LLM Self-Consistency Certification on AIME 2026
82.5Bonferroni
CITE/W-CITE Framework
Updated 26d ago
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
2026.05 | 82.5 | 94.4 | 94.6 | 95.9 | 79.8 | 25.7 | 113.7 | 113.2 | 57.8 | |
2026.05 | 82.5 | 82.9 | 82.9 | 84.8 | 55.4 | 44 | 176.7 | 126.6 | 91 | |
2026.05 | 79.9 | 81.1 | 81.2 | 83 | 53.8 | 34.8 | 84.2 | 80.7 | 58 | |
2026.05 | 77 | 90.2 | 90.6 | 92.9 | 77.9 | 19.9 | 81.8 | 82 | 45.6 | |
2026.05 | 74 | 76.5 | 77.8 | 79.5 | 51.3 | 25.7 | 62.8 | 60.7 | 40.3 | |
2026.05 | 68.3 | 83.3 | 83.8 | 88 | 74.4 | 14.5 | 49.1 | 55.5 | 32.1 | |
2026.05 | 65.9 | 69.3 | 70.1 | 74.2 | 48.6 | 18.3 | 43.8 | 43.5 | 27.9 | |
2026.05 | 59.7 | 78.1 | 78.5 | 82.7 | 70.5 | 10.2 | 32.4 | 33 | 22.5 | |
2026.05 | 56.3 | 59.6 | 59.9 | 66.6 | 45.9 | 12.6 | 29.3 | 29.6 | 21 | |
2026.05 | 49.9 | 72.4 | 72.7 | 77.8 | 67.1 | 7.1 | 21.8 | 22.8 | 15.9 |