Vision-centric Token Compression in Large Language Model

Xing, Ling, Wang, Alex Jinpeng, Yan, Rui · arXiv · 2025