Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Adversarial Robustness on Jailbreak Structured-based
Loading...
2.62
Perplexity
Vanilla
1.488
9.129
16.77
24.411
Mar 23, 2026
Perplexity
Updated 25d ago
Evaluation Results
Method
Method
Links
Perplexity
Vanilla
Base Model=MiniGPT-4
2026.03
2.62
NullSteer
Base Model=MiniGPT-4
2026.03
3.65
Vanilla
Base Model=LLaVA-v1.5
2026.03
3.82
NullSteer
Base Model=LLaVA-v1.5
2026.03
4.26
ASTRA
Base Model=MiniGPT-4
2026.03
4.29
ASTRA
Base Model=LLaVA-v1.5
2026.03
4.61
Vanilla
Base Model=Qwen2-VL
2026.03
30
NullSteer
Base Model=Qwen2-VL
2026.03
30.48
ASTRA
Base Model=Qwen2-VL
2026.03
30.92
Feedback
Search any
task
Search any
task