Question Answering on MAQA
0.635AccuracyVariational Credal
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Variational Credal2026.02 | 0.635 | 0.07 | 0.38 | 0.45 | 0.74 | |
| Deep Ensembles2026.02 | 0.631 | 0.78 | 0.31 | 0.25 | 0.65 | |
| CBDL2026.02 | 0.63 | 0.6 | 0.33 | 0.31 | 0.68 | |
| MC Dropout2026.02 | 0.628 | 0.77 | 0.3 | 0.24 | 0.64 | |
| CreINNs2026.02 | 0.628 | 0.63 | 0.32 | 0.28 | 0.66 | |
| Semantic Entropy2026.02 | 0.624 | 0.82 | 0.29 | 0.22 | 0.63 | |
| P(True)2026.02 | 0.618 | 0.76 | 0.27 | 0.2 | 0.61 |