迈向遥感中的可信推理：一种基于感知的地理空间思维链用于视觉-语言模型
Towards Faithful Reasoning in Remote Sensing: A Perceptually-Grounded GeoSpatial Chain-of-Thought for Vision-Language Models
AI总结：
本研究提出Geo-CoT框架，通过两阶段策略提升遥感VLMs的推理能力，实现可验证的多步骤分析，显著优于现有模型。
摘要：
遥感中的视觉-语言模型（VLMs）在复杂分析任务中常常失败，这种局限性源于其端到端训练范式，这种范式绕过了关键的推理步骤，导致输出不可验证。为了解决这一局限性，我们引入了感知导向的地理空间思维链（Geo-CoT），一种将遥感分析建模为可验证、多步骤过程的框架。我们通过一种两阶段对齐策略来植入这一分析过程，利用Geo-CoT380k，首个大规模的结构化Geo-CoT推理数据集。该策略首先使用监督微调（SFT）来植入基础认知架构，然后利用分组奖励策略优化（GRPO）来优化模型的推理策略，使其更接近事实正确性。最终模型RSThinker能够输出最终答案及其可验证的分析轨迹。这种能力使其在广泛的任务上均显著优于最先进模型。本研究发布时公开我们的Geo-CoT380k数据集和RSThinker模型，为从模糊感知向结构化、可验证推理的地球观测提供了具体路径。
英文摘要：
Vision-Language Models (VLMs) in remote sensing often fail at complex analytical tasks, a limitation stemming from their end-to-end training paradigm that bypasses crucial reasoning steps and leads to unverifiable outputs. To address this limitation, we introduce the Perceptually-Grounded Geospatial Chain-of-Thought (Geo-CoT), a framework that models remote sensing analysis as a verifiable, multi-step process. We instill this analytical process through a two-stage alignment strategy, leveraging Geo-CoT380k, the first large-scale dataset of structured Geo-CoT rationales. This strategy first employs supervised fine-tuning (SFT) to instill the foundational cognitive architecture, then leverages Group Reward Policy Optimization (GRPO) to refine the model's reasoning policy towards factual correctness. The resulting model, RSThinker, outputs both a final answer and its justifying, verifiable analytical trace. This capability yields dominant performance, significantly outperforming state-of-the-art models across a comprehensive range of tasks. The public release of our Geo-CoT380k dataset and RSThinker model upon publication serves as a concrete pathway from opaque perception towards structured, verifiable reasoning for Earth Observation.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
