VLMShield: Efficient and Robust Defense of Vision-Language Models against Malicious Prompts

Add code
Apr 07, 2026

Share this with someone who'll enjoy it:

View paper onarxiv icon

Share this with someone who'll enjoy it: