List your company
ResearchFormerly WizWand

Instruction Following on IFBench-P

17.4AccuracyCE-GPPO
11.78413.24214.716.158Sep 25, 2025
Updated 6mo ago

Evaluation results

MethodLinks
CE-GPPOBase Model=DS-R1-Distill-Qwen-7B, Training Algorithm=CE-GPPO, Beta1 Hyperparameter=0.5, Beta2 Hyperparameter=12025.0917.4
CE-GPPOBase Model=DS-R1-Distill-Qwen-7B, Training Algorithm=CE-GPPO, Beta1 Hyperparameter=0.75, Beta2 Hyperparameter=12025.0917.4
DS-R1-Distill-Qwen-7BBase Model=DS-R1-Distill-Qwen-7B2025.0916.8
GRPOBase Model=DS-R1-Distill-Qwen-7B, Training Algorithm=GRPO2025.0916.6
DAPOBase Model=DS-R1-Distill-Qwen-7B, Training Algorithm=DAPO2025.0916.5
CE-GPPOBase Model=DS-R1-Distill-Qwen-1.5B, Training Algorithm=CE-GPPO, Beta1 Hyperparameter=0.75, Beta2 Hyperparameter=12025.0913.8
CE-GPPOBase Model=DS-R1-Distill-Qwen-1.5B, Training Algorithm=CE-GPPO, Beta1 Hyperparameter=0.5, Beta2 Hyperparameter=12025.0913.7
DAPOBase Model=DS-R1-Distill-Qwen-1.5B, Training Algorithm=DAPO2025.0912.8
GRPOBase Model=DS-R1-Distill-Qwen-1.5B, Training Algorithm=GRPO2025.0912.2
DS-R1-Distill-Qwen-1.5BBase Model=DS-R1-Distill-Qwen-1.5B2025.0912