PLPHP: Per-Layer Per-Head Vision Token Pruning for Efficient Large Vision-Language Models

Add code
Feb 20, 2025

Share this with someone who'll enjoy it:

View paper onarxiv icon

Share this with someone who'll enjoy it: