GeoSolver: Scaling Test-Time Reasoning in Remote Sensing with Fine-Grained Process Supervision

• URL: http://arxiv.org/abs/2603.09551v1
• Date: Tue, 10 Mar 2026 11:59:05 GMT
• Title: GeoSolver: Scaling Test-Time Reasoning in Remote Sensing with Fine-Grained Process Supervision
• Abstract summary: We introduce Geor, a framework that transitions remote sensing reasoning toward verifiable, process-supervised learning.<n>We train GeoPRM, a token-level process reward model (PRM), that provides granular faithfulness feedback.<n>To effectively leverage these verification signals, we propose Process-Aware Tree-GRPO.<n>Our resulting model, Geor-9B, integrates state-of-the-art performance across diverse remote sensing benchmarks.
• Score: 10.62942603434055
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: While Vision-Language Models (VLMs) have significantly advanced remote sensing interpretation, enabling them to perform complex, step-by-step reasoning remains highly challenging. Recent efforts to introduce Chain-of-Thought (CoT) reasoning to this domain have shown promise, yet ensuring the visual faithfulness of these intermediate steps remains a critical bottleneck. To address this, we introduce GeoSolver, a novel framework that transitions remote sensing reasoning toward verifiable, process-supervised reinforcement learning. We first construct Geo-PRM-2M, a large-scale, token-level process supervision dataset synthesized via entropy-guided Monte Carlo Tree Search (MCTS) and targeted visual hallucination injection. Building upon this dataset, we train GeoPRM, a token-level process reward model (PRM) that provides granular faithfulness feedback. To effectively leverage these verification signals, we propose Process-Aware Tree-GRPO, a reinforcement learning algorithm that integrates tree-structured exploration with a faithfulness-weighted reward mechanism to precisely assign credit to intermediate steps. Extensive experiments demonstrate that our resulting model, GeoSolver-9B, achieves state-of-the-art performance across diverse remote sensing benchmarks. Crucially, GeoPRM unlocks robust Test-Time Scaling (TTS). Serving as a universal geospatial verifier, it seamlessly scales the performance of GeoSolver-9B and directly enhances general-purpose VLMs, highlighting its remarkable cross-model generalization.

Related papers

• GeoSearcher: Anchor-Guided Progressive Reasoning for Remote Sensing Visual Grounding with Process Supervision [12.717046652815027]
  High-resolution remote sensing images usually cover large-scale scenes, where targets are often extremely small and surrounded by numerous visually similar distractors.<n>Existing MLLM-based methods usually formulate RSVG as one-step coordinate generation, which may lead to unstable predictions for small-object localization and complex queries.<n>We propose GeoSearcher, which reformulates RSVG as an anchor-guided progressive reasoning process and realizes it through two coupled stages.<n>In ACR-SFT, anchor-centric reasoning data are used to teach the model to represent key visual clues as anchors and progressively integrate location, relational, and attribute
  arXiv  Detail & Related papers  (2026-07-01T15:12:51Z)
• StainFlow: Entity-Stain Tracking and Evidence Linking for Process Rewards in GUI Agents [67.03593791535786]
  Reinforcement Learning (RL) has become a promising approach for improving GUI Agents in long-horizon, digital environments.<n>To mitigate this issue, recent studies introduce Process Reward Models (PRMs)<n>PRMs provide finer-grained training feedback through global milestone verification or local step-level evaluation.<n>We propose StainFlow, an entity-stain-flow process reward model for GUI Agents.
  arXiv  Detail & Related papers  (2026-06-05T08:17:28Z)
• GeoAgentBench: A Dynamic Execution Benchmark for Tool-Augmented Agents in Spatial Analysis [16.604040127938955]
  We introduce GeoAgentBench (GABench), a dynamic and interactive evaluation benchmark tailored for tool-augmented GIS agents.<n>GABench provides a realistic sandbox integrating 117 atomic GIS tools, encompassing 53 typical spatial analysis tasks.<n>We develop a novel agent architecture, Plan-and-React, that mimics expert cognitive anomalies by decoupling global orchestration from step-wise reactive execution.
  arXiv  Detail & Related papers  (2026-04-15T13:55:34Z)
• Understanding by Reconstruction: Reversing the Software Development Process for LLM Pretraining [66.89012795621349]
  Large Language Models (LLMs) often struggle with the deep, long-horizon reasoning required for complex software engineering.<n>We propose a novel paradigm: understanding via reconstruction.<n>We introduce a framework that synthesizes latent agentic trajectories using a multi-agent simulation.
  arXiv  Detail & Related papers  (2026-03-11T09:23:20Z)
• SpotAgent: Grounding Visual Geo-localization in Large Vision-Language Models through Agentic Reasoning [31.665287327579026]
  SpotAgent is a framework that formalizes geo-localization into an agentic reasoning process.<n>It actively explores and verifies visual cues by leveraging external tools (e.g., web search, maps) through a ReAct diagram.<n>It achieves state-of-the-art performance, effectively mitigating hallucinations while delivering precise and verifiable geo-localization.
  arXiv  Detail & Related papers  (2026-02-10T06:57:12Z)
• GeoDiT: A Diffusion-based Vision-Language Model for Geospatial Understanding [14.436063587920005]
  We introduce GeoDiT, the first diffusion-based vision-language model tailored for the geospatial domain.<n>It achieves significant gains in image captioning, visual grounding, and multi-object detection.<n>Our work validates that aligning the generative process with the data's intrinsic structure is key to unlocking superior performance in complex geospatial analysis.
  arXiv  Detail & Related papers  (2025-12-02T07:59:46Z)
• GeoVista: Web-Augmented Agentic Visual Reasoning for Geolocalization [53.080882980294795]
  Current research on agentic visual reasoning enables deep multimodal understanding but primarily focuses on image manipulation tools.<n>In this work, we revisit the geolocalization task, which requires not only nuanced visual grounding but also web search to confirm or refine hypotheses.<n>Since existing geolocalization benchmarks fail to meet the need for high-resolution imagery and the localization challenge for deep agentic reasoning, we curate GeoBench.<n>We propose GeoVista, an agentic model that seamlessly integrates tool invocation within the reasoning loop, including an image-zoom-in tool to magnify regions of interest and a web-search tool to retrieve related
  arXiv  Detail & Related papers  (2025-11-19T18:59:22Z)
• Detect Anything via Next Point Prediction [51.55967987350882]
  Rex- Omni is a 3B-scale MLLM that achieves state-of-the-art object perception performance.<n>On benchmarks like COCO and LVIS, Rex- Omni attains performance comparable to or exceeding regression-based models.
  arXiv  Detail & Related papers  (2025-10-14T17:59:54Z)
• Towards Faithful Reasoning in Remote Sensing: A Perceptually-Grounded GeoSpatial Chain-of-Thought for Vision-Language Models [8.021952962029165]
  Vision-Language Models (VLMs) in remote sensing often fail at complex analytical tasks.<n>We introduce the Perceptually-Grounded Geospatial Chain-of-Thought (Geo-CoT)<n>Geo-CoT is a framework that models remote sensing analysis as a verifiable, multi-step process.
  arXiv  Detail & Related papers  (2025-09-26T11:34:42Z)
• Enhancing the Geometric Problem-Solving Ability of Multimodal LLMs via Symbolic-Neural Integration [57.95306827012784]
  We propose GeoGen, a pipeline that can automatically generate step-wise reasoning paths for geometry diagrams.<n>By leveraging the precise symbolic reasoning, textbfGeoGen produces large-scale, high-quality question-answer pairs.<n>We train textbfGeoLogic, a Large Language Model (LLM), using synthetic data generated by GeoGen.
  arXiv  Detail & Related papers  (2025-04-17T09:13:46Z)
• Thinking Longer, Not Larger: Enhancing Software Engineering Agents via Scaling Test-Time Compute [61.00662702026523]
  We propose a unified Test-Time Compute scaling framework that leverages increased inference-time instead of larger models.<n>Our framework incorporates two complementary strategies: internal TTC and external TTC.<n>We demonstrate our textbf32B model achieves a 46% issue resolution rate, surpassing significantly larger models such as DeepSeek R1 671B and OpenAI o1.
  arXiv  Detail & Related papers  (2025-03-31T07:31:32Z)
• Without Paired Labeled Data: End-to-End Self-Supervised Learning for Drone-view Geo-Localization [20.603433987118837]
  Drone-view Geo-Localization (DVGL) aims to achieve accurate localization of drones by retrieving the most relevant GPS-tagged satellite images.<n>Existing methods heavily rely on strictly pre-paired drone-satellite images for supervised learning.<n>We propose a novel end-to-end self-supervised learning method with a shallow backbone network.
  arXiv  Detail & Related papers  (2025-02-17T02:53:08Z)
• Chain-of-Retrieval Augmented Generation [91.02950964802454]
  This paper introduces an approach for training o1-like RAG models that retrieve and reason over relevant information step by step before generating the final answer.<n>Our proposed method, CoRAG, allows the model to dynamically reformulate the query based on the evolving state.
  arXiv  Detail & Related papers  (2025-01-24T09:12:52Z)
• Salient Object Detection in Optical Remote Sensing Images Driven by Transformer [69.22039680783124]
  We propose a novel Global Extraction Local Exploration Network (GeleNet) for Optical Remote Sensing Images (ORSI-SOD) Specifically, GeleNet first adopts a transformer backbone to generate four-level feature embeddings with global long-range dependencies. Extensive experiments on three public datasets demonstrate that the proposed GeleNet outperforms relevant state-of-the-art methods.
  arXiv  Detail & Related papers  (2023-09-15T07:14:43Z)

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.