Model Calibration
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
1.1ECE
150
Jun 25, 2026
0.82Expected Calibration Error
92
Jun 4, 2026
82.4AUROC
84
Feb 26, 2026
0.74ECE
68
May 26, 2026
0.76ECE
68
Apr 10, 2026
0.68ECE
60
Jun 4, 2026
1.22ECE
57
Jun 30, 2026
0.451ECE
56
Feb 26, 2026
0.24ECE
40
Feb 26, 2026
5.413MAE
20
Feb 26, 2026
20.84MAE
20
Feb 26, 2026
1.77Average ECE
13
Feb 26, 2026
0.002ECE
13
Jul 3, 2026
3.3ECE
12
Mar 27, 2026
8.26ECE
11
Feb 26, 2026
8.77ECE
11
Feb 26, 2026
9.46ECE
11
Feb 26, 2026
6.68ECE
11
Feb 26, 2026
9.67ECE
11
Feb 26, 2026
6.23ECE
11
Feb 26, 2026
6.01ECE
11
Feb 26, 2026
0.086ECE
10
Feb 26, 2026
1.52AdaECE
10
Feb 26, 2026
4.515MAE
10
Feb 26, 2026
54.63MAE
10
Feb 26, 2026