List your company

Verification on GSM8K

70.17AUROCCRV
51.668456.471761.27566.0783Oct 10, 2025

Evaluation results

MethodLinks
CRVParadigm=White-Box2025.1070.1714.379.61
EnergyParadigm=Black-Box2025.1062.559.1186.34
EntropyParadigm=Black-Box2025.1056.677.2987.08
CoT-KineticsParadigm=Gray-Box2025.1056.547.3586.83
MLP ProbeParadigm=Gray-Box2025.1056.028.6393.94
LR ProbeParadigm=Gray-Box2025.1055.867.9990.32
PPLParadigm=Black-Box2025.1055.468.1290.69
MaxProbParadigm=Black-Box2025.1054.917.9991.86
Temp. ScalingParadigm=Black-Box2025.1054.428.2492.28
CoE-CParadigm=Gray-Box2025.1053.5710.896.33
CoE-RParadigm=Gray-Box2025.1052.388.3496.2