Mathematical Reasoning on MATH500 (Pass@8/Avg@8)
79.3Avg Score (@8)vOPDtop-k
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| vOPDtop-kModel Setup=Qwen3-4B → Qwen3-4B-Base2026.05 | 79.3 | 93 | 45.3 | |
| vOPDfull-VModel Setup=Qwen3-4B → Qwen3-4B-Base2026.05 | 78.9 | 92.4 | 45.4 | |
| OPDfull-VModel Setup=Qwen3-4B → Qwen3-4B-Base2026.05 | 78.6 | 91.8 | 45.1 | |
| OPDModel Setup=Qwen3-4B → Qwen3-4B-Base2026.05 | 75.2 | 90.8 | 42.9 | |
| OPDtop-kModel Setup=Qwen3-4B → Qwen3-4B-Base2026.05 | 75 | 91.1 | 42.8 | |
| vOPDtop-kModel Setup=Qwen3-1.7B → Qwen3-1.7B-Base2026.05 | 64.9 | 84.8 | 33 | |
| OPDfull-VModel Setup=Qwen3-1.7B → Qwen3-1.7B-Base2026.05 | 64.6 | 85.4 | 33 | |
| vOPDfull-VModel Setup=Olmo-3-7B-Think → Olmo-3-7B-Base2026.05 | 64.4 | 83.2 | 32.9 | |
| vOPDfull-VModel Setup=Qwen3-1.7B → Qwen3-1.7B-Base2026.05 | 64 | 84.6 | 33.1 | |
| vOPDtop-kModel Setup=Olmo-3-7B-Think → Olmo-3-7B-Base2026.05 | 64 | 81 | 33.1 | |
| OPDfull-VModel Setup=Olmo-3-7B-Think → Olmo-3-7B-Base2026.05 | 62.8 | 83.6 | 31.8 | |
| OPDModel Setup=Olmo-3-7B-Think → Olmo-3-7B-Base2026.05 | 61.2 | 78.6 | 29.9 | |
| OPDtop-kModel Setup=Olmo-3-7B-Think → Olmo-3-7B-Base2026.05 | 58.8 | 75.4 | 29.5 | |
| OPDModel Setup=Qwen3-1.7B → Qwen3-1.7B-Base2026.05 | 58.7 | 82.6 | 29.8 | |
| OPDtop-kModel Setup=Qwen3-1.7B → Qwen3-1.7B-Base2026.05 | 58 | 84 | 30.2 | |
| StudentModel Setup=Qwen3-4B → Qwen3-4B-Base2026.05 | 51.1 | 86.4 | 28.4 | |
| StudentModel Setup=Olmo-3-7B-Think → Olmo-3-7B-Base2026.05 | 43.1 | 83.4 | 20.2 | |
| StudentModel Setup=Qwen3-1.7B → Qwen3-1.7B-Base2026.05 | 42.3 | 79.8 | 20.6 |