Visual attractiveness perception—an individual’s capacity to recognise and evaluate the visual appeal of urban scene safety—has direct implications for well-being, economic vitality, and social cohesion. However, most empirical studies rely on single-source metrics or algorithm-centric pipelines that under-represent human perception. Addressing this gap, we introduce a fully reproducible, multimodal framework that measures and models this domain-specific facet of human intelligence by coupling Generative Pre-trained Transformer 4o (GPT-4o) with 1000 Street View images. The pipeline first elicits pairwise aesthetic judgements from GPT-4o, converts them into a latent attractiveness scale via Thurstone’s law of comparative judgement, and then validates the scale against 1.17 M crowdsourced ratings from MIT’s Place Pulse 2.0 benchmark (Spearman ρ = 0.76, p < 0.001). Compared with a Siamese CNN baseline (ρ = 0.60), GPT-4o yields both higher criterion validity and an 88% reduction in inference time, underscoring its superior capacity to approximate human evaluative reasoning. In this study, we introduce a standardised and reproducible streetscape evaluation pipeline using GPT-4o. We then combine the resulting attractiveness scores with network-based accessibility modelling to generate a “aesthetic–accessibility map” of urban central districts in Chongqing, China. Cluster analysis reveals four statistically distinct street types—Iconic Core, Liveable Rings, Transit-Rich but Bland, and Peripheral Low-Appeal—providing actionable insights for landscape design, urban governance, and tourism planning.

The complexity of urban street vitality is reflected in the interaction of multiple factors. A deep understanding of the multi-dimensional driving mechanisms behind it is crucial to enhancing urban street vitality. However, existing studies lack comprehensive interpretative analyses of urban multi-source data, making it difficult to uncover these drivers’ nonlinear relationships and interaction effects fully. This study introduces an interpretable machine learning framework, using Nanchang, China as a case study. It utilizes urban multi-source data to explore how these variables influence different dimensions of street vitality. This study’s innovation lies in employing an integrated measurement approach which reveals the complex nonlinearities and interaction effects between data, providing a more comprehensive explanation. The results not only demonstrate the strong explanatory power of the measurement approach but also reveal that (1) built environment indicators play a key role in influencing street vitality, showing significant spatial positive correlations; (2) different dimensions of street vitality exhibit nonlinear characteristics, with transit station density being the most influential one; and (3) cluster analysis revealed distinct built environment and socioeconomic characteristics across various street vitality types. This study provides urban planners with a data-driven quantitative tool to help formulate more effective strategies for enhancing street vitality.

Intensified human activities and changes in land-use patterns have led to numerous eco-environmental challenges. A comprehensive understanding of the eco-environmental effects of land-use transitions and their driving mechanisms is essential for developing scientifically sound and sustainable environmental management strategies. However, existing studies often lack a comprehensive analysis of these mechanisms due to methodological limitations. This study investigates the eco-environmental effects of land-use transitions in the Poyang Lake Region over the past 30 years from the perspective of the production-living-ecological space (PLES) framework. Additionally, a geographically explainable artificial intelligence (GeoXAI) framework is introduced to further explore the mechanisms underlying these eco-environmental effects. The GeoXAI framework effectively addresses the challenges of integrating nonlinear relationships and spatial effects, which are often not adequately captured by traditional models. The results indicate that (1) the conversion of agricultural space to forest and lake spaces is the primary factor contributing to eco-environmental improvement. Conversely, the occupation of forest and lake spaces by agricultural and residential uses constitutes the main driver of eco-environmental degradation. (2) The GeoXAI demonstrated excellent performance by incorporating geographic variables to address the absence of spatial causality in traditional machine learning. (3) High-altitude and protected water areas are more sensitive to human activities. In contrast, geographic factors have a greater impact on densely populated urban areas. The results and methodology presented here can serve as a reference for eco-environmental assessment and decision-making in other areas facing similar land-use transformation challenges.

Estimating Visual Quality of Street Space (VQoSS) is pivotal for urban design, environmental sustainability, civic engagement, etc. Recent advancements, notably in deep learning, have enabled large-scale analysis. However, traditional deep learning approaches are hampered by extensive data annotation requirements and limited adaptability across diverse VQoSS tasks. Multimodal Large Language Models (MLLMs) have recently demonstrated proficiency in various computer vision tasks, positioning them as promising tools for automated VQoSS assessment. In this paper, we pioneer the application of MLLMs to VQoSS change estimation, with our empirical findings affirming their effectiveness. In addition, we introduce Street Quality GPT (SQ-GPT), a model that distills knowledge from the current most powerful but inaccessible (not free) GPT-4, requiring no human efforts. SQ-GPT approaches GPT-4’s performance and is viable for large-scale VQoSS change estimation. In a case study of Nanjin, we showcase the practicality of SQ-GPT and knowledge distillation pipeline. Our work promises to be a valuable asset for future urban studies research.

Traditional villages are of great importance for understanding human–environment interactions. However, with the advancement of urbanization, the gradual disappearance of spatial characteristics and the homogenization of landscapes have emerged as prominent issues. Existing studies primarily adopt a single-disciplinary perspective to analyze villages spatial morphology and its influencing factors, relying heavily on qualitative analysis methods. These efforts are often constrained by the lack of digital infrastructure and insufficient data. To address the current research limitations, this paper proposes a Hierarchical Graph Neural Network (HGNN) model that integrates multi-source data to conduct an in-depth analysis of villages' spatial morphology. The framework includes two types of nodes—input nodes and communication nodes—and two types of edges—static input edges and dynamic communication edges. By combining Graph Convolutional Networks (GCN) and Graph Attention Networks (GAT), the proposed model efficiently integrates multimodal features under a two-stage feature update mechanism. Additionally, based on existing principles for classifying villages spatial morphology, the paper introduces a relational pooling mechanism and implements a joint training strategy across 17 subtypes. Experimental results demonstrate that this method achieves significant performance improvements over existing approaches in multimodal fusion and classification tasks. Additionally, the proposed joint optimization of all sub-types lifts mean accuracy/F1 from 0.71 / 0.83 (independent models) to 0.82 / 0.90, driven by a 6% gain for parcel tasks. Our method provides scientific evidence for exploring villages' spatial patterns and generative logic.