Do We Really Need Curated Malicious Data for Safety Alignment in Multi-modal Large Language Models?

Add code
Apr 14, 2025

Share this with someone who'll enjoy it:

View paper onarxiv icon

Share this with someone who'll enjoy it: