Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Safety Alignment on AdvBench
Loading...
99
Wins
Chain-of-Thought
30.36
48.18
66
83.82
Jun 26, 2025
Wins
Ties
Losses
Updated 4d ago
Evaluation Results
Method
Method
Links
Wins
Ties
Losses
Chain-of-Thought
Main Model=GPT-4o-mini...
2025.06
99
0
1
Best-of-N
Main Model=GPT-4o-mini...
2025.06
99
1
0
Multi-Agent Debate
Main Model=GPT-4o-mini...
2025.06
99
1
0
Self-Refine (Debate)
training_context=after...
2025.06
33
61
6
RLAIF
training_context=after...
2025.06
33
64
3
Feedback
Search any
task
Search any
task