List your company
ResearchFormerly WizWand

Mathematical Reasoning on CHMath24 (avg@16)

89.39Average Score @16Baseline
19.179637.407355.63573.8627May 16, 2025
Updated 7mo ago

Evaluation results

MethodLinks
BaselineBackbone=QwQ-32B2025.0589.39
GRPOBackbone=QwQ-32B2025.0588.75
SGPOBackbone=QwQ-32B, Judge Model=DeepSeek-V3-03242025.0588.33
SGPOBackbone=DeepSeek-R1-Distill-Qwen-7B, Judge Model=DeepSeek-V3-03242025.0577.29
SGPOBackbone=DeepSeek-R1-Distill-Qwen-7B, Judge Model=o4-mini-04162025.0575
SGPOBackbone=DeepSeek-R1-Distill-Qwen-7B, Judge Model=QwQ-32B2025.0575
SGPOBackbone=DeepSeek-R1-Distill-Qwen-7B, Judge Model=QwQ-32B, Stability Parameters=w/o {beta,gamma}2025.0574.58
BaselineBackbone=DeepSeek-R1-Distill-Qwen-7B2025.0573.75
SGPOBackbone=DeepSeek-R1-Distill-Qwen-7B, Judge Model=Qwen3-235B-A22B2025.0573.33
GRPOBackbone=DeepSeek-R1-Distill-Qwen-7B2025.0571.04
BaselineBackbone=DeepSeek-R1-Distill-Llama-8B2025.0561.46
SGPOBackbone=DeepSeek-R1-Distill-Llama-8B, Judge Model=Claude-3.72025.0558.96
GRPOBackbone=DeepSeek-R1-Distill-Llama-8B2025.0557.08
SGPOBackbone=Qwen2.5-32B-Instruct, Judge Model=o4-mini-04162025.0529.79
BaselineBackbone=Qwen2.5-32B-Instruct2025.0526.87
SGPOBackbone=Qwen2.5-14B-Instruct, Judge Model=o4-mini-04162025.0522.92
GRPOBackbone=Qwen2.5-32B-Instruct2025.0522.5
BaselineBackbone=Qwen2.5-14B-Instruct2025.0521.88
GRPOBackbone=Qwen2.5-14B-Instruct2025.0521.88