arXiv paper: Caption-once, Frames-on-Demand: Visual-Need Routing for Budget-Aware Agentic Long Video Understanding
A new arXiv AI paper by Weitong Cai, Hang Zhang, and Yukai Huang, and 6 more studies Caption-once, Frames-on-Demand: Visual-Need Routing for Budget-Aware Agentic Long Video Understanding.
ResearchAI
Follow arXiv AI/ML to make it a durable For You signal.