Model Selection on 36 Benchmarks Average (test)
78.28Selected-Model AccuracyBestModel
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| BestModelCriterion=BestModel (Oracle)2026.06 | 78.28 | — | — | — | — | — | |
| OSTBCriterion=OSTB2026.06 | 77.03 | 0.807 | 0.68 | 66.7 | 91.7 | 1.249 | |
| Cross-EntropyCriterion=Cross-Entropy2026.06 | 75.25 | 0.711 | 0.623 | 50 | 66.7 | 3.029 | |
| KL DivergenceCriterion=KL Divergence2026.06 | 75.22 | 0.705 | 0.619 | 52.8 | 66.7 | 3.058 | |
| ConfidenceCriterion=Confidence2026.06 | 68.4 | 0.436 | 0.355 | 27.8 | 66.7 | 9.875 | |
| EntropyCriterion=Entropy2026.06 | 68.04 | 0.386 | 0.305 | 19.4 | 63.9 | 10.242 |