Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Calibration on SQuAD

5.87ECE

Temp. Scaling

3.0921.85540.6259.385Jan 6, 2026Jan 12, 2026Jan 18, 2026Jan 25, 2026Jan 31, 2026Feb 6, 2026Feb 13, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.01
5.870.1769
2026.01
9.110.1818
2026.01
9.660.2007
2026.01
21.310.2216
2026.01
23.930.2357
2026.01
29.560.304
2026.01
29.590.2755
2026.02
30.29-
2026.01
31.340.2816
2026.01
32.150.298
2026.02
35.69-
2026.01
36.030.3208
2026.02
37.92-
2026.01
39.060.3558
2026.01
40.410.3821
2026.01
40.850.3569
2026.01
44.590.3733
2026.01
46.450.4003
2026.01
480.4646
2026.01
48.220.4582
2026.01
52.550.4546
2026.01
55.320.5094
2026.01
56.660.565
2026.01
60.120.555
2026.02
65.13-
2026.02
67.03-
2026.02
69.23-
2026.02
69.84-
2026.02
71.13-
2026.02
71.24-
2026.02
75.37-