List your company

Combined Win Rate Evaluation on PKU-SafeRLHF Sev-3 (n=100)

60.9Combined Win Rate (CVaR 0.125)IPO Entropic τ=10
-0.4615.4731.447.33May 11, 2026

Evaluation results

MethodLinks
IPO Entropic τ=10tau (τ)=10, objective=Entropic2026.0560.9
IPO k = 1 Neutralk=1, objective=Neutral2026.0557.2
EG Entropic τ=5tau (τ)=5, objective=Entropic2026.0557.2
IPO CVaR α=0.125alpha (α)=0.125, objective=CVaR2026.0556.8
IPO CVaR α=0.25alpha (α)=0.25, objective=CVaR2026.0556.5
EG k = 1 Neutralk=1, objective=Neutral2026.0554.9
IPO Entropic τ=5tau (τ)=5, objective=Entropic2026.0554.8
NMD k = 1 Neutralk=1, objective=Neutral2026.0552.5
EGPO† (HF)source=Hugging Face2026.0511
SFT Basetype=Baseline2026.051.9