Uncertainty Quantification on CNN/DailyMail

0.745Hamming AUC

Structure-Aware Minimum Bayes Risk Decoding

Updated 4d ago

Evaluation Results

Method
Structure-Aware Minimum Bayes Risk Decoding 2026.01	0.745	0.617	0.607
Structure-Aware Minimum Bayes Risk Decoding 2026.01	0.732	0.586	0.603
Structure-Aware Minimum Bayes Risk Decoding 2026.01	0.709	0.585	0.66
Structure-Aware Minimum Bayes Risk Decoding 2026.01	0.672	0.556	0.592
COCOA 2026.01	0.628	0.645	0.512
SAR 2026.01	0.625	0.657	0.51
MSP 2026.01	0.605	0.592	0.517
SAR 2026.01	0.604	0.563	0.526
SAR 2026.01	0.594	0.619	0.574
KLE 2026.01	0.589	0.57	0.502
COCOA 2026.01	0.582	0.576	0.525
COCOA 2026.01	0.577	0.569	0.589
SE 2026.01	0.577	0.548	0.504
SE 2026.01	0.576	0.573	0.475
SE 2026.01	0.572	0.534	0.501
SAR 2026.01	0.57	0.576	0.542
KLE 2026.01	0.57	0.555	0.492
SE 2026.01	0.563	0.556	0.527
COCOA 2026.01	0.563	0.561	0.526
MSP 2026.01	0.546	0.55	0.52
KLE 2026.01	0.539	0.51	0.51
KLE 2026.01	0.534	0.518	0.479
MSP 2026.01	0.526	0.519	0.574
MSP 2026.01	0.523	0.543	0.515
P(True) 2026.01	0.517	0.483	0.539
P(True) 2026.01	0.515	0.514	0.499
P(True) 2026.01	0.479	0.484	0.504
P(True) 2026.01	0.432	0.458	0.474