{"slug": "caption-once-frames-on-demand-visual-need-routing-for-budget-aware-agentic-long", "title": "Caption-once, Frames-on-Demand: Visual-Need Routing for Budget-Aware Agentic Long Video Understanding", "summary": "Researchers submitted a paper to arXiv on 10 September 2026 proposing Caption-once, Frames-on-Demand (CFD), a budget-aware edge-cloud agentic framework for long-video understanding that runs a single offline captioning pass to build a dual-track narrative index and uses a cloud-side Visual-Need Router to trigger bounded keyframe retrieval only for perceptual questions. CFD caps per-query frame consumption regardless of video length, and experiments on long-video benchmarks showed strong accuracy-efficiency trade-offs while substantially reducing online visual processing.", "body_md": "# Computer Science > Computer Vision and Pattern Recognition\n\n  [Submitted on 10 Sep 2026]\n\n# Title:Caption-once, Frames-on-Demand: Visual-Need Routing for Budget-Aware Agentic Long Video Understanding\n\n[View PDF](/pdf/2609.11899v1)\n\n[HTML (experimental)](https://arxiv.org/html/2609.11899v1)\n\nAbstract:Long-video understanding on edge devices must reason over hours of content under tight compute and bandwidth budgets. Subsampling visual tokens loses temporal structure, while text-only video memories lose fine-grained visual attributes. We observe a visual-textual duality: language memories carry long-range temporal structure better than dense frames, while pixels remain decisive for attribute-level perception. Building on this insight, we propose Caption-once, Frames-onDemand (CFD), a budget-aware edge-cloud agentic framework. The edge runs a single offline captioning pass that builds a dual-track narrative index, an event-level story skeleton plus a clip-level micro-log, cached and reused across queries without re-captioning. At query time, a cloud-side MLLM reasons over the index in a story-first loop centered on a lightweight Visual-Need Router: a per-query gating module that triggers bounded keyframe retrieval only for perceptual questions (appearance, on-screen text, attribute disambiguation) and keeps temporal-structural questions in language space. The router turns visual access into a first-class, query-conditioned cost, capping per-query frame consumption regardless of video length. Experiments on long-video benchmarks demonstrate strong accuracy-efficiency trade-offs while substantially reducing online visual processing.\n    \n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer \n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers \n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps \n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations \n\n*(*[What are Smart Citations?](https://www.scite.ai/))\n# Code, Data and Media Associated with this Article\n\nalphaXiv \n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers \n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub \n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub \n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face \n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast \n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))\n# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower \n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender \n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))\n# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [**Learn more about arXivLabs**](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/caption-once-frames-on-demand-visual-need-routing-for-budget-aware-agentic-long", "canonical_source": "http://arxiv.org/abs/2609.11899v1", "published_at": "2026-09-11 14:32:19+00:00", "updated_at": "2026-09-11 14:43:48.777736+00:00", "lang": "en", "topics": ["computer-vision", "artificial-intelligence", "ai-research", "ai-agents", "ai-infrastructure"], "entities": ["arXiv", "Caption-once, Frames-on-Demand", "CFD", "Visual-Need Router"], "alternates": {"html": "https://wpnews.pro/news/caption-once-frames-on-demand-visual-need-routing-for-budget-aware-agentic-long", "markdown": "https://wpnews.pro/news/caption-once-frames-on-demand-visual-need-routing-for-budget-aware-agentic-long.md", "text": "https://wpnews.pro/news/caption-once-frames-on-demand-visual-need-routing-for-budget-aware-agentic-long.txt", "jsonld": "https://wpnews.pro/news/caption-once-frames-on-demand-visual-need-routing-for-budget-aware-agentic-long.jsonld"}}