An Image Is Worth 1000 Lies: Adversarial Transferability across Prompts on Vision-Language Models

Add code
Mar 14, 2024

Share this with someone who'll enjoy it:

View paper onarxiv icon

Share this with someone who'll enjoy it: