Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Adversarial Robustness on Jailbreak Perturbation-based
Loading...
2.58
Perplexity
NullSteer
2.3288
4.0244
5.72
7.4156
Mar 23, 2026
Perplexity
Updated 25d ago
Evaluation Results
Method
Method
Links
Perplexity
NullSteer
Base Model=MiniGPT-4
2026.03
2.58
Vanilla
Base Model=LLaVA-v1.5
2026.03
3.68
Vanilla
Base Model=MiniGPT-4
2026.03
3.95
ASTRA
Base Model=MiniGPT-4
2026.03
5.82
Vanilla
Base Model=Qwen2-VL
2026.03
6.8
NullSteer
Base Model=Qwen2-VL
2026.03
6.95
NullSteer
Base Model=LLaVA-v1.5
2026.03
7.23
ASTRA
Base Model=LLaVA-v1.5
2026.03
8.59
ASTRA
Base Model=Qwen2-VL
2026.03
8.86
Feedback
Search any
task
Search any
task