Mathematical Reasoning on AIME Aya (Complete)
65.8AccuracyInput-all
Evaluation Results
| Method | Links | |
|---|---|---|
| Input-allk=5, Summarizer=Aya2026.05 | 65.8 | |
| Model-first Greedyk=5, Summarizer=Aya2026.05 | 65.4 | |
| Truth-prediction Greedyk=5, Summarizer=Aya2026.05 | 61.1 | |
| Oracle-surrogate Greedyk=5, Summarizer=Aya2026.05 | 60.7 | |
| MoAk=5, Summarizer=Aya2026.05 | 58 | |
| Conditioned-diversityk=5, Summarizer=Aya2026.05 | 48.3 | |
| GPT5.2-judgek=5, Summarizer=Aya2026.05 | 44.8 | |
| Aya-judgek=5, Summarizer=Aya2026.05 | 43.5 | |
| Top-accuracyk=5, Summarizer=Aya2026.05 | 37.7 | |
| Best-modelk=5, Summarizer=Aya2026.05 | 34.9 |