Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
CVaR(0.125) Combined Win Rate Evaluation on PKU-SafeRLHF (n=100 Conflict Prompts)
Loading...
40.3
CVaR(0.125) Combined Win Rate
IPO Entropic τ=10
0.884
11.117
21.35
31.583
May 11, 2026
CVaR(0.125) Combined Win Rate
Updated 21d ago
Evaluation Results
Method
Method
Links
CVaR(0.125) Combined Win Rate
IPO Entropic τ=10
tau (τ)=10, objective=...
2026.05
40.3
IPO CVaR α=0.25
alpha (α)=0.25, object...
2026.05
37.2
IPO Entropic τ=5
tau (τ)=5, objective=E...
2026.05
37.1
IPO CVaR α=0.125
alpha (α)=0.125, objec...
2026.05
34.2
EG Entropic τ=5
tau (τ)=5, objective=E...
2026.05
34.1
IPO k = 1 Neutral
k=1, objective=Neutral
2026.05
33.8
EG k = 1 Neutral
k=1, objective=Neutral
2026.05
32.4
NMD k = 1 Neutral
k=1, objective=Neutral
2026.05
28
EGPO† (HF)
source=Hugging Face
2026.05
4.8
SFT Base
type=Baseline
2026.05
2.4
Feedback
Search any
task
Search any
task