Tingwei Zhang
Tingwei Zhang
Home
Publications
News
CV
Light
Dark
Automatic
3
Soft Prompts Go Hard: Steering Visual Language Models with Hidden Meta-Instructions
We introduce a new type of indirect injection vulnerabilities in language models that operate on images: hidden meta-instructions that …
Tingwei Zhang
,
Collin Zhang
,
John X Morris
,
Eugene Bagdasaryan
,
Vitaly Shmatikov
PDF
Cite
Code
Cite
×