Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Over-Prudence Evaluation on VLGuard
Loading...
4.48
RR (Before)
Mixed-SFT
0.1952
1.3076
2.42
3.5324
Mar 14, 2025
RR (Before)
RR (After)
Updated 1mo ago
Evaluation Results
Method
Method
Links
RR (Before)
RR (After)
Mixed-SFT
Base Model=LLaVA-1.5-7...
2025.03
4.48
91.76
Posthoc-SFT
Base Model=LLaVA-1.5-7...
2025.03
2.69
90.83
NPO-Unlearning
Base Model=LLaVA-1.5-7...
2025.03
2.51
11.69
RMU-Unlearning
Base Model=LLaVA-1.5-7...
2025.03
1.25
7.56
LLaVA-1.5-7B
Fine-tuning Strategy=F...
2025.03
0.36
0.36
Unsafe-Filter
Base Model=LLaVA-1.5-7...
2025.03
0.36
0
Feedback
Search any
task
Search any
task