Reasoning with Pixel-level Precision: QVLM Architecture and SQuID Dataset for Quantitative Geospatial Analytics

• URL: http://arxiv.org/abs/2601.13401v1
• Date: Mon, 19 Jan 2026 21:14:34 GMT
• Title: Reasoning with Pixel-level Precision: QVLM Architecture and SQuID Dataset for Quantitative Geospatial Analytics
• Abstract summary: Current Vision-Language Models (VLMs) fail at quantitative spatial reasoning because their architectures destroy pixel-level information required for counting and measurements.<n>We present two contributions to address this fundamental limitation. First, we introduce SQuID, a benchmark of 2,000 satellite image Question-Answer pairs with both numerical range and categorical answers.<n>Second, we propose QVLM, a code-generation architecture that maintains pixel precision by decoupling language understanding from visual analysis.
• Score: 0.2291770711277359
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Current Vision-Language Models (VLMs) fail at quantitative spatial reasoning because their architectures destroy pixel-level information required for counting and measurements. Vision encoders compress images through patch embeddings, reducing spatial indexing and losing the precise pixel-level tracking required for accurate counting. We present two contributions to address this fundamental limitation. First, we introduce SQuID (Satellite Quantitative Intelligence Dataset), a benchmark of 2,000 satellite image Question-Answer pairs with both numerical range and categorical answers, designed to evaluate quantitative spatial reasoning. The dataset spans three difficulty tiers with annotations automatically generated from human labels and their learned variability. Second, we propose QVLM (Quantitative Vision-Language Model), a code-generation architecture that maintains pixel precision by decoupling language understanding from visual analysis. Instead of encoding images into embeddings, QVLM generates executable code that first calls a segmentation model to obtain pixel-level masks, then operates directly on these masks, preserving spatial indexing throughout the reasoning process. Our experiments show that QVLM using GPT-5 as coder achieves 42.0% accuracy on SQuID compared to 28.1% for a VLM prompted with image-question pairs. Our work reveals that, for quantitative spatial reasoning, architectural decoupling enables better accuracy on quantitative tasks.

Related papers

• Dual-Pathway Geometry-Aware MLLM for Spatial Intelligence [63.11502936651325]
  We propose GAMSI, a dual-pathway Geometry-Aware MLLM for Spatial Intelligence.<n>It takes only RGB images as input while internalizing both forms of geometric prior within a unified autoregressive backbone.<n>GAMSI achieves state-of-the-art performance on seven spatial intelligence benchmarks.
  arXiv  Detail & Related papers  (2026-05-25T01:33:19Z)
• Geospatial-Temporal Sensemaking of Remote Sensing Activity Detections with Multimodal Large Language Model [3.569867801312132]
  We introduce SMARTHC-VQA, a visual question answering dataset derived from the IARPA SMART Heavy Construction dataset.<n>The dataset transforms construction-type labels, temporal-phase labels, geographic metadata, and observation relationships into natural language question-answer triplets.<n>We describe an implemented multi-image MLLM training framework based on LVA-NeXT Mistral-7B, adapted to accept multiple dated image inputs and train on metadata-derived VQA examples.
  arXiv  Detail & Related papers  (2026-05-11T15:42:09Z)
• PinpointQA: A Dataset and Benchmark for Small Object-Centric Spatial Understanding in Indoor Videos [23.951493516327016]
  PinpointQA is the first dataset and benchmark for small object-centric spatial understanding in indoor videos.<n>It comprises 1,024 scenes and 10,094 QA pairs organized into four progressively challenging tasks.<n>Experiments reveal a consistent capability gap along the progressive chain, with SSP remaining particularly difficult.
  arXiv  Detail & Related papers  (2026-04-10T05:53:47Z)
• Nüwa: Mending the Spatial Integrity Torn by VLM Token Pruning [82.39668822222386]
  Vision token pruning has proven to be an effective acceleration technique for the efficient Vision Language Model (VLM)<n>We propose $textNwa$, a two-stage token pruning framework that enables efficient feature aggregation while maintaining spatial integrity.<n>Experiments demonstrate that $textNwa$ achieves SOTA performance on multiple VQA benchmarks (from 94% to 95%) and yields substantial improvements on visual grounding tasks (from 7% to 47%)
  arXiv  Detail & Related papers  (2026-02-03T00:51:03Z)
• EarthVL: A Progressive Earth Vision-Language Understanding and Generation Framework [37.80969053377522]
  A progressive Earth vision-language understanding framework is proposed, including a multi-task dataset (EarthVLSet) and a semantic-guided network (EarthVLNet)<n>EarthVLSet includes 10.9k sub-meter resolution remote sensing images, land-cover masks, and 761.5k textual pairs.<n>In an object-centric way, EarthVLNet is proposed to progressively achieve semantic segmentation, relational reasoning, and comprehensive understanding.
  arXiv  Detail & Related papers  (2026-01-06T07:41:44Z)
• Video2Layout: Recall and Reconstruct Metric-Grounded Cognitive Map for Spatial Reasoning [19.549136366694572]
  Video2 is a framework for reconstructing metric-grounded spatial layouts from video.<n>The framework employs continuous object boundary coordinates to quantify inter-object physical and object size.<n>Our model, V2LO-7B, achieves an average improvement of 4.92% over the model trained on grid maps.
  arXiv  Detail & Related papers  (2025-11-20T08:57:14Z)
• SD-VLM: Spatial Measuring and Understanding with Depth-Encoded Vision-Language Models [75.64836077468722]
  Vision language models (VLMs) excel in 2D semantic visual understanding, but their ability to quantitatively reason about 3D spatial relationships remains under-explored.<n>We propose SD-VLM, a novel framework that significantly enhances fundamental spatial perception abilities of VLMs.<n>We have trained SD-VLM, a strong generalist VLM which shows superior quantitative spatial measuring and understanding capability.
  arXiv  Detail & Related papers  (2025-09-22T12:08:12Z)
• GEOBench-VLM: Benchmarking Vision-Language Models for Geospatial Tasks [84.86699025256705]
  We present GEOBench-VLM, a benchmark specifically designed to evaluate Vision-Language Models (VLMs) on geospatial tasks.<n>Our benchmark features over 10,000 manually verified instructions and spanning diverse visual conditions, object types, and scales.<n>We evaluate several state-of-the-art VLMs to assess performance on geospatial-specific challenges.
  arXiv  Detail & Related papers  (2024-11-28T18:59:56Z)
• A Hitchhikers Guide to Fine-Grained Face Forgery Detection Using Common Sense Reasoning [9.786907179872815]
  The potential of vision and language remains underexplored in face forgery detection. There is a need for a methodology that converts face forgery detection to a Visual Question Answering (VQA) task. We propose a multi-staged approach that diverges from the traditional binary decision paradigm to address this gap.
  arXiv  Detail & Related papers  (2024-10-01T08:16:40Z)
• Pluralistic Salient Object Detection [108.74650817891984]
  We introduce pluralistic salient object detection (PSOD), a novel task aimed at generating multiple plausible salient segmentation results for a given input image. We present two new SOD datasets "DUTS-MM" and "DUS-MQ", along with newly designed evaluation metrics.
  arXiv  Detail & Related papers  (2024-09-04T01:38:37Z)
• Coarse Correspondences Boost Spatial-Temporal Reasoning in Multimodal Language Model [51.83436609094658]
  We introduce Coarse Correspondences, a simple lightweight method that enhances MLLMs' spatial-temporal reasoning with 2D images as input. Our method uses a lightweight tracking model to identify primary object correspondences between frames in a video or across different image viewpoints. We demonstrate that this simple training-free approach brings substantial gains to GPT4-V/O consistently on four benchmarks.
  arXiv  Detail & Related papers  (2024-08-01T17:57:12Z)
• SpatialBot: Precise Spatial Understanding with Vision Language Models [12.67089704185187]
  Vision Language Models (VLMs) have achieved impressive performance in 2D image understanding.<n>They are still struggling with spatial understanding which is the foundation of Embodied AI.<n>In this paper, we propose SpatialBot for better spatial understanding by feeding both RGB and depth images.
  arXiv  Detail & Related papers  (2024-06-19T15:41:30Z)
• SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities [59.39858959066982]
  understanding and reasoning about spatial relationships is a fundamental capability for Visual Question Answering (VQA) and robotics. We develop an automatic 3D spatial VQA data generation framework that scales up to 2 billion VQA examples on 10 million real-world images. By training a VLM on such data, we significantly enhance its ability on both qualitative and quantitative spatial VQA.
  arXiv  Detail & Related papers  (2024-01-22T18:01:01Z)
• ClusTR: Exploring Efficient Self-attention via Clustering for Vision Transformers [70.76313507550684]
  We propose a content-based sparse attention method, as an alternative to dense self-attention. Specifically, we cluster and then aggregate key and value tokens, as a content-based method of reducing the total token count. The resulting clustered-token sequence retains the semantic diversity of the original signal, but can be processed at a lower computational cost.
  arXiv  Detail & Related papers  (2022-08-28T04:18:27Z)

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.