List your company

Safety Alignment on PKU-SafeRLHF-10K (test)

356.32HV ScoreCAGE
216.5856252.8628289.14325.4172May 8, 2026

Evaluation results

MethodLinks
CAGEBase Model=Alpaca-65B, Reward Model Size=7B2026.05356.3282.8
GenARMBase Model=Alpaca-65B, Reward Model Size=7B2026.05262.2372.2
PARMBase Model=Alpaca-65B, Reward Model Size=7B2026.05221.9665.7