Picture for Tzu-Tao Chang

Tzu-Tao Chang

LV-XAttn: Distributed Cross-Attention for Long Visual Inputs in Multimodal Large Language Models

Add code
Feb 04, 2025
Viaarxiv icon