The study evaluates whether Earth‑observation foundation models encode geographic location information by attempting to predict coordinates from their embeddings. Using 284 verified European solar farms, the authors tested Tessera v1, Tessera v1.1, and AlphaEarth, finding that all three models contain recoverable geographic data. AlphaEarth showed the strongest correlation between embedding distance and geographic distance, while both Tessera variants outperformed Sentinel‑2 controls, suggesting that geographic information should be considered when auditing such models.
arXiv:2606.08918v2 Announce Type: replace Abstract: Worldwide image geo-localization aims to determine where on Earth a single image was captured. However, visually similar scenes may lie thousands o...
arXiv:2608. 04792v1 Announce Type: new Abstract: Accurate estimation of Above-Ground Biomass (AGB) from satellite imagery is essential for the large-scale monitoring of carbon stocks, yet it remains a challenging regression task at global scale.
Location encoders transform geographic coordinates into high‑dimensional embeddings for machine learning, yet it is unclear how well these embeddings capture interpretable spatial effects. This study benchmarks GeoShapley—a game‑theoretic explainer treating all location features as a single joint player—against eleven TorchSpatial encoders on a synthetic process with known coefficients, across grid, county, and global scales, with and without raw coordinates and under different training regimes. The results show that primary coefficient recovery is consistently high across encoders, while secondary coefficient recovery varies more with scale, especially at the global level, and raw‑coordinate baselines remain competitive throughout.
The study evaluates how the length of observation windows affects the performance of Tessera embeddings for land‑use/land‑cover mapping. By freezing the encoder and recomputing embeddings from a full year down to a single day, the authors benchmark linear probes and UNet heads on LUCAS, DynamicEarthNet, and PASTIS‑R datasets. Results show that embeddings are highly task‑dependent: for phenology‑driven classes (PASTIS‑R) they outperform from‑scratch models by ~46%, while for temporally stable classes (DynamicEarthNet, LUCAS) they match only with full supervision, yet remain more label‑efficient across all datasets.
arXiv:2609.13868v1 Announce Type: cross Abstract: Earth-observation (EO) foundation models have become exceptionally effective at learning se mantic, high-dimensional geospatial embeddings, while mod...
arXiv:2606. 20034v1 Announce Type: new Abstract: Understanding urban spatial morphology is critical for climate modeling, risk assessment, and sustainable urban design, and Local Climate Zone (LCZ) mapping provides the basic framework for this.
GeoContext is a new vision‑language geolocation benchmark that introduces two tasks: GeoHint, where a model must localize an image given a coarse location hint, and GeoVerify, where a model must decide if an image was taken within 150 m of a claimed place. The benchmark builds a context ladder by stratifying nearby reference points by distance and referenceability, allowing the same image to be evaluated under varying context. Evaluation of five models on 109 sites in 30 cities shows that hint repetition is low, localization error grows with hint distance, and models struggle to achieve high discriminability in GeoVerify, with many false acceptances reported with high confidence.
The paper introduces MIND, a method that distills specialist geospatial model embeddings into a single generalist coordinate embedding with adjustable spatial granularity, using nested supervision across multiple embedding dimensions. MIND’s design allows downstream predictors to use only leading chunks or apply a Chunked Penalty to downweight finer details without retraining the INR. The authors evaluate MIND on CoordBench, a large INR benchmark of 52 datasets and 78 targets, and report that MIND and its Chunked Penalty variant achieve the highest regression and classification scores, especially under regional holdout, establishing a new state‑of‑the‑art for geographic implicit neural representations.
arXiv:2606. 24997v1 Announce Type: new Abstract: Geographic implicit neural representations (INRs) learn to map any coordinate on Earth to a location embedding, implicitly encoding geospatial data into the weights of a neural network.
arXiv:2607. 05207v1 Announce Type: cross Abstract: Self-supervised learning (SSL) is designed to learn generic, transferable representations rather than representations optimized for a single task.
GTPred is a new benchmark for geo‑temporal prediction that evaluates multi‑modal large language models (MLLMs) on 370 images taken across 120 years worldwide. It assesses predictions by matching both the year and a hierarchical location sequence, and includes annotated reasoning chains to test intermediate reasoning. Experiments on 15 MLLMs show that while visual perception is strong, models still lack world knowledge and geo‑temporal reasoning, and that adding temporal data improves location inference.
arXiv:2609. 21872v1 Announce Type: cross Abstract: We introduce Chronosphere, a spatio-temporal neural field that learns representations of climate.