Calibration
HealthBench Professional → ARC-AGI-3
ARC-AGI-3 is estimated from HealthBench Professional with a offset logistic curve fitted on 9 models measured on both: y = 0.0340 + (1.2000 − 0.0340) / (1 + exp(−75.59·(x − 0.6465))), R² = 0.85, cross-validated error 11.5 pp. It is used for 0 estimates.