List your company

Mathematical Reasoning on HMMT25 (Pass@16)

53.3Pass@16Qwen3-4B + DASD
22.130.238.346.4May 21, 2026

Evaluation results

MethodLinks
Qwen3-4B + DASDModel Scale=4B, Method=DASD2026.0553.3
Qwen3-8B + DASDModel Scale=8B, Method=DASD2026.0553.3
Qwen3-4B + GRPOModel Scale=4B, Method=GRPO2026.0550
Qwen3-8B + GRPOModel Scale=8B, Method=GRPO2026.0546.7
Qwen3-4B + HEPOModel Scale=4B, Method=HEPO2026.0543.3
Qwen3-4B + SRPOModel Scale=4B, Method=SRPO2026.0543.3
Qwen3-8B + HEPOModel Scale=8B, Method=HEPO2026.0543.3
Qwen3-8B + OPSDModel Scale=8B, Method=OPSD2026.0543.3
Qwen3-8B + RLSDModel Scale=8B, Method=RLSD2026.0543.3
Qwen3-8B + SRPOModel Scale=8B, Method=SRPO2026.0543.3
Qwen3-1.7B + DASDModel Scale=1.7B, Method=DASD2026.0540
Qwen3-4B + OPSDModel Scale=4B, Method=OPSD2026.0540
Qwen3-4B + RLSDModel Scale=4B, Method=RLSD2026.0540
Qwen3-8B + SDPOModel Scale=8B, Method=SDPO2026.0540
Qwen3-1.7B + HEPOModel Scale=1.7B, Method=HEPO2026.0536.7
Qwen3-1.7B + OPSDModel Scale=1.7B, Method=OPSD2026.0536.7
Qwen3-4BModel Scale=4B, Method=Base2026.0536.7
Qwen3-4B + SDPOModel Scale=4B, Method=SDPO2026.0536.7
Qwen3-1.7BModel Scale=1.7B, Method=Base2026.0533.3
Qwen3-1.7B + SDPOModel Scale=1.7B, Method=SDPO2026.0533.3
Qwen3-1.7B + RLSDModel Scale=1.7B, Method=RLSD2026.0533.3
Qwen3-8BModel Scale=8B, Method=Base2026.0533.3
Qwen3-1.7B + SRPOModel Scale=1.7B, Method=SRPO2026.0530
Qwen3-1.7B + GRPOModel Scale=1.7B, Method=GRPO2026.0523.3