Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Preference Optimization on Alpaca-GPT4 Style
Loading...
85.3
Win Rate
F-beta
83.9272
84.2836
84.64
84.9964
Jun 5, 2025
Win Rate
PPA (α)
Updated 3mo ago
Evaluation Results
Method
Method
Links
Win Rate
PPA (α)
F-beta
beta (β)=10^0, Backbon...
2025.06
85.3
36.35
DPO
Backbone=Qwen2.5-3B-In...
2025.06
84.78
37.86
F-beta
beta (β)=10^-4, Backbo...
2025.06
84.32
41.97
F-beta
beta (β)=10^-2, Backbo...
2025.06
84.25
36.37
F-beta
beta (β)=0, Backbone=Q...
2025.06
83.98
50.12
Feedback
Search any
task
Search any
task
