Vision-centric Token Compression in Large Language Model

NeurIPSSpotlight2025

Authors: Ling Xing, Alex Jinpeng Wang, Rui Yan, Xiangbo Shu, Jinhui Tang
Venue: NeurIPS 2025
Track: Spotlight

TL;DR

Real-world applications are stretching context windows to hundreds of thousand of tokens while Large Language Models (LLMs) swell from billions to trillions of parameters…

Opening excerpt from the authors’ abstract. source

Read the paper

OpenReview

Topics

large language model language model llm

← All NeurIPS 2025 Spotlight papers · Browse the whole archive