mlsanityData to 7 Oct 2026

GeoEyes: On-Demand Visual Focusing for Evidence-Grounded Understanding of Ultra-High-Resolution Remote Sensing Imagery

arXiv 2602.14201 ↗First posted 15 Feb 2026cs.CV cs.AIcited by 9

On the map: Multimodal LLMs / vision-language models → Embodied and 3D vision-language models

Abstract (arXiv)

The "thinking-with-images" paradigm enables multimodal large language models (MLLMs) to actively explore visual scenes via zoom-in tools. This is essential for ultra-high-resolution (UHR) remote sensing VQA, where task-relevant cues are sparse and tiny. However, we observe a consistent failure mode in existing zoom-enabled MLLMs: Tool Usage Homogenization, where tool calls collapse into task-agnostic patterns, limiting effective evidence acquisition. To address this, we propose GeoEyes, a staged training framework consisting of (1) a cold-start SFT dataset, UHR Chain-of-Zoom (UHR-CoZ), which covers diverse zooming regimes, and (2) an agentic reinforcement learning method, AdaZoom-GRPO, that explicitly rewards evidence gain and answer improvement during zoom interactions. The resulting model learns on-demand zooming with proper stopping behavior and achieves substantial improvements on UHR remote sensing benchmarks, with 54.23% accuracy on XLRS-Bench.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Placed in Embodied and 3D vision-language models, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 40 of its 48 references that are in the map. First posted on arXiv 15 Feb 2026.

Builds on

40 of 48 references.

Closest ancestors · 17

Same direction, newest first

  1. 2026Thinking with Blueprints: Assisting Vision-Language Models in Spatial Reasoning via Structured Object Representation5 Jan
  2. 2025SpaceTools: Tool-Augmented Spatial Reasoning via Double Interactive RL3 Dec
  3. 2025GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning1 Jul
  4. 2025EarthMind: Leveraging Cross-Sensor Data for Advanced Earth Observation Interpretation with a Unified Multimodal LLM2 Jun
  5. 2025GeoLLaVA-8K: Scaling Remote-Sensing Multimodal Large Language Models to 8K Resolution27 May
  6. 2025DeepEyes: Incentivizing "Thinking with Images" via Reinforcement Learning20 May
  7. 2025XLRS-Bench: Could Your Multimodal LLMs Understand Extremely Large Ultra-High-Resolution Remote Sensing Imagery?31 Mar
  8. 2025When Large Vision-Language Model Meets Large Remote Sensing Imagery: Coarse-to-Fine Text-Guided Token Pruning10 Mar
  9. 2024EarthDial: Turning Multi-sensory Earth Observations to Interactive Dialogues19 Dec
  10. 2024RSUniVLM: A Unified Vision Language Model for Remote Sensing via Granularity-oriented Mixture of Experts7 Dec
  11. 2024LHRS-Bot-Nova: Improved Multimodal Large Language Model for Remote Sensing Vision-Language Interpretation14 Nov
  12. 2024EarthMarker: A Visual Prompting Multi-modal Large Language Model for Remote Sensing18 Jul
5 more
  1. 2024VHM: Versatile and Honest Vision Language Model for Remote Sensing Image Analysis29 Mar
  2. 2024EarthGPT: A Universal Multi-modal Large Language Model for Multi-sensor Image Comprehension in Remote Sensing Domain30 Jan
  3. 2024SkyEyeGPT: Unifying Remote Sensing Vision-Language Tasks via Instruction Tuning with Large Language Model18 Jan
  4. 2023GeoChat: Grounded Large Vision-Language Model for Remote Sensing24 Nov
  5. 2023RSGPT: A Remote Sensing Vision Language Model and Benchmark28 Jul

Other references · 23

  1. 2023Visual Instruction Tuningcited by 7,792
  2. 2025Qwen2.5-VL Technical Reportcited by 5,418
  3. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,151
  4. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,655
  5. 2025Qwen3-VL Technical Reportcited by 2,534
  6. 2025InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Modelscited by 1,769
  7. 2025Pixel Reasoner: Incentivizing Pixel-Space Reasoning with Curiosity-Driven Reinforcement Learningcited by 267
  8. 2024Enhancing the Reasoning Ability of Multimodal Large Language Models via Mixed Preference Optimizationcited by 238
  9. 2024Reinforcement learningcited by 236
  10. 2020RSVQA: Visual Question Answering for Remote Sensing Datacited by 197
  11. 2025Thinking with Images for Multimodal Reasoning: Foundations, Methods, and Future Frontierscited by 193
  12. 2025DeepEyesV2: Toward Agentic Multimodal Modelcited by 119
11 more
  1. 2024ZoomEye: Enhancing Multimodal LLMs with Human-Like Zooming Capabilities through Tree-Based Image Explorationcited by 99
  2. 2025Zebra-CoT: A Dataset for Interleaved Vision Language Reasoningcited by 78
  3. 2025DyFo: A Training-Free Dynamic Focus Visual Search for Enhancing LMMs in Fine-Grained Visual Understandingcited by 47
  4. 2024Large Language Models for Captioning and Retrieving Remote Sensing Imagescited by 30
  5. 2025ZoomEarth: Active Perception for Ultra-High-Resolution Geospatial Vision-Language Taskscited by 19
  6. 2025Thinking With Bounding Boxes: Enhancing Spatio-Temporal Video Grounding via Reinforcement Fine-Tuningcited by 11
  7. 2025Deep But Reliable: Advancing Multi-turn Reasoning for Thinking with Imagescited by 8
  8. 2026Thinking with Map: Reinforced Parallel Map-Augmented Agent for Geolocalizationcited by 8
  9. 2025VisReason: A Large-Scale Dataset for Visual Chain-of-Thought Reasoningcited by 5
  10. 2026EarthVL: A Progressive Earth Vision-Language Understanding and Generation Frameworkcited by 5
  11. 2025VICoT-Agent: A Vision-Interleaved Chain-of-Thought Framework for Interpretable Multimodal Reasoning and Scalable Remote Sensing Analysiscited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 9.

9 papers.

  1. 2026Artifact-Bench: Evaluating MLLMs on Detecting and Assessing the Artifacts of AI-Generated Videoscited by 4
  2. 2026WeaveEarth: Structured Evidence Construction and Reasoning for Training-Free UHR Remote Sensing Understandingcited by 2
  3. 2026Thinking With Tools, Not With Pixels: Tool Calls as Text Scaffolds for Visual Reasoningcited by 2
  4. 2026Beyond GSD-as-Token: Continuous Scale Conditioning for Remote Sensing VLMscited by 1
  5. 2026UHR-Micro: Diagnosing and Mitigating the Resolution Illusion in Earth Observation VLMscited by 1
  6. 2026Beyond Zooming: Learning Multi-Tool Visual Reasoning for Ultra-High-Resolution Remote Sensingcited by 1
  7. 2026Decoding the Delta: Unifying Remote Sensing Change Detection and Understanding with Multimodal Large Language Modelscited by 0
  8. 2026More with Less: a Large Scale Remote Sensing VLM with a Simple Recipecited by 0
  9. 2026RS-OPSD: Reliable Privileged On-Policy-Self-Distillation for Ultra-High-Resolution Remote Sensing VQAcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026The Earth in One Gaze: Training-Free Active Focus for UHR Remote Sensing Understanding
  2. 2026Text Before Vision: Staged Knowledge Injection Matters for Agentic RLVR in Ultra-High-Resolution Remote Sensing Understanding
  3. 2026GeoVista: Visually Grounded Active Perception for Vision-Language Understanding of Ultra-High-Resolution Remote Sensing Images
  4. 2026SA-GEM: Scale-Adaptive and Geospatial Evidence-Modulated Token Pruning for Efficient Remote Sensing Large Vision-Language Models
  5. 2025Asking like Socrates: Socrates helps VLMs understand remote sensing images
  6. 2025A Benchmark for Ultra-High-Resolution Remote Sensing MLLMs
  7. 2025GeoZero: Incentivizing Reasoning from Scratch on Geospatial Scenes
  8. 2025RemoteReasoner: Towards Unifying Geospatial Reasoning Workflow