Mathematical Reasoning on AIME Out-of-distribution 2024 (test)
14.8AccuracyRLTR
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| RLTRevaluation_samples=642026.02 | 14.8 | 18.9 | 20 | 21.1 | |
| RLVRevaluation_samples=642026.02 | 11.6 | 14.4 | 17.8 | 18.9 | |
| Base modelevaluation_samples=642026.02 | 9.8 | 10 | 13.3 | 16.7 |