# GeoGNN: Time Series Geo-Localization using Two-Tower Graph Neural Networks

Source: [GeoGNN: Time Series Geo-Localization using Two-Tower Graph Neural Networks](<https://www.sota2.com/research/paper/6a279e4990966ade919bf933>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-10T05:00:46.690Z

Authors: Toan Tran, Waqwoya Abebe, Abhishek Potnis, Supriya Chinthavali, Cyrus Shahabi, Li Xiong, Dalton Lunga

Published: 2026-06-06T19:10:56.000Z

## Abstract

This paper investigates a novel concept of time series geolocalization, where the goal is to infer the geographic origin of each raw time series. Successful geolocalization can provide spatial context to time series, enabling downstream location-aware applications. We formalize the problem, adapt core ideas from image geolocalization to establish strong baselines, and propose GeoGNN, a two-tower architecture. During training, GeoGNN's spatial tower learns embeddings of geographic cell candidates by leveraging the geographic adjacency graph, while the temporal tower extracts informative representations from time series. During inference, each temporal representation is matched against candidate geographic embeddings using dot-product similarity, combined with an auxiliary classification head, to predict the time series' associated geographic origin. Experiments on large-scale, countrywide electricity-consumption datasets demonstrate that GeoGNN achieves the best performance across datasets and enhances both fine- and coarse-grained geolocalization accuracy by ~27% on average.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Time Series Geo-Localization on Spain](<https://www.sota2.com/research/sota/time-series-geo-localization-on-spain.md>): Time Series Geo-Localization; Spain; Accuracy @ 1km (Street): 17.473.

- [Time Series Geo-Localization on BuildingsBench Northeast US](<https://www.sota2.com/research/sota/time-series-geo-localization-on-buildingsbench-northeast-us.md>): Time Series Geo-Localization; BuildingsBench Northeast US; Accuracy@1km (Street): 13.372.

- [Time Series Geo-Localization on BuildingsBench US](<https://www.sota2.com/research/sota/time-series-geo-localization-on-buildingsbench-us.md>): Time Series Geo-Localization; BuildingsBench US; Accuracy @ 1km (Street): 11.17.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2606.08303>)

[PDF](<https://arxiv.org/pdf/2606.08303.pdf>)