Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Robustness to Jailbreak Attacks on JBB Paraphrased
Loading...
16.1
Harmful Reasoning Ratio
IPO
13.96
28.405
42.85
57.295
Sep 29, 2025
Harmful Reasoning Ratio
Harmful Response Ratio
Updated 1mo ago
Evaluation Results
Method
Method
Links
Harmful Reasoning Ratio
Harmful Response Ratio
IPO
Backbone=DS-8B
2025.09
16.1
1.4
RealSafe
Backbone=DS-8B
2025.09
20.4
3.2
Base
Backbone=DS-8B
2025.09
69.6
44.6
Feedback
Search any
task
Search any
task