List your company

Mathematical Reasoning on GSM8K (Pass@8, Avg@8)

83.85Average Score @8PowerOPD
75.904477.967280.0382.0928Jun 15, 2026

Evaluation results

MethodLinks
PowerOPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base, alpha=5002026.0683.8594.16322
PowerOPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base, alpha=1002026.0683.0693.93319
Vanilla OPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base, Stabilization=Z-Score2026.0680.2695.22323
Full-vocabulary KL OPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base2026.0680.2586.54,095
PowerOPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base, alpha=0.12026.0679.9694.31319
PowerOPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base, alpha=52026.0679.5494.16312
PowerOPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base, alpha=102026.0678.7993.93321
PowerOPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base, alpha=0.52026.0678.2294.77308
Vanilla OPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base, Stabilization=Tanh2026.0678.0294.92321
Vanilla OPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base2026.0677.5595.07323
Vanilla OPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base, Stabilization=Clip2026.0676.9694.84308
PowerOPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base, alpha=12026.0676.2194.31311