List your company

Overall Performance Evaluation on Aggregate All Domains

38.14Average AccuracyVRPO
23.184827.067430.9534.8326Aug 5, 2025

Evaluation results

MethodLinks
VRPOModel Scale=Qwen3-8B2025.0838.14
λ-GRPOModel Scale=Qwen3-8B2025.0837.95
Reinforce++Model Scale=Qwen3-8B2025.0836.6
Dr.GRPOModel Scale=Qwen3-8B2025.0836.59
GRPOModel Scale=Qwen3-8B2025.0835.19
BaseModel Scale=Qwen3-8B2025.0834.64
VRPOModel Scale=Qwen3-1.7B2025.0834.56
PPOModel Scale=Qwen3-8B2025.0834.55
KTAEModel Scale=Qwen3-8B2025.0834.55
Reinforce++Model Scale=Qwen3-1.7B2025.0833.26
λ-GRPOModel Scale=Qwen3-1.7B2025.0832.68
PPOModel Scale=Qwen3-1.7B2025.0832.51
GRPOModel Scale=Qwen3-1.7B2025.0832.43
Dr.GRPOModel Scale=Qwen3-1.7B2025.0832.19
BaseModel Scale=Qwen3-1.7B2025.0830.78
KTAEModel Scale=Qwen3-1.7B2025.0823.76