ToDRE: Visual Token Pruning via Diversity and Task Awareness for Efficient Large Vision-Language Models

Add code
May 24, 2025

Share this with someone who'll enjoy it:

View paper onarxiv icon

Share this with someone who'll enjoy it: