List your company

Language Understanding on MMLU-Pro (Brier score)

19.26Brier ScoreCAGE-CAL
18.61222.98627.3631.734May 28, 2026

Evaluation results

MethodLinks
CAGE-CALCalibration category=Trained calibrators2026.0519.26
DiscoUQ-LLMCalibration category=Trained calibrators2026.0524.02
+ Scaling-bin.Calibration category=Post-hoc plurality calibrators2026.0524.21
+ PlattCalibration category=Post-hoc plurality calibrators2026.0524.3
+ IsotonicCalibration category=Post-hoc plurality calibrators2026.0524.76
Plurality shareCalibration category=Post-hoc plurality calibrators2026.0525.01
Scalar + GBTCalibration category=Trained calibrators2026.0525.96
GraphCalCalibration category=Trained calibrators2026.0526.28
Collab. Cal.Calibration category=LLM-elicited confidence estimators2026.0533.71
LLM-Cal (+topo)Calibration category=LLM-elicited confidence estimators2026.0534.6
LLM-Cal (no topo)Calibration category=LLM-elicited confidence estimators2026.0535.46