Visual Mathematical Reasoning on MathVerse (test)
68.1AccuracyQwen3-VL-8B-Instruct + DUPL
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen3-VL-8B-Instruct + DUPLModel Size=8B, Optimization=DUPL2025.10 | 68.1 | |
| Qwen3-VL-4B-Instruct + DUPLModel Size=4B, Optimization=DUPL2025.10 | 65.1 | |
| Qwen3-VL-8B-Instruct + GRPOModel Size=8B, Optimization=GRPO2025.10 | 63.4 | |
| Qwen3-VL-4B-Instruct + GRPOModel Size=4B, Optimization=GRPO2025.10 | 62 | |
| Qwen3-VL-8B-InstructModel Size=8B2025.10 | 58.6 | |
| Qwen3-VL-4B-InstructModel Size=4B2025.10 | 57.6 |