Image-Text Reasoning on MathVerse
61.52AccuracyGRPO + ReMind
Evaluation Results
| Method | Links | |
|---|---|---|
| GRPO + ReMindBase Model=Qwen3-VL-8B-Instruct2026.06 | 61.52 | |
| DAPO + ReMindBase Model=Qwen3-VL-8B-Instruct2026.06 | 60.37 | |
| DAPOBase Model=Qwen3-VL-8B-Instruct2026.06 | 59.55 | |
| GRPOBase Model=Qwen3-VL-8B-Instruct2026.06 | 59.52 | |
| RePOBase Model=Qwen3-VL-8B-Instruct2026.06 | 59.39 | |
| RLEPBase Model=Qwen3-VL-8B-Instruct2026.06 | 58.91 | |
| ExGRPOBase Model=Qwen3-VL-8B-Instruct2026.06 | 56.89 | |
| Base ModelBase Model=Qwen3-VL-8B-Instruct2026.06 | 54.61 |