Puffin is a probabilistic framework that disaggregates coarse spatial totals into high‑resolution estimates by leveraging satellite embeddings as covariates. Unlike deterministic methods, it learns a probability distribution for each fine‑scale subregion and trains with an aggregation‑aware likelihood, ensuring that predictions sum to the observed regional total. The approach provides calibrated uncertainty and does not require fine‑resolution labels, and it has been evaluated on German and US census, employment, and election data.
DeepC4 is a deep learning-based spatial disaggregation method that uses local census statistics as cluster-level constraints and incorporates multiple conditional label relationships in a multitask learning framework. Applied to Rwandan urban morphology, it achieves macro‑F1 scores of 0.63, 0.78, and 0.45 for roof, wall, and height prediction, respectively, and estimates national dwelling and occupant counts within about 1.1% error compared to census records. The approach outperforms existing GEM and METEOR methods and covers 32‑49% more 500‑meter grid pixels across provinces.
By Joshua Dimasaka, Christian Gei{\ss}, Emily So
arXiv:2608. 08064v1 Announce Type: new Abstract: Learning fine-grained spatial patterns from coarse-resolution data is challenging, especially in causal settings where high-resolution effects must be inferred from aggregated interventions and outcomes.
By Gerrit Gro{\ss}mann, Sumantrak Mukherjee, Sebastian J. Vollmer
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.
By Isaac Corley, Arjun Rao, Esther Rolf, Konstantin Klemmer, Evan Shelhamer, Nils Lehmann, Marc Ru{\ss}wurm, Gengchen Mai, Nathan Jacobs, Hannah Kerner
The paper presents an uncertainty‑aware machine‑learning approach for mapping poverty in Africa using satellite imagery. By combining simultaneous quantile regression with a novel conformal prediction technique, the authors generate statistically guaranteed prediction intervals for neighborhood‑level International Wealth Index estimates, achieving high explanatory power (R² = 0.75) while acknowledging broader uncertainty. They also propose a risk‑controlled aid allocation procedure that leverages both survey data and model predictions, showing in simulations that it can deliver more aid per eligible recipient than alternative strategies.
By Markus B. Pettersson, James Bailie, Mohammad Kakooei, Eagon Meng, Adel Daoud
The paper introduces a multi‑view contrastive learning framework that creates low‑dimensional spatial embeddings by combining satellite imagery and OpenStreetMap data across Europe. These embeddings align with coordinate‑based encodings, allowing any dataset with latitude‑longitude pairs to be enriched with meaningful spatial features without needing the original spatial inputs. In case studies on French real‑estate prices and Belgian flood claim counts, models using the embeddings outperform those using raw coordinates, improving predictive accuracy and territorial risk classification while offering explainable spatial effects.
By Freek Holvoet, Christopher Blier-Wong, Katrien Antonio
The paper introduces DSP‑BART‑HS, a Dynamic Spatial Panel Bayesian Additive Regression Trees model with Horseshoe shrinkage, designed for high‑dimensional spatio‑temporal panel data. Across nine simulated scenarios, the model outperforms or matches a wide range of spatial econometric, non‑parametric machine learning, and small‑area estimators, especially when individual‑level non‑linearity drives outcome variance. The authors validate the method on two U.S. county‑level applications—intergenerational economic mobility and geographic income inequality—showing strong predictive accuracy even under unseen‑region, random, and temporal holdouts, while noting a temporal extrapolation advantage for a simpler autoregressive model.
By Hammed A. Olayinka, Saheed O. Olayemi
arXiv:2602. 04525v2 Announce Type: replace-cross Abstract: Rapid urban expansion has fueled the growth of informal settlements in major cities of low- and middle-income countries, with Lahore and Karachi in Pakistan and Mumbai in India serving as prominent examples.
By Muhammad Taha Mukhtar, Syed Musa Ali Kazmi, Khola Naseem, Muhammad Ali Chattha, Andreas Dengel, Sheraz Ahmed, Muhammad Naseer Bajwa, Muhammad Imran Malik
arXiv:2607. 20559v1 Announce Type: cross Abstract: Monitoring poverty and food security indicators is imperative for addressing socioeconomic challenges in developing nations.
By Aditya Dutt, Paul Gader, Aditya Singh
MoRA is a human‑centric geospatial representation learning framework that uses a large mobility graph as its backbone to fuse spatial tokenization, graph neural networks, and asymmetric contrastive learning. It aligns over 100 million points of interest, massive remote sensing imagery, and structured demographic data with a billion‑edge mobility graph, producing compact 128‑dimensional embeddings that capture socio‑economic context and functional roles of locations. On a benchmark of nine downstream social and economic prediction tasks, MoRA outperforms state‑of‑the‑art models by an average of 12.9% and demonstrates scaling behavior analogous to large language models.
By Ya Wen, Jixuan Cai, Qiyao Ma, Linyan Li, Xinhua Chen, Chris Webster, Yulun Zhou
arXiv:2607. 10410v1 Announce Type: cross Abstract: Reliable forecasting of several interrelated environmental variables - such as regional precipitation and temperature, or other correlated geophysical fields - across many locations calls for accurate predictions accompanied by trustworthy statements of their uncertainty.
By Jongwook Kim, Jong-Min Kim
PlacePulse is a structural geospatial foundation model that transforms 572 attributes—expanded to 2,288 through H3 neighborhood aggregates—into 256‑dimensional unit‑norm embeddings for over a million H3 resolution‑7 cells across the United States. The model achieves high intrinsic reconstruction quality (mean $R^2 = 0.853$) and strong spatial organization (Moran's $I = 0.93$), while outperforming a dimensionality‑matched baseline on 74 held‑out targets from 12 external sources and competing with Google’s Population Dynamics Foundation Model on various health, economic, election, mobility, and social‑vulnerability outcomes. These results demonstrate that static structural composition can serve as a distinct and complementary modality for geospatial foundation models.
By Luc\'ia Garc\'ia-Duarte, Miguel \'Alvarez, Diego Vicente, Javier P\'erez Trufero, Javier de la Torre