SOURCE-LINKED INTELLIGENCE

CoFiE: Coarse-to-Fine Evidence Selection for Efficient Streaming Video Understanding

arXiv · AI, language, vision and robotics · article · Sep 3, 2026 · UTC

Streaming video understanding requires Vision Language Models (VLLMs) to process growing video streams and answer user questions under tight latency constraints. Existing methods improve efficiency through token pruning and memory-bank schemes, but mainly reduce visual tokens after visual encoding. Consequently, downstream token pruning alone cannot substantially reduce end-to-end latency because the expensive frame encoding cost has already been incurred. We propose CoFiE, a Coarse-to-Fine Evidence Selection framework that decouples evidence selection into a coarse, query-agnostic filtering s

Read original source ↗ Open in workspace

recordType

paper

region

Global