List your company

Mathematical Reasoning on MinervaMath (Pass@1)

23.64Pass@1 AccuracyPROGRS-4
14.269616.702319.13521.5677Feb 8, 2026

Evaluation results

MethodLinks
PROGRS-4rollout_budget=42026.0223.641,081.73
PROGRS-8rollout_budget=82026.0222.131,043.08
PROGRS-8 (αcoh = 0)rollout_budget=8, coherence_penalty=disabled2026.0220.181,012.28
PROGRS-8 (No Centering)rollout_budget=8, centering=none2026.0220.151,016.18
DAPO-16rollout_budget=162026.0218.791,110.73
DAPO-8rollout_budget=82026.0214.631,168