List your company

Safe and Helpful Response Generation on SafeRLHF-30K (test)

94.3Safe Response RateSafeDPO
40.94854.79968.6582.501May 26, 2025

Evaluation results

MethodLinks
SafeDPO2025.0594.349.9
SafeRLHF2025.0593.249.7
SFT2025.054386.8