Allocation Before Ranking: Decoupled Token Compression for OmniLLMs

Zhenghui Guo, Yilin Yang, Yuanbin Man, Miao Yin, Weidong Shi, Rabimba Karanjai, Omprakash Gnawali, Chengming Zhang Jan 1, 2026
Scholar
Read Paper

Paper coming soon. Check back or search on Google Scholar.

Search on Scholar →

Abstract

Preprint: We decouple token compression from routing in multi-modal systems, improving processing speeds by up to 30%.

Type Preprint
Venue NeurIPS 2026 (Under Review)
Date Jan 2026
LLMs Token Compression Efficiency
LLMs Token Compression Efficiency