | |
| Base ModelBase Model=Qwen3-VL-8B-Instruct2026.06 | 51.75 |