VisInject: Disruption != Injection -- A Dual-Dimension Evaluation of Universal Adversarial Attacks on Vision-Language Models
Paper • 2605.01449 • Published
Jeff Liu
jeffliulab
·
AI & ML interests
DL, RL, MLOps
Recent Activity
updated a Space about 11 hours ago
jeffliulab/visinject authored a paper about 23 hours ago
VisInject: Disruption != Injection -- A Dual-Dimension Evaluation of Universal Adversarial Attacks on Vision-Language Models updated a Space 1 day ago
jeffliulab/predict-powerOrganizations
None yet
authored a
paper about 23 hours ago