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

Calibration on MathQA

0.344ECE' Gain

S

Updated 1mo ago

Evaluation Results

Method	Links
S 2026.04		0.344	4	0.59	0	1.19	2	-0.2	0
GIRB 2026.04		0.185	0	0.749	1	1.173	2	0.043	4
QAB 2026.04		0.145	0	0.696	0	1.134	0	0.036	0
S-QAB 2026.04		0.059	0	0.828	3	1.144	0	-0.003	0
HS-QAB 2026.04		0.017	0	0.747	0	1.129	0	0.009	0
None 2026.04		0	0	0	0	0	0	0	0
B 2026.04		-0.014	0	0.329	0	1.116	0	0.017	0
S-B 2026.04		-0.061	0	0.281	0	1.038	0	-0.23	0