SOURCE-LINKED INTELLIGENCE

VideoMM: Adaptive Macro-Micro Inference for Efficient Video MLLMs

arXiv · AI, language, vision and robotics · article · Sep 15, 2026 · UTC

Scaling Multimodal Large Language Models (MLLMs) to long-form video understanding is bottlenecked by the explosion of visual tokens, which saturates context windows and incurs prohibitive costs. Current solutions predominantly rely on auxiliary models for token reduction but face a fundamental dilemma: lightweight encoder-driven approaches often overlook critical semantic information, whereas heavyweight MLLM-driven reduction negates the efficiency gains. {In this work, we identify a more fundamental inefficiency underlying this dilemma: while fine-grained visual details are essential for deta

Read original source ↗ Open in workspace

recordType

paper

region

Global