Calibration
ARC-AGI-3 → Vals MMLU-Pro
Vals MMLU-Pro is estimated from ARC-AGI-3 with a Michaelis–Menten curve fitted on 11 models measured on both: y = 0.8978·x / (0.00004 + x), R² = 0.44, cross-validated error 1.7 pp. It is used for 0 estimates.
Vals MMLU-Pro is estimated from ARC-AGI-3 with a Michaelis–Menten curve fitted on 11 models measured on both: y = 0.8978·x / (0.00004 + x), R² = 0.44, cross-validated error 1.7 pp. It is used for 0 estimates.