List your company

Mathematical Reasoning on GSM8K (AR%, Risk%)

100AR (%)Always-Act
5.3629.9354.579.07May 18, 2026

Evaluation results

MethodLinks
Always-Actalpha=0.052026.051005
ACIalpha=0.052026.0599.74.8
SAOCPalpha=0.052026.0599.54.7
Fixed-Thresholdalpha=0.052026.0599.54.7
Naive-Tuningalpha=0.052026.0599.54.7
NEX-Confalpha=0.052026.0597.64.4
CSAalpha=0.052026.0571.72.6
CRCalpha=0.052026.0528.94.9
LTTalpha=0.052026.0590.3
ConfFactalpha=0.052026.0590.3