List your company

Question Answering on MMLU (F1, AUARC)

4.37F1MAD
2.26922.81463.363.9054Jun 11, 2026

Evaluation results

MethodLinks
MADConfidence Estimator=Sequence Probability, Calibration=Beta2026.064.377.63
MADConfidence Estimator=Self-Report, Calibration=Beta2026.064.3710.05
WSVConfidence Estimator=Self-Report, Calibration=Beta2026.064.239.89
WSVConfidence Estimator=Sequence Probability, Calibration=Beta2026.063.748.95
CGAConfidence Estimator=Self-Report, Calibration=Beta2026.063.583.51
HIDConfidence Estimator=Self-Report, Calibration=Beta2026.063.364.26
MAD-DConfidence Estimator=Sequence Probability, Calibration=Beta2026.063.142.48
MAD-DConfidence Estimator=Self-Report, Calibration=Beta2026.063.146.14
HIDConfidence Estimator=Sequence Probability, Calibration=Beta2026.062.545.76
CGAConfidence Estimator=Sequence Probability, Calibration=Beta2026.062.354.73