Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Confidence Estimation on Global-MMLU (test)

0.78AUROC

Last Answer

0.48880.56440.640.7156May 29, 2026
Updated 2d ago

Evaluation Results

MethodLinks
2026.05
0.780.580.190.16
2026.05
0.750.480.270.27
2026.05
0.740.520.230.21
2026.05
0.730.40.20.17
2026.05
0.720.60.270.24
2026.05
0.720.430.390.42
2026.05
0.720.440.310.31
2026.05
0.710.490.210.16
2026.05
0.680.410.270.25
2026.05
0.680.350.230.21
2026.05
0.640.510.310.26
2026.05
0.640.320.20.17
2026.05
0.50.630.260.26
2026.05
0.50.630.190.06