让模型像人一样看图推理、验证，提升地理定位准确性
Perceive to Hypothesize, Verify to Ground: An Agentic Reasoning Framework for Open-World Geo-Localization
- 分两步走：先根据图像生成假设，再用证据验证或修正
- 在真实用户签到数据上测试，定位准确率显著提升
- 适合需要可信推理的地理应用，如导航、城市规划
开放世界地理定位需模型通过多步推理和外部知识对模糊视觉线索进行判断。尽管大型视觉语言模型具备强大多模态推理能力，现有方法仍因缺乏显式证据验证而存在感知幻觉与上下文漂移问题。本文将地理定位重构为类人‘感知-验证’推理任务，提出GeoPAVE（地理定位感知与验证引擎）——一种双层代理框架，包含单次遍历生成感知假设，以及基于证据支持、反驳或修正决策的动作。为支持严格评估，我们构建了PAVED数据集，源自真实用户签到数据，包含多跳查询、多轮工具调用及结构化感知-验证轨迹。代码与数据已公开于 https://github.com/Arandinglv/GeoPAVE。
原文摘要 · Abstract (English)
Open-world geo-localization requires models to reason over ambiguous visual cues through multi-step reasoning and external knowledge grounding. While recent large vision-language models exhibit strong multimodal reasoning capabilities, existing approaches still suffer from perceptual hallucination and context drift due to the lack of explicit evidence-grounded verification. In this work, we reformulate geo-localization as a human-like perceive-then-verify reasoning problem and propose GeoPAVE (Geo-localization Perception-and-Verification-Engine), a bi-level agentic framework that contains perception-based hypothesis generation via single-pass rollouts and verification-based evidence grounding for decision actions: support, refute, and refine. To support rigorous evaluation, we further introduce PAVED, a novel dataset derived from real-world user check-in data, equipped with comprehensive reasoning trajectories featuring multi-hop queries, multi-round tool invocations, and structured perception-verification traces. The dataset and code are available at https://github.com/Arandinglv/GeoPAVE.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。