List your company

Math reasoning on HMMT25 Nov.

15.83Mean@8MOPD
9.3311.017512.70514.3925May 12, 2026

Evaluation results

MethodLinks
MOPDBase Model=Qwen3-4B2026.0515.8322.47
GRPOBase Model=Qwen3-4B2026.0513.1619.53
SDPOBase Model=Qwen3-4B2026.0512.0118.49
Qwen3-4Bstatus=base model2026.059.5813.91