Question Answering on BoolQ (F1 and AUARC)
1.4F1 ScoreWSV
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| WSVConfidence Estimator=Self-Report, Calibration=Beta2026.06 | 1.4 | 2.85 | |
| HIDConfidence Estimator=Self-Report, Calibration=Beta2026.06 | 1.01 | 4.57 | |
| WSVConfidence Estimator=Sequence Probability, Calibration=Beta2026.06 | 0.77 | 7.28 | |
| CGAConfidence Estimator=Self-Report, Calibration=Beta2026.06 | 0.62 | 5.46 | |
| HIDConfidence Estimator=Sequence Probability, Calibration=Beta2026.06 | 0.59 | 7.81 | |
| CGAConfidence Estimator=Sequence Probability, Calibration=Beta2026.06 | 0.5 | 8.5 | |
| MAD-DConfidence Estimator=Sequence Probability, Calibration=Beta2026.06 | -0.19 | -2.19 | |
| MAD-DConfidence Estimator=Self-Report, Calibration=Beta2026.06 | -0.19 | 2.99 | |
| MADConfidence Estimator=Sequence Probability, Calibration=Beta2026.06 | -5.53 | -1.04 | |
| MADConfidence Estimator=Self-Report, Calibration=Beta2026.06 | -5.53 | -2.56 |