Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Jailbreak Attack Evaluation on DeepIn
Loading...
68
ASR
ICD
-2.72
15.64
34
52.36
Aug 9, 2025
ASR
Updated 18d ago
Evaluation Results
Method
Method
Links
ASR
ICD
Base LLM=Vicuna
2025.08
68
No Defense
Base LLM=Vicuna
2025.08
56
Self-Examination
Base LLM=Vicuna
2025.08
44
Self-Reminder
Base LLM=Vicuna
2025.08
40
No Defense
Base LLM=ChatGPT
2025.08
20
Self-Reminder
Base LLM=ChatGPT
2025.08
16
Self-Examination
Base LLM=ChatGPT
2025.08
4
Context Filtering
Base LLM=Vicuna
2025.08
2
No Defense
Base LLM=Llama2
2025.08
2
Intention Analysis
Base LLM=Vicuna
2025.08
0
Safe Decoding
Base LLM=Vicuna
2025.08
0
Self-Reminder
Base LLM=Llama2
2025.08
0
ICD
Base LLM=Llama2
2025.08
0
Self-Examination
Base LLM=Llama2
2025.08
0
Intention Analysis
Base LLM=Llama2
2025.08
0
Safe Decoding
Base LLM=Llama2
2025.08
0
Context Filtering
Base LLM=Llama2
2025.08
0
ICD
Base LLM=ChatGPT
2025.08
0
Intention Analysis
Base LLM=ChatGPT
2025.08
0
Context Filtering
Base LLM=ChatGPT
2025.08
0
Feedback
Search any
task
Search any
task