Multiple Choice Question Answering
124.35AACHUQ
Evaluation Results
| Method | Links | |
|---|---|---|
| HUQ2026.04 | 124.35 | |
| MC Dropout2026.04 | 123.12 | |
| Test-time augmentations2026.04 | 117.48 | |
| Single model2026.04 | 116.56 | |
| Entropy and MI2026.04 | 111.05 | |
| Prediction variance2026.04 | 110.93 | |
| Total variance, base2026.04 | 110.32 | |
| Total variance, CAEUses experts=training2026.04 | 100.85 | |
| Total variance, expertsUses experts=inference2026.04 | 66.11 |