Delta-LLaVA: Base-then-Specialize Alignment for Token-Efficient Vision-Language Models

Add code
Dec 21, 2025

Share this with someone who'll enjoy it:

View paper onarxiv icon

Share this with someone who'll enjoy it: