UHR-Micro: Diagnosing and Mitigating the Resolution Illusion in Earth Observation VLMs
- URL: http://arxiv.org/abs/2605.12237v1
- Date: Tue, 12 May 2026 15:07:30 GMT
- Title: UHR-Micro: Diagnosing and Mitigating the Resolution Illusion in Earth Observation VLMs
- Abstract summary: Vision-Language Models (VLMs) operate on ultra-high-resolution (UHR) Earth observation imagery.<n>These models are vulnerable to a severe scale mismatch between large-scale scene context and micro-scale targets.<n>We benchmark this challenge with UHR-Micro, a benchmark comprising 11,253 instructions grounded in 1,212 UHR images.
- Score: 40.3198846405438
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Vision-Language Models (VLMs) increasingly operate on ultra-high-resolution (UHR) Earth observation imagery, yet they remain vulnerable to a severe scale mismatch between large-scale scene context and micro-scale targets. We refer to this empirical gap as a "resolution illusion": higher input resolution provides the appearance of richer visual detail, but does not necessarily yield reliable perception of spatially small, task-relevant evidence. To benchmark this challenge, we introduce UHR-Micro, a benchmark comprising 11,253 instructions grounded in 1,212 UHR images, designed to evaluate VLMs at the spatial limits of native Earth observation imagery. UHR-Micro spans diverse micro-target scales, context requirements, task families, and visual conditions, and provides diagnostic annotations that support controlled evaluation and fine-grained error attribution. Experiments with representative high-resolution VLMs show substantial failures in spatial grounding and evidence parsing, despite access to high-resolution inputs. Further analysis suggests that these failures are not fully resolved by increasing model capacity, but are closely tied to insufficient guidance in locating and using task-relevant micro-evidence. Motivated by this finding, we propose Micro-evidence Active Perception (MAP), a reference agent that decomposes queries into evidence-seeking steps, actively inspects candidate regions, and grounds its answers in localized observations. MAP-Agent improves micro-level perception by making high-resolution reasoning evidence-centered rather than image-centered. Together, UHR-Micro and MAP-Agent provide a diagnostic platform for evaluating, understanding, and advancing high-resolution reasoning in Earth observation VLMs. Datasets and source code were released at https://github.com/MiliLab/UHR-Micro.
Related papers
- Benchmarking Vision-Language Models for Microscopic Plant Image Understanding [34.461133608619015]
We present PlantMicro, a benchmark for evaluating vision-language models (VLMs) in microscopic plant imagery.<n>We design a set of complementary tasks that capture different facets of microscopic image understanding.<n>Experiments show that current VLMs struggle with fine-grained recognition and biologically grounded reasoning.
arXiv Detail & Related papers (2026-06-21T13:39:23Z) - Benchmarking Large Vision-Language Models on Fine-Grained Image Tasks: From Evaluation to Diagnosis [63.49939245944781]
We introduce FG-BMK, a comprehensive fine-grained evaluation benchmark containing 1.01 million questions and 0.28 million images.<n>We find that current LVLMs remain inadequate fine-grained recognizers, with failures arising from intertwined bottlenecks in visual representations, semantic grounding, modality alignment, and category-level knowledge.
arXiv Detail & Related papers (2026-06-17T13:23:26Z) - UltraVR: A Diagnostic Ultra-Resolution Image-VQA Benchmark for Evidence-Grounded Reasoning [32.37070955496229]
We introduce UltraVR, a diagnostic benchmark for evidence-grounded visual reasoning over ultra-resolution images.<n>Each instance includes a structured ground-truth chain of thought with step-level questions, intermediate answers, and reasoning labels.<n>We show that current models remain far from reliable on ultra-resolution reasoning.
arXiv Detail & Related papers (2026-06-04T01:51:16Z) - Micro-Defects Expose Macro-Fakes: Detecting AI-Generated Images via Local Distributional Shifts [48.75898711300078]
generative models can produce images that appear highly realistic, raising challenges in distinguishing real and AI-generated images.<n>We propose Micro-Defects expose Macro-Fakes (MDMF), a local distribution-aware detection framework that amplifies micro-scale statistical irregularities into macro-level distributional discrepancies.<n>Our theory-grounded analysis shows that patch-wise modeling yields provably larger discrepancies when localized forensic signals are present in generated images.
arXiv Detail & Related papers (2026-05-10T03:44:09Z) - ESPIRE: A Diagnostic Benchmark for Embodied Spatial Reasoning of Vision-Language Models [37.686748605290795]
ESPIRE is a diagnostic benchmark for embodied spatial reasoning.<n>It physically grounds vision-language models and evaluates them on spatial-reasoning-centric robotic tasks.<n>We decompose each task into localization and execution, and frame both as generative problems.
arXiv Detail & Related papers (2026-03-13T14:43:00Z) - Rethinking Evaluation of Infrared Small Target Detection [105.59753496831739]
This paper introduces a hybrid-level metric incorporating pixel- and target-level performance, proposing a systematic error analysis method, and emphasizing the importance of cross-dataset evaluation.<n>An open-source toolkit has be released to facilitate standardized benchmarking.
arXiv Detail & Related papers (2025-09-21T02:45:07Z) - Beyond Semantics: Rediscovering Spatial Awareness in Vision-Language Models [13.768090541138571]
Vision Language Models (VLMs) excel at identifying and describing objects but often fail at spatial reasoning.<n>Our analysis reveals a key imbalance: vision token embeddings have much larger norms than text tokens.<n>Tools uncover that vision tokens and system prompts dominate attention.
arXiv Detail & Related papers (2025-03-21T17:51:14Z) - SegSub: Evaluating Robustness to Knowledge Conflicts and Hallucinations in Vision-Language Models [6.52323086990482]
Vision language models (VLM) demonstrate sophisticated multimodal reasoning yet are prone to hallucination when confronted with knowledge conflicts.<n>This research introduces segsub, a framework for applying targeted image perturbations to investigate VLM resilience against knowledge conflicts.
arXiv Detail & Related papers (2025-02-19T00:26:38Z) - Ensemble Learning for Microbubble Localization in Super-Resolution Ultrasound [1.912429179274357]
Super-resolution ultrasound (SR-US) is a powerful imaging technique for capturing microvasculature and blood flow at high spatial resolution.<n>In this paper, we explore the potential of ensemble learning techniques to enhance microbubble (MB) localization.
arXiv Detail & Related papers (2024-11-11T21:26:36Z) - Improving Vision Anomaly Detection with the Guidance of Language
Modality [64.53005837237754]
This paper tackles the challenges for vision modality from a multimodal point of view.
We propose Cross-modal Guidance (CMG) to tackle the redundant information issue and sparse space issue.
To learn a more compact latent space for the vision anomaly detector, CMLE learns a correlation structure matrix from the language modality.
arXiv Detail & Related papers (2023-10-04T13:44:56Z) - Video-based Facial Micro-Expression Analysis: A Survey of Datasets,
Features and Algorithms [52.58031087639394]
micro-expressions are involuntary and transient facial expressions.
They can provide important information in a broad range of applications such as lie detection, criminal detection, etc.
Since micro-expressions are transient and of low intensity, their detection and recognition is difficult and relies heavily on expert experiences.
arXiv Detail & Related papers (2022-01-30T05:14:13Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.