Thinking with Map: Reinforced Parallel Map-Augmented Agent for Geolocalization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ji, Yuxiang, Wang, Yong, Ma, Ziyu, Hu, Yiming, Huang, Hailang, Hu, Xuecai, Chen, Guanhua, Wu, Liaoni, Chu, Xiangxiang
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917191429914624
author Ji, Yuxiang
Wang, Yong
Ma, Ziyu
Hu, Yiming
Huang, Hailang
Hu, Xuecai
Chen, Guanhua
Wu, Liaoni
Chu, Xiangxiang
author_facet Ji, Yuxiang
Wang, Yong
Ma, Ziyu
Hu, Yiming
Huang, Hailang
Hu, Xuecai
Chen, Guanhua
Wu, Liaoni
Chu, Xiangxiang
contents The image geolocalization task aims to predict the location where an image was taken anywhere on Earth using visual clues. Existing large vision-language model (LVLM) approaches leverage world knowledge, chain-of-thought reasoning, and agentic capabilities, but overlook a common strategy used by humans -- using maps. In this work, we first equip the model \textit{Thinking with Map} ability and formulate it as an agent-in-the-map loop. We develop a two-stage optimization scheme for it, including agentic reinforcement learning (RL) followed by parallel test-time scaling (TTS). The RL strengthens the agentic capability of model to improve sampling efficiency, and the parallel TTS enables the model to explore multiple candidate paths before making the final prediction, which is crucial for geolocalization. To evaluate our method on up-to-date and in-the-wild images, we further present MAPBench, a comprehensive geolocalization training and evaluation benchmark composed entirely of real-world images. Experimental results show that our method outperforms existing open- and closed-source models on most metrics, specifically improving Acc@500m from 8.0\% to 22.1\% compared to \textit{Gemini-3-Pro} with Google Search/Map grounded mode.
format Preprint
id arxiv_https___arxiv_org_abs_2601_05432
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Thinking with Map: Reinforced Parallel Map-Augmented Agent for Geolocalization
Ji, Yuxiang
Wang, Yong
Ma, Ziyu
Hu, Yiming
Huang, Hailang
Hu, Xuecai
Chen, Guanhua
Wu, Liaoni
Chu, Xiangxiang
Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
The image geolocalization task aims to predict the location where an image was taken anywhere on Earth using visual clues. Existing large vision-language model (LVLM) approaches leverage world knowledge, chain-of-thought reasoning, and agentic capabilities, but overlook a common strategy used by humans -- using maps. In this work, we first equip the model \textit{Thinking with Map} ability and formulate it as an agent-in-the-map loop. We develop a two-stage optimization scheme for it, including agentic reinforcement learning (RL) followed by parallel test-time scaling (TTS). The RL strengthens the agentic capability of model to improve sampling efficiency, and the parallel TTS enables the model to explore multiple candidate paths before making the final prediction, which is crucial for geolocalization. To evaluate our method on up-to-date and in-the-wild images, we further present MAPBench, a comprehensive geolocalization training and evaluation benchmark composed entirely of real-world images. Experimental results show that our method outperforms existing open- and closed-source models on most metrics, specifically improving Acc@500m from 8.0\% to 22.1\% compared to \textit{Gemini-3-Pro} with Google Search/Map grounded mode.
title Thinking with Map: Reinforced Parallel Map-Augmented Agent for Geolocalization
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2601.05432