List your company

Mathematical Reasoning on MinervaMath (maj@8, maj@16, pass@128)

39maj@8DARS-HW-Breadth
33.69635.07336.4537.827Aug 19, 2025

Evaluation results

MethodLinks
DARS-HW-BreadthBase Model=Qwen2.5-Math-7B2025.083939.464.3
DARS-HWBase Model=Qwen2.5-Math-7B2025.0838.639.362.1
RLVR BaselineBase Model=Qwen2.5-Math-7B2025.0838.539.658.5
DARS-HWBase Model=Qwen2.5-Math-1.5B2025.0834.936.158.8
DARS-HW-BreadthBase Model=Qwen2.5-Math-1.5B2025.0834.635.461
RLVR BaselineBase Model=Qwen2.5-Math-1.5B2025.0833.935.458.8