List your company

Instruction Following on Custom (test)

71.4SSRPT-GRPO
54.13658.61863.167.582Jun 4, 2026

Evaluation results

MethodLinks
PT-GRPOModel=Gemma-2-2B-Instruct, Group Size (G)=82026.0671.430.6
DA-PT-GRPOModel=Gemma-2-2B-Instruct, Group Size (G)=82026.0670.833.4
DA-GRPOModel=Gemma-2-2B-Instruct, Group Size (G)=82026.0670.732.6
MDP-GRPOModel=Gemma-2-2B-Instruct, Group Size (G)=82026.0670.332.8
MT-GRPOModel=Gemma-2-2B-Instruct, Group Size (G)=82026.0668.629.6
GRPOModel=Gemma-2-2B-Instruct, Group Size (G)=82026.0668.429
DA-PT-GRPOModel=Llama-3.2-3B-Instruct, Group Size (G)=82026.0666.325.4
DA-GRPOModel=Llama-3.2-3B-Instruct, Group Size (G)=82026.0666.124.9
MDP-GRPOModel=Llama-3.2-3B-Instruct, Group Size (G)=82026.0665.825.2
MT-GRPOModel=Llama-3.2-3B-Instruct, Group Size (G)=82026.0665.424.6
GRPOModel=Llama-3.2-3B-Instruct, Group Size (G)=82026.0665.124.8
PT-GRPOModel=Llama-3.2-3B-Instruct, Group Size (G)=82026.0664.723.6
BaselineModel=Llama-3.2-3B-Instruct, Group Size (G)=82026.0660.320.8
BaselineModel=Gemma-2-2B-Instruct, Group Size (G)=82026.0654.818.8