Learning to Wander: Improving the Global Image Geolocation Ability of LMMs via Actionable Reasoning

2026-03-11

AI Summary

Current large multimodal language models lack the ability to integrate world knowledge with actionable reasoning for image geolocation tasks. To address this limitation, this work proposes GeoAoT, a novel framework that reframes geolocation from static recognition to interactive exploration by generating executable actions—such as rotation and movement—to actively reduce localization uncertainty. The study also introduces WanderBench, the first embodied benchmark for global geolocation, along with a new evaluation paradigm that jointly assesses localization accuracy and difficulty-aware question-answering capability. Experiments across 19 large models demonstrate that GeoAoT significantly improves fine-grained localization accuracy and generalization in dynamic environments, validating the effectiveness of the proposed approach.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Natural Language Processing: Language Grounding & Multi-modal NLP
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services

Abstract

Geolocation, the task of identifying the geographic location of an image, requires abundant world knowledge and complex reasoning abilities. Though advanced large multimodal models (LMMs) have shown superior aforementioned capabilities, their performance on the geolocation task remains unexplored. To this end, we introduce \textbf{WanderBench}, the first open access global geolocation benchmark designed for actionable geolocation reasoning in embodied scenarios. WanderBench contains over 32K panoramas across six continents, organized as navigable graphs that enable physical actions such as rotation and movement, transforming geolocation from static recognition into interactive exploration. Building on this foundation, we propose \textbf{GeoAoT} (Action of Thought), a \underline{Geo}location framework with \underline{A}ction of \underline{T}hough, which couples reasoning with embodied actions. Instead of generating textual reasoning chains, GeoAoT produces actionable plans such as, approaching landmarks or adjusting viewpoints, to actively reduce uncertainty. We further establish an evaluation protocol that jointly measures geolocation accuracy and difficulty-aware geolocation questioning ability. Experiments on 19 large multimodal models show that GeoAoT achieves superior fine-grained localization and stronger generalization in dynamic environments. WanderBench and GeoAoT define a new paradigm for actionable, reasoning driven geolocation in embodied visual understanding.

Problem

• geolocation
• large multimodal models
• embodied reasoning
• actionable exploration
• visual understanding

Innovation

• actionable reasoning
• embodied geolocation
• navigable panoramas
• GeoAoT
• WanderBench