Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Safety Alignment on SafeRLHF
Loading...
83
Win Rate
Chain-of-Thought
55.96
62.98
70
77.02
Jun 26, 2025
Win Rate
Tie Rate
Loss Rate
Updated 4d ago
Evaluation Results
Method
Method
Links
Win Rate
Tie Rate
Loss Rate
Chain-of-Thought
Main Model=Vicuna, Inf...
2025.06
83
8
9
Best-of-N
Main Model=Vicuna, Inf...
2025.06
77
10
13
Multi-Agent Debate
Main Model=Vicuna, Inf...
2025.06
76
10
14
RLAIF
training_context=after...
2025.06
72
22
6
Chain-of-Thought
Main Model=GPT-4o-mini...
2025.06
71
6
23
Best-of-N
Main Model=GPT-4o-mini...
2025.06
62
20
18
Multi-Agent Debate
Main Model=GPT-4o-mini...
2025.06
61
21
18
Self-Refine (Debate)
training_context=after...
2025.06
57
32
11
Feedback
Search any
task
Search any
task