Resilient road infrastructure is a cornerstone of the UN Sustainable Development Goals. Yet a primary indicator of network functionality and resilience is critically lacking: a comprehensive global baseline of road surface information. Here, we overcome this gap by applying a deep learning framework to a global mosaic of Planetscope satellite imagery from 2020 and 2024. The result is the first global multi-temporal dataset of road pavedness and width for 9.2 million km of critical arterial roads, achieving 95.5% coverage where nearly half the network was previously unclassified. This dataset reveals a powerful multi-scale geography of human development. At the planetary scale, we show that the rate of change in pavedness is a robust proxy for a country's development trajectory (correlation with HDI = 0.65). At the national scale, we quantify how unpaved roads constitute a fragile backbone for economic connectivity. We further synthesize our data into a global Humanitarian Passability Matrix with direct implications for humanitarian logistics. At the local scale, case studies demonstrate the framework's versatility: in Ghana, road quality disparities expose the spatial outcomes of governance; in Pakistan, the data identifies infrastructure vulnerabilities to inform climate resilience planning. Together, this work delivers both a foundational dataset and a multi-scale analytical framework for monitoring global infrastructure, from the dynamics of national development to the realities of local governance, climate adaptation, and equity. Unlike traditional proxies such as nighttime lights, which reflect economic activity, road surface data directly measures the physical infrastructure that underpins prosperity and resilience - at higher spatial resolution.

STEAM education in many parts of the Global South remains abstract and weakly connected to learners sociocultural realities. This study examines how human experts evaluate the capacity of Generative AI (GenAI) to contextualize STEAM instruction in these settings. Using a convergent mixed-methods design grounded in human-centered and culturally responsive pedagogy, four STEAM education experts reviewed standardized Ghana NaCCA lesson plans and GenAI-generated lessons created with a customized Culturally Responsive Lesson Planner (CRLP). Quantitative data were collected with a validated 25-item Culturally Responsive Pedagogy Rubric assessing bias awareness, cultural representation, contextual relevance, linguistic responsiveness, and teacher agency. Qualitative reflections provided additional insight into the pedagogical and cultural dynamics of each lesson. Findings show that GenAI, especially through the CRLP, improved connections between abstract standards and learners lived experiences. Teacher Agency was the strongest domain, while Cultural Representation was the weakest. CRLP-generated lessons were rated as more culturally grounded and pedagogically engaging. However, GenAI struggled to represent Ghana's cultural diversity, often producing surface-level references, especially in Mathematics and Computing. Experts stressed the need for teacher mediation, community input, and culturally informed refinement of AI outputs. Future work should involve classroom trials, broader expert participation, and fine-tuning with Indigenous corpora.

X-ray scattering measurements of in situ human brain tissue encode structural signatures of pathological cross-$β$ inclusions, yet systematic exploitation of these data for automated detection remains challenging due to substrate contamination, strong inter-feature correlations, and limited sample sizes. This work develops a three-stage classification framework for identifying cross-$β$ structural inclusions-a hallmark of Alzheimer's disease-in X-ray scattering profiles of post-mortem human brain. Stage 1 employs a Bayes-optimal classifier to separate mica substrate from tissue regions on the basis of their distinct scattering signatures. Stage 2 introduces a multicollinearityaware, class-conditional correlation pruning scheme with formal guarantees on the induced Bayes risk and approximation error, thereby reducing redundancy while retaining class-discriminative information. Stage 3 trains a compact neural network on the pruned feature set to detect the presence or absence of cross-$β$ fibrillar ordering. The top-performing model, optimized with a composite loss combining Focal and Dice objectives, attains a test F1-score of 84.30% using 11 of 211 candidate features and 174 trainable parameters. The overall framework yields an interpretable, theory-grounded strategy for data-limited classification problems involving correlated, high-dimensional experimental measurements, exemplified here by X-ray scattering profiles of neurodegenerative tissue.

In recent years, the sizes of both machine-learning models and datasets have been increasing rapidly. To accelerate the training, it is common to distribute the computation on multiple machines.

In algorithmically fair prediction problems, a standard goal is to ensure the equality of fairness metrics across multiple overlapping groups simultaneously.

As predictive algorithms grow in popularity, using the same dataset to both train and test a new model has become routine across research, policy, and industry. Sample-splitting attains valid inference on model properties by using separate subsamples to estimate the model and to evaluate it. However, this approach has two drawbacks, since each task uses only part of the data, and different splits can lead to widely different estimates. Averaging across multiple splits, I develop an inference approach that uses more data for training, uses the entire sample for testing, and improves reproducibility. I address the statistical dependence from reusing observations across splits by proving a new central limit theorem for a large class of split-sample estimators under arguably mild and general conditions. Importantly, I make no restrictions on model complexity or convergence rates. I show that confidence intervals based on the normal approximation are valid for many applications, but may undercover in important cases of interest, such as comparing the performance between two models. I develop a new inference approach for such cases, explicitly accounting for the dependence across splits. Moreover, I provide a measure of reproducibility for p-values obtained from split-sample estimators. Finally, I apply my results to two important problems in development and public economics: predicting poverty and learning heterogeneous treatment effects in randomized experiments. I show that my inference approach with repeated cross-fitting achieves better power than previous alternatives, often enough to find statistical significance that would otherwise be missed.

Foundation Models (FMs) are large-scale, pre-trained artificial intelligence (AI) systems that have revolutionized natural language processing and computer vision, and are now advancing geospatial analysis and Earth Observation (EO). They promise improved generalization across tasks, scalability, and efficient adaptation with minimal labeled data. However, despite the rapid proliferation of geospatial FMs, their real-world utility and alignment with global sustainability goals remain underexplored. We introduce SustainFM, a comprehensive benchmarking framework grounded in the 17 Sustainable Development Goals with extremely diverse tasks ranging from asset wealth prediction to environmental hazard detection. This study provides a rigorous, interdisciplinary assessment of geospatial FMs and offers critical insights into their role in attaining sustainability goals. Our findings show: (1) While not universally superior, FMs often outperform traditional approaches across diverse tasks and datasets. (2) Evaluating FMs should go beyond accuracy to include transferability, generalization, and energy efficiency as key criteria for their responsible use. (3) FMs enable scalable, SDG-grounded solutions, offering broad utility for tackling complex sustainability challenges. Critically, we advocate for a paradigm shift from model-centric development to impact-driven deployment, and emphasize metrics such as energy efficiency, robustness to domain shifts, and ethical considerations.

Context utilisation, the ability of Language Models (LMs) to incorporate relevant information from the provided context when generating responses, remains largely opaque to users, who cannot determine whether models draw from parametric memory or provided context, nor identify which specific context pieces inform the response. Highlight explanations (HEs) offer a natural solution as they can point the exact context pieces and tokens that influenced model outputs. However, no existing work evaluates their effectiveness in accurately explaining context utilisation. We address this gap by introducing the first gold standard HE evaluation framework for context attribution, using controlled test cases with known ground-truth context usage, which avoids the limitations of existing indirect proxy evaluations. To demonstrate the framework's broad applicability, we evaluate four HE methods -- three established techniques and MechLight, a mechanistic interpretability approach we adapt for this task -- across four context scenarios, four datasets, and five LMs. Overall, we find that MechLight performs best across all context scenarios. However, all methods struggle with longer contexts and exhibit positional biases, pointing to fundamental challenges in explanation accuracy that require new approaches to deliver reliable context utilisation explanations at scale.