Mathematical Reasoning on IMO Shortlist
63.8AccuracyNaive Ensemble
Evaluation results
| Method | Links | |
|---|---|---|
| Naive EnsembleBest-of-n samples=502026.04 | 63.8 | |
| FUSEBest-of-n samples=502026.04 | 63.8 | |
| WEAVERBest-of-n samples=502026.04 | 62.1 | |
| Logistic RegressionBest-of-n samples=502026.04 | 60.2 | |
| Oracle Best VerifierBest-of-n samples=50, Backbone=gpt-oss-20b2026.04 | 59.7 | |
| Naive BayesBest-of-n samples=502026.04 | 59.1 | |
| Majority VoteBest-of-n samples=502026.04 | 57.7 | |
| Pass@1Best-of-n samples=502026.04 | 53.3 |