{"ID":2837041,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.20460","arxiv_id":"2511.20460","title":"Look Where It Matters: Training-Free Ultra-HR Remote Sensing VQA via Adaptive Zoom Search","abstract":"With advances in satellite constellations, sensor technologies, and imaging pipelines, ultra-high-resolution (Ultra-HR) remote sensing imagery is becoming increasingly widespread. However, current remote sensing foundation models are ill-suited to such inputs: full-image encoding exhausts token and memory budgets, while resize-based preprocessing loses fine-grained and answer-critical details. In this context, guiding the model look where it matters before prediction becomes crucial. Therefore, we present ZoomSearch, a training-free, plug-and-play pipeline that decouples 'where to look' from 'how to answer' for Ultra-HR Remote Sensing Visual Question Answering (RS-VQA). ZoomSearch combines Adaptive Multi-Branch Zoom Search, which performs a hierarchical search over image patches to localize query-relevant regions, with Layout-Aware Patch Reassembly, which reorganizes the selected patches into a compact, layout-faithful canvas. We conduct comprehensive experiments on Ultra-HR RS-VQA benchmarks MME-RealWorld-RS and LRS-VQA, comparing against (i) strong general foundation models, (ii) remote sensing foundation models, (iii) Ultra-HR RS-VQA methods, and (iv) plug-and-play search-based VQA methods. When integrated with LLaVA-ov, ZoomSearch attains state-of-the-art accuracy across diverse tasks, improving the LLaVA-ov baseline by 26.3% on LRS-VQA and 114.8% on MME-RealWorld-RS. Meanwhile, it achieves much higher inference efficiency, outperforming prior search-based methods by 20%~44% in speed.","short_abstract":"With advances in satellite constellations, sensor technologies, and imaging pipelines, ultra-high-resolution (Ultra-HR) remote sensing imagery is becoming increasingly widespread. However, current remote sensing foundation models are ill-suited to such inputs: full-image encoding exhausts token and memory budgets, whil...","url_abs":"https://arxiv.org/abs/2511.20460","url_pdf":"https://arxiv.org/pdf/2511.20460v2","authors":"[\"Yunqi Zhou\",\"Chengjie Jiang\",\"Chun Yuan\",\"Jing Li\"]","published":"2025-11-25T16:25:54Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Generative Adversarial Network\"]","has_code":false}