Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Safety Evaluation on FigStep (test)

0ASR

LLaVA-1.5-7B + Mixed-SFT

-3.4419.784366.22Mar 14, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.03
0
2025.03
0
2025.03
0
2025.03
0
2025.03
2
2025.03
4
2025.03
4
2025.03
6
2025.03
10
2025.03
12
2025.03
12
2025.03
16
2025.03
20
2025.03
28
2025.03
36
2025.03
40
2025.03
62
2025.03
62
2025.03
72
2025.03
74
2025.03
84
2025.03
84
2025.03
84
2025.03
86