With the widespread adoption of large language models (LLMs) in code generation tasks, geospatial code generation has emerged as a critical frontier in the integration of artificial intelligence and geoscientific analysis. This growing trend underscores the urgent need for systematic evaluation methodologies to assess the generation capabilities of LLMs in geospatial contexts. In particular, geospatial computation and visualization tasks in the JavaScript environment rely heavily on the orchestration of diverse frontend libraries and ecosystems, posing elevated demands on a model’s semantic comprehension and code synthesis capabilities. To address this challenge, we propose GeoJSEval—the first multimodal, function-level automatic evaluation framework for LLMs in JavaScript-based geospatial code generation tasks. The framework comprises three core components: a standardized test suite (GeoJSEval-Bench), a code submission engine, and an evaluation module. It includes 432 function-level tasks and 2071 structured test cases, spanning five widely used JavaScript geospatial libraries that support spatial analysis and visualization functions, as well as 25 mainstream geospatial data types. GeoJSEval enables multidimensional quantitative evaluation across metrics such as accuracy, output stability, resource consumption, execution efficiency, and error type distribution. Moreover, it integrates boundary testing mechanisms to enhance robustness and evaluation coverage. We conduct a comprehensive assessment of 20 state-of-the-art LLMs using GeoJSEval, uncovering significant performance disparities and bottlenecks in spatial semantic understanding, code reliability, and function invocation accuracy. GeoJSEval offers a foundational methodology, evaluation resource, and practical toolkit for the standardized assessment and optimization of geospatial code generation models, with strong extensibility and promising applicability in real-world scenarios. This manuscript represents the peer-reviewed version of our earlier preprint previously made available on arXiv.

AlphaEarth Foundations (AEF) unifies multi-modal and multi-temporal observations into analysis-ready low-dimensional representations, introducing a representation-driven paradigm that addresses key limitations of traditional task-centric approaches, including high engineering complexity, limited cross-regional transferability, and strong dependence on labeled data. This paper presents the first systematic review of AEF, synthesizing 23 research articles published up to March 2026 from three perspectives: conceptual positioning, technical framework, and application practices. It first clarifies the distinctions between AEF and related paradigms, including foundation models, remote sensing large models, and existing unified embedding approaches. It then summarizes the organization of its data system and key technical components. Based on a systematic literature survey, the paper further provides a structured synthesis of current studies in terms of thematic and regional distribution, scientific questions, usage patterns, and evaluation methods. The analysis indicates that AEF represents an important step toward a paradigm shift from task-driven to representation-driven Earth observation, offering clear advantages in reducing engineering barriers, enabling cross-regional transfer, and establishing a unified environmental semantic foundation. However, limitations remain in temporal resolution, semantic interpretability, and performance in specific task scenarios. Future work should focus on enhancing dynamic representation capabilities, cross-domain adaptation mechanisms, multi-source integration frameworks, systematic evaluation and annotation systems, and asset-oriented applications. By delineating the capability boundaries and applicable contexts of AEF from both methodological and empirical perspectives, this study provides a systematic reference for the standardized development and rational application of unified surface embeddings as an emerging geospatial information infrastructure.

Pixel-level slum mapping remains constrained by limited cross-city generalisation, the lack of continuous estimates of intra-pixel slum coverage, and weak comparability across geographic contexts. AlphaEarth Foundations (AEF), which provides globally consistent 64-dimensional annual surface embeddings at 10 m resolution, offers an analysis-ready representation for scalable and lightweight slum monitoring. However, its applicability to slum mapping remains unclear because slum conditions are only indirectly observable from Earth observation data and are jointly shaped by built-form and socioeconomic processes. We systematically evaluate AEF for binary slum classification and sub-pixel slum cover fraction estimation across 12 cities and 69 city–year pairs from 2017 to 2024, using high-resolution GRAM-derived pseudo-masks as supervisory labels. The evaluation spans four training strategies, two protocols (random split and 3 × 3 spatial block cross-validation), six auxiliary feature configurations, and five baseline models, complemented by representation-level analyses (PCA, SHAP) and full-AOI mapping. Five findings emerge. (1) Same-city cross-year training is optimal under both protocols (median spatial F1 = 0.616, R² = 0.466); temporal expansion outperforms cross-city transfer, indicating city-scale representational drift. (2) Regression R² is driven primarily by zero/non-zero boundary discrimination: positive-pixel R² is consistently negative across all cities, revealing limited capacity to model intra-pixel variation in sub-pixel slum cover fraction at 10 m. (3) PC36 is consistently top-ranked across tasks; classification saturates at k = 32 while regression remains unsaturated at k = 64. (4) POI features yield the largest density gain (ΔR² = +0.064). (5) For six cities meeting dual-task usability thresholds, full-AOI inference across 2017–2024 preserves slum cluster structure (mean SSIM = 0.926). The study delineates the capabilities and complementarity needs of foundation-model embeddings for slum monitoring.