Scale-Induced Semantic Shift Enables Visual Prompt Injection in VLMs
Researchers describe Scale-Induced Semantic Shift (SISS), a security failure mode in which image resizing or compression changes machine-readable semantics and makes hidden textual instructions recoverable after aggressive downsampling. Their closed-loop black-box framework reaches a 58.75% average attack success rate on a fixed 300-case benchmark across four closed-source vision-language models, though the effect is bounded to aggressive compression. In a downstream web-agent study, hidden-instruction perception increases but absolute behavioral compromise remains about 4%. The proposed ScaleGuard consistency detector reduces attack success, while generic input transformations provide little protection.
Useful for AI security teams because it documents a reproducible, bounded preprocessing vulnerability in vision-language models and evaluates both attack and defense.
Comments & Questions (0)
No comments yet
Be the first to comment!