List your company

Instruction Following on MT-Bench Math

55.14Original RatioPUAUDIT
50.293651.551852.8154.0682Jun 17, 2026

Evaluation results

MethodLinks
PUAUDITModel=Qwen, n=91452026.0655.1456.851.71
PUAUDITModel=Mistral, n=83612026.0650.4854.163.67