List your company

Mathematical Reasoning on MATH 500 (avg@16, pass@32)

84Average Score @16ISPO
78.38479.84281.382.758Jun 7, 2026

Evaluation results

MethodLinks
ISPOBackbone=Qwen2.5-7B2026.068495.6
ProGRPOBackbone=Qwen2.5-7B2026.0682.894.2
Scaf-GRPOBackbone=Qwen2.5-7B2026.0680.193.6
Vanilla GRPOBackbone=Qwen2.5-7B2026.0678.692.4