Mathematical Reasoning on AIME25, AMC23, MATH500, Minerva Aggregate
72.16Average Score
GRPO w/ Structure Reward
Updated 2mo ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.03 | 72.16 | 7.65 | |
2026.03 | 71.66 | 7.15 | |
| 71.45 | 6.94 | ||
| 71.12 | 6.61 | ||
2026.03 | 70.38 | 5.87 | |
2026.03 | 70.18 | 5.67 | |
2026.03 | 64.51 | - |