EvoComp: Learning Visual Token Compression for Multimodal Large Language Models via Semantic-Guided Evolutionary Labeling

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Song, Jiafei, Zhou, Fengwei, Qu, Jin, Li, Wenjin Jason, Wu, Tong, Xue, Gengjian, Zhao, Zhikang, Wei, Daomin, Lu, Yichao, Na, Bailin
Format:	Preprint
Published:	2026
Subjects:	Computer Vision and Pattern Recognition Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2604.17087