List your company
ResearchFormerly WizWand

Code Reasoning on LCB v6

53.6AccuracyCE-GPPO
18.177627.373836.5745.7662Sep 25, 2025Nov 4, 2025Dec 14, 2025Jan 24, 2026Mar 5, 2026Apr 14, 2026May 25, 2026
Updated 5mo ago

Evaluation results

MethodLinks
CE-GPPOBase Model=DS-R1-Distill-Qwen-7B, Training Algorithm=CE-GPPO, Beta1 Hyperparameter=0.75, Beta2 Hyperparameter=12025.0953.6
DAPOBase Model=DS-R1-Distill-Qwen-7B, Training Algorithm=DAPO2025.0952.2
CE-GPPOBase Model=DS-R1-Distill-Qwen-7B, Training Algorithm=CE-GPPO, Beta1 Hyperparameter=0.5, Beta2 Hyperparameter=12025.0952.2
GRPOBase Model=DS-R1-Distill-Qwen-7B, Training Algorithm=GRPO2025.0949.2
DS-R1-Distill-Qwen-7BBase Model=DS-R1-Distill-Qwen-7B2025.0949
MCPO-DAPOBackbone=Qwen3-8B2026.0537.53
MCPO-GRPOBackbone=Qwen3-8B2026.0534.29
MCPO-DAPOBackbone=Qwen3-4B2026.0532.71
MGSBackbone=Qwen3-8B2026.0532.63
MT-GRPOBackbone=Qwen3-8B2026.0532.18
DAPOBackbone=Qwen3-8B2026.0531.73
CE-GPPOBase Model=DS-R1-Distill-Qwen-1.5B, Training Algorithm=CE-GPPO, Beta1 Hyperparameter=0.5, Beta2 Hyperparameter=12025.0931.7
CE-GPPOBase Model=DS-R1-Distill-Qwen-1.5B, Training Algorithm=CE-GPPO, Beta1 Hyperparameter=0.75, Beta2 Hyperparameter=12025.0931.1
MCPO-GRPOBackbone=Qwen3-4B2026.0530.77
CLIPOBackbone=Qwen3-8B2026.0530.64
DAPOBase Model=DS-R1-Distill-Qwen-1.5B, Training Algorithm=DAPO2025.0930.5
GRPOBackbone=Qwen3-8B2026.0530.41
MGSBackbone=Qwen3-4B2026.0528.53
MT-GRPOBackbone=Qwen3-4B2026.0528.34
DAPOBackbone=Qwen3-4B2026.0527.93
GRPOBackbone=Qwen3-4B2026.0527.91
CLIPOBackbone=Qwen3-4B2026.0527.25
GRPOBase Model=DS-R1-Distill-Qwen-1.5B, Training Algorithm=GRPO2025.0927.1
DS-R1-Distill-Qwen-1.5BBase Model=DS-R1-Distill-Qwen-1.5B2025.0925.1
Base ModelBackbone=Qwen3-8B2026.0521.68
Base ModelBackbone=Qwen3-4B2026.0519.54