Classification on MAQA (test)
63.5AccuracyVar. Credal
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Var. Credal2026.02 | 63.5 | 0.07 | 0.38 | 0.45 | 0.76 | |
| Deep Ensembles2026.02 | 63.1 | 0.78 | 0.31 | 0.25 | 0.73 | |
| MC Dropout2026.02 | 62.8 | 0.77 | 0.3 | 0.24 | 0.72 | |
| Semantic Entropy2026.02 | 62.4 | 0.82 | 0.29 | 0.22 | 0.74 | |
| P(True)2026.02 | 61.8 | 0.76 | 0.27 | 0.2 | 0.71 |