List your company

Mathematical Problem Solving on ORZ-57K, AIME24/25, Minerva, Math500, AMC, OlympiadBench (test)

31.66Accuracy (Minerva)ACPO
12.586417.538222.4927.4418Jun 21, 2026

Evaluation results

MethodLinks
ACPOBackbone=Qwen2.5-7B, Policy Regime=OffP.2026.0631.6680.4851.216.3510.142.1738.66
ACPOBackbone=Qwen2.5-7B, Policy Regime=N-OnP.2026.0631.0279.7854.3716.7714.3742.8339.86
Base ModelBackbone=Qwen2.5-7B, Policy Regime=N-OnP.2026.0613.3241.4826.20.830.6218.0216.75