List your company

Mathematical Reasoning on AIME 24 (avg@16, pass@32)

35Avg@16ISPO
13.57619.13824.730.262Jun 7, 2026

Evaluation results

MethodLinks
ISPOBackbone=Qwen2.5-7B2026.063567
ProGRPOBackbone=Qwen2.5-7B2026.0626.753.3
Scaf-GRPOBackbone=Qwen2.5-7B2026.0622.550
Vanilla GRPOBackbone=Qwen2.5-7B2026.0614.446.7