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

Model Calibration on MATH

3.3ECE

Verbalized (steered)

0.43219.79139.1558.509Mar 26, 2026
Updated 22d ago

Evaluation Results

MethodLinks
2026.03
3.32.19.3
2026.03
3.72.410.5
2026.03
10.73.511.8
2026.03
14.98.420.2
2026.03
19.525.941.4
2026.03
24.320.129.1
2026.03
35.115.940
2026.03
3616.736.9
2026.03
46.340.653.6
2026.03
68.450.568.5
2026.03
73.557.573.5
2026.03
7558.374.7