Omni-Prune: Query-Aware Unified Token Pruning for Efficient Omnimodal Large Language Models 文章

ArXiv CS.CV2026-07-28PAPERen作者: Yiming Zhong, Chang Nie, Caifeng Shan

Omni-Prune: Query-Aware Unified Token Pruning for Efficient Omnimodal Large Language Models · 相关技术