List your company
ResearchFormerly WizWand

Mathematical Reasoning on Kaoyan (pass@1)

73.37pass@1 ScoreSGPO
27.381239.320651.2663.1994May 16, 2025
Updated 8mo ago

Evaluation results

MethodLinks
SGPOBackbone=QwQ-32B, Judge Model=DeepSeek-V3-03242025.0573.37
GRPOBackbone=QwQ-32B2025.0571.36
BaselineBackbone=QwQ-32B2025.0564.32
SGPOBackbone=DeepSeek-R1-Distill-Qwen-7B, Judge Model=QwQ-32B, Stability Parameters=w/o {beta,gamma}2025.0558.29
SGPOBackbone=DeepSeek-R1-Distill-Qwen-7B, Judge Model=o4-mini-04162025.0557.79
SGPOBackbone=DeepSeek-R1-Distill-Qwen-7B, Judge Model=Qwen3-235B-A22B2025.0556.78
GRPOBackbone=DeepSeek-R1-Distill-Qwen-7B2025.0555.78
SGPOBackbone=DeepSeek-R1-Distill-Qwen-7B, Judge Model=DeepSeek-V3-03242025.0554.77
SGPOBackbone=DeepSeek-R1-Distill-Qwen-7B, Judge Model=QwQ-32B2025.0552.26
BaselineBackbone=DeepSeek-R1-Distill-Qwen-7B2025.0550.25
GRPOBackbone=Qwen2.5-32B-Instruct2025.0548.24
SGPOBackbone=Qwen2.5-32B-Instruct, Judge Model=o4-mini-04162025.0548.24
BaselineBackbone=Qwen2.5-32B-Instruct2025.0545.73
GRPOBackbone=Qwen2.5-14B-Instruct2025.0543.22
SGPOBackbone=DeepSeek-R1-Distill-Llama-8B, Judge Model=Claude-3.72025.0539.7
SGPOBackbone=Qwen2.5-14B-Instruct, Judge Model=o4-mini-04162025.0538.69
BaselineBackbone=Qwen2.5-14B-Instruct2025.0537.69
GRPOBackbone=DeepSeek-R1-Distill-Llama-8B2025.0535.68
BaselineBackbone=DeepSeek-R1-Distill-Llama-8B2025.0529.15