List your company

Confidence Estimation on Internal Verification Suite GPT-4.1-mini (test)

0.912Q1 Attribution Present (AUROC)Logprob
0.801760.830380.8590.88762May 11, 2026

Evaluation results

MethodLinks
LogprobBase Model=GPT-4.1-mini2026.050.9120.7160.7060.840.9050.9950.602
VerbalizedBase Model=GPT-4.1-mini2026.050.8510.632—0.61——0.735
VERDI CVBase Model=GPT-4.1-mini, Evaluation Protocol=5-fold CV2026.050.8060.8250.9150.8770.90510.459