From Local Evidence to Safety Verdicts: Causal Tracing in Vision-Language Models
A vision-language model may need to combine an image with a prompt to recognize a safety risk that neither reveals alone. Where does this joint safety judgment become accessible inside the model? We introduce SSU-Bench, a dataset of matched safe and unsafe image-text combinations constructed using single-item prompt ed...
Faezeh Dehghan Tarzjani, Mevan Wijewardena, A. Romanus et al.
· 0 citations