Visual Grounding Safety in Vision-Language Models
Vision-language models (VLMs) are increasingly trained to generate structured outputs like points and bounding boxes that downstream interfaces, agents, and robots can act on, yet safety alignment of this output channel has not been systematically analyzed. We study visual grounding safety by repurposing three safety b...
Erfan Shayegani, Kundan Krishna, Yue Dong et al.
· 0 citations