List your company

Mathematical Reasoning on AIME 24 (maj@8, maj@16, pass@128)

43.3maj@8DARS-HW-Breadth
17.71624.3583137.642Aug 19, 2025

Evaluation results

MethodLinks
DARS-HW-BreadthBase Model=Qwen2.5-Math-7B2025.0843.344.776.7
DARS-HWBase Model=Qwen2.5-Math-7B2025.0837.343.376.7
RLVR BaselineBase Model=Qwen2.5-Math-7B2025.0827.333.373.3
DARS-HW-BreadthBase Model=Qwen2.5-Math-1.5B2025.0822.727.366.7
RLVR BaselineBase Model=Qwen2.5-Math-1.5B2025.0819.322.756.7
DARS-HWBase Model=Qwen2.5-Math-1.5B2025.0818.722.760