Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Helpfulness on GPT-4 Evaluation Template T2 (overall)
Loading...
91.6
Win Rate
SafeDPO
57.5712
66.4056
75.24
84.0744
May 26, 2025
Win Rate
Tie Rate
Lose Rate
Updated 1mo ago
Evaluation Results
Method
Method
Links
Win Rate
Tie Rate
Lose Rate
SafeDPO
Judge Model=GPT-4, Eva...
2025.05
91.6
0.64
7.76
SafeRLHF
Judge Model=GPT-4, Eva...
2025.05
85.51
1.42
13.07
DPO-SAFEBETTER
Judge Model=GPT-4, Eva...
2025.05
75.95
11.27
12.78
DPO-HARMLESS
Judge Model=GPT-4, Eva...
2025.05
72.58
8.67
18.75
DPO-HELPFUL
Judge Model=GPT-4, Eva...
2025.05
58.88
16.73
24.39
Feedback
Search any
task
Search any
task