Calibration

MedXpertQA (MM) → ERQA

Name: benchgap: LLM benchmark scores, measured and estimated
Creator: benchgap

ERQA is estimated from MedXpertQA (MM) with a inverse Michaelis–Menten curve fitted on 7 models measured on both: y = 1.08756·(x − 0.0000) / (0.0000 + 2.0000 − x), R² = 0.63, cross-validated error 6.0 pp. It is used for 0 estimates.