List your company

Calibration on TruthfulQA

2.132GainS
-1.03064-0.209570.61151.43257Apr 19, 2026

Evaluation results

MethodLinks
SEvaluation Metric=ECE'2026.042.1324———
SEvaluation Metric=GECE'2026.042.0733———
BEvaluation Metric=GECE'2026.042.0140———
GIRBEvaluation Metric=GECE'2026.042.011———
QABEvaluation Metric=GECE'2026.041.9840———
BEvaluation Metric=ECE'2026.041.9360———
QABEvaluation Metric=ECE'2026.041.9010———
GIRBEvaluation Metric=ECE'2026.041.8960———
SEvaluation Metric=Brier'2026.041.482———
BEvaluation Metric=Brier'2026.041.4640———
GIRBEvaluation Metric=Brier'2026.041.4332———
QABEvaluation Metric=Brier'2026.041.4160———
S-BEvaluation Metric=ECE'2026.041.0080———
HS-QABEvaluation Metric=GECE'2026.040.9940———
S-BEvaluation Metric=GECE'2026.040.9810———
HS-QABEvaluation Metric=ECE'2026.040.9740———
S-QABEvaluation Metric=GECE'2026.040.9540———
S-QABEvaluation Metric=ECE'2026.040.9240———
S-BEvaluation Metric=Brier'2026.040.8860———
HS-QABEvaluation Metric=Brier'2026.040.7960———
S-QABEvaluation Metric=Brier'2026.040.7760———
GIRBEvaluation Metric=AUAC2026.040.2261———
QABEvaluation Metric=AUAC2026.040.111———
S-QABEvaluation Metric=AUAC2026.040.0760———
HS-QABEvaluation Metric=AUAC2026.040.0531———
NoneEvaluation Metric=ECE'2026.0400———
NoneEvaluation Metric=GECE'2026.0400———
NoneEvaluation Metric=Brier'2026.0400———
NoneEvaluation Metric=AUAC2026.0401———
S-BEvaluation Metric=AUAC2026.04-0.5540———
BEvaluation Metric=AUAC2026.04-0.6510———
SEvaluation Metric=AUAC2026.04-0.9090———
BaseModel=LLAMA3-8B, Temperature Scaling=false2026.05————46.2
BaseModel=LLAMA3-8B, Temperature Scaling=true2026.05————24.5
BaseModel=QWEN2-7B, Temperature Scaling=false2026.05————47.5
BaseModel=QWEN2-7B, Temperature Scaling=true2026.05————47.5
CAGE-CALType=Trained2026.05——6.579.66—
Collab. Cal.Type=LLM-elicited2026.05——52.8778.24—
DiscoUQ-LLMType=Trained2026.05——6.9971.93—
GraphCalType=Trained2026.05——33.6663.75—
LLM-Cal (+topo)Type=LLM-elicited, Topology=Included2026.05——54.5278.48—
LLM-Cal (no topo)Type=LLM-elicited, Topology=None2026.05——54.8878.5—
Plurality shareType=Post-hoc plurality2026.05——25.2371.16—
Plurality share + IsotonicType=Post-hoc plurality, Technique=Isotonic regression2026.05——8.4770.45—
Plurality share + PlattType=Post-hoc plurality, Technique=Platt scaling2026.05——9.8271.16—
Plurality share + Scaling-bin.Type=Post-hoc plurality, Technique=Scaling-binning2026.05——8.4170.83—
PPT (Distribution)Model=LLAMA3-8B, Temperature Scaling=false2026.05————27.5
PPT (Distribution)Model=LLAMA3-8B, Temperature Scaling=true2026.05————22.3
PPT (Distribution)Model=QWEN2-7B, Temperature Scaling=false2026.05————38.1
PPT (Distribution)Model=QWEN2-7B, Temperature Scaling=true2026.05————38.3
PPT (Mean)Model=LLAMA3-8B, Temperature Scaling=false2026.05————37.7
PPT (Mean)Model=QWEN2-7B, Temperature Scaling=false2026.05————47.5
Scalar + GBTType=Trained2026.05——10.2866.81—
VerbalizedModel=LLAMA3-8B, Temperature Scaling=false2026.05————47