Mathematical Problem Solving on MATH Hard
84.4Ac (alpha=0.10)MV
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| MVModel=Qwen3-8B, ncal=200, m=16, splits=202026.05 | 84.4 | 53.7 | — | 49.8 | 90.1 | 39.3 | |
| J-R1Q14BModel=Qwen3-8B, ncal=200, m=16, splits=202026.05 | 84.3 | 54.7 | 44.1 | 53.6 | 90.1 | 43.6 | |
| R-L8BModel=Qwen3-8B, ncal=200, m=16, splits=202026.05 | 83.7 | 53.6 | 48.9 | 54.8 | 90.5 | 44.8 | |
| J-G26BModel=Qwen3-8B, ncal=200, m=16, splits=202026.05 | 83.7 | 53.8 | 44.8 | 51.5 | 89.6 | 39.7 | |
| PplModel=Qwen3-8B, ncal=200, m=16, splits=202026.05 | 83.5 | 53.8 | 48.8 | 55 | 90.5 | 45.3 | |
| R-Q8BModel=Qwen3-8B, ncal=200, m=16, splits=202026.05 | 83.4 | 53.5 | 48.7 | 55 | 90.5 | 44.3 | |
| Ppl-StdModel=Qwen3-8B, ncal=200, m=16, splits=202026.05 | 83.1 | 54.1 | 47.6 | 54.2 | 90.6 | 43.9 | |
| J-Q30BModel=Qwen3-8B, ncal=200, m=16, splits=202026.05 | 82.7 | 53.1 | 43.9 | 50.6 | 89.6 | 36.7 | |
| OracleModel=Qwen3-8B, ncal=200, m=16, splits=202026.05 | 69.9 | — | 69.9 | 100 | 69.9 | 100 | |
| GreedyModel=Qwen3-8B, ncal=200, m=16, splits=202026.05 | 44 | — | — | 100 | 44 | 100 |