Mathematical Reasoning on MATH-500 2023
13.6Peak AccuracyPessimism Only
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Pessimism Only2026.04 | 13.6 | 11.9 | 1.7 | |
| RM + Pessimism2026.04 | 12.3 | 10.1 | 2.3 | |
| Reward Model2026.04 | 11.5 | 8.5 | 3 |
| Method | Links | |||
|---|---|---|---|---|
| Pessimism Only2026.04 | 13.6 | 11.9 | 1.7 | |
| RM + Pessimism2026.04 | 12.3 | 10.1 | 2.3 | |
| Reward Model2026.04 | 11.5 | 8.5 | 3 |