List your company

Mathematical Reasoning on HMMT25 (Avg@32)

17.9Avg@32MHPO
-0.702484.127018.956513.78599Mar 14, 2026Mar 30, 2026Apr 16, 2026May 2, 2026May 19, 2026Jun 4, 2026Jun 21, 2026

Evaluation results

MethodLinks
MHPOBackbone=Qwen2.5-Math-7B-Instruct2026.0317.9
MHPOBackbone=Qwen2.5-Math-7B-Instruct2026.0317.9
MHPOBackbone=Qwen2.5-7B-Instruct2026.0313.8
DAPOBackbone=Qwen2.5-Math-7B-Instruct2026.0311.5
DAPOBackbone=Qwen2.5-Math-7B-Instruct2026.0311.5
GSPOBackbone=Qwen2.5-Math-7B-Instruct2026.0310.9
GSPOBackbone=Qwen2.5-Math-7B-Instruct2026.0310.9
SAPOBackbone=Qwen2.5-Math-7B-Instruct2026.0310
SAPOBackbone=Qwen2.5-Math-7B-Instruct2026.0310
GRPOBackbone=Qwen2.5-Math-7B-Instruct2026.039.6
GRPOBackbone=Qwen2.5-Math-7B-Instruct2026.039.6
DAPOBackbone=Qwen2.5-7B-Instruct2026.039.1
GSPOBackbone=Qwen2.5-7B-Instruct2026.038.5
SAPOBackbone=Qwen2.5-7B-Instruct2026.038
GRPOBackbone=Qwen2.5-7B-Instruct2026.037.8
BaselineBackbone=Qwen2.5-Math-7B-Instruct2026.037.5
BaselineBackbone=Qwen2.5-Math-7B-Instruct2026.037.5
BaselineBackbone=Qwen2.5-7B-Instruct2026.035.6
Qwen3-30B-A3B-Instruct-2507Role=Teacher Model2026.060.442
IW-OPDStudent=Qwen3-4B2026.060.287
IW-OPDStudent=Qwen3-4B, Training Step=102026.060.273
OPDStudent=Qwen3-4B2026.060.271
OPDStudent=Qwen3-4B, Training Step=102026.060.235
IW-OPDStudent=Qwen3-1.7B2026.060.164
IW-OPDStudent=Qwen3-1.7B, Training Step=102026.060.156
OPDStudent=Qwen3-1.7B2026.060.155
OPDStudent=Qwen3-1.7B, Training Step=102026.060.144
BaseStudent=Qwen3-4B2026.060.1
IW-OPDStudent=Qwen3-0.6B2026.060.08
IW-OPDStudent=Qwen3-0.6B, Training Step=102026.060.076
OPDStudent=Qwen3-0.6B2026.060.071
BaseStudent=Qwen3-1.7B2026.060.068
OPDStudent=Qwen3-0.6B, Training Step=102026.060.065
BaseStudent=Qwen3-0.6B2026.060.013