Mathematical Reasoning on MATH (test) (AUROC, ECE, Recovery %)
66.8AUROC
Post-solution probe
Updated 16d ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.07 | 66.8 | 23.2 | - | |
| 65.2 | 27.2 | - | ||
2026.07 | 63.4 | 30.6 | - |
Post-solution probe
| Method | Links | |||
|---|---|---|---|---|
2026.07 | 66.8 | 23.2 | - | |
| 65.2 | 27.2 | - | ||
2026.07 | 63.4 | 30.6 | - |