UHR-BAT: Budget-Aware Token Compression Vision-Language model for Ultra-High-Resolution Remote Sensing

arXiv cs.CV / 4/16/2026

💬 OpinionSignals & Early TrendsIdeas & Deep AnalysisModels & Research

共有:

Key Points

UHR-BAT is a budget-aware token compression framework for ultra-high-resolution remote sensing that aims to prevent the quadratic growth of visual tokens while preserving query-critical details.
The method uses text-guided, multi-scale importance estimation to select the most relevant visual tokens under a strict context budget.
It applies region-wise preserve-and-merge strategies to reduce redundancy among tokens without sacrificing information in small objects.
The authors report state-of-the-art performance on multiple benchmarks and indicate that the code will be released via GitHub.

Abstract

Ultra-high-resolution (UHR) remote sensing imagery couples kilometer-scale context with query-critical evidence that may occupy only a few pixels. Such vast spatial scale leads to a quadratic explosion of visual tokens and hinders the extraction of information from small objects. Previous works utilize direct downsampling, dense tiling, or global top-k pruning, which either compromise query-critical image details or incur unpredictable compute. In this paper, we propose UHR-BAT, a query-guided and region-faithful token compression framework to efficiently select visual tokens under a strict context budget. Specifically, we leverage text-guided, multi-scale importance estimation for visual tokens, effectively tackling the challenge of achieving precise yet low-cost feature extraction. Furthermore, by introducing region-wise preserve and merge strategies, we mitigate visual token redundancy, further driving down the computational budget. Experimental results show that UHR-BAT achieves state-of-the-art performance across various benchmarks. Code will be available at https://github.com/Yunkaidang/UHR.