Risk Awareness Injection: Calibrating Vision-Language Models for Safety without Compromising Utility

Add code
Feb 03, 2026

Share this with someone who'll enjoy it:

View paper onarxiv icon

Share this with someone who'll enjoy it: