Form-finding is a process in architectural design. Architects create and manipulate the morphology of a building by finding the form using digital tools and algorithms, such as machine learning. Recent research indicates that existing machine learning methods for architectural form-finding are not efficient for training and cannot generate multiple 3D forms under the constraints of users. Therefore, in this research, we develop a method to train and apply low-rank adaptation (LoRA) models in Stable Diffusion (SD) to generate 3D architectural forms based on morphological heat maps. Furthermore, the generated 3D forms can be directly used to precisely control the generation of realistic architectural renderings using pre-trained LoRA and SD models. In conclusion, our method can help architects generate 3D architectural models with consistent renderings. It can serve as a useful tool to improve efficiency and creativity in the architectural design practice of form-finding.

A universal set of mathematical qualities combine complexity with geometry to define “living geometry”. Those properties apply equally to organisms as to inanimate matter. A special type of geometry that merges approximate fractal structure with nested symmetries on all scales describes most biological and natural forms. Christopher Alexander discovered 15 geometrical properties in the context of architecture adapted to human feelings, and those provide a practical toolkit for creating living geometry in artifacts, buildings, and urban spaces. AI can be used to both validate the importance of living geometry for human health, and to evaluate architectural designs before and after they are built. Human beings evolved their neurological system to interpret living geometry defined by the ancestral natural environment, and the same interpretative reference is used to navigate today's built environment. This knowledge helps to understand how geometry affects cognition, emotions, and health. Asking large-language models about the best learning environments for creative thought produces a list of suggestions for renovating lecture and studio spaces in architecture schools. A severe disconnect between the contemporary urban experience and living geometry has negative effects on human health, intelligence, and the learning process. AI helps to completely rethink design aesthetics and design-making strategies.

Architectural plan generation via pix2pix series algorithms faces dual challenges: the absence of domain-specific evaluation metrics and a lack of systematic insights into the joint impact of training configurations. To address the limitations of pix2pix-based models adaptation to architectural design, we designed a training regimen involving 12 experiments with varying training set sizes, dataset characteristics, and algorithms. These experiments utilized our self-built, high-quality, large-volume synthetic dataset of architectural-like plans. By saving intermediate models, we obtained 240 generative models for evaluation on a fixed test set. To quantify model performance, we developed a dual-aspect evaluation method that assesses predictions through pixel similarity (principle adherence) and segmentation line continuity (vectorization quality). Analysis revealed algorithm choice and training set size as primary factors, with larger sets enhancing the benefits of high-resolution and enhanced-annotation datasets. The optimal model achieved high-quality predictions, demonstrating strict adherence to predefined principles (0.81 similarity) and effective vectorization (0.86 segmentation line continuity). Testing on 7695 samples of varying complexity confirmed the model's robustness, strong generative capability, and controlled innovation within defined principles, validated through 3D model conversion. This work provides a domain-adapted framework for training and evaluating pix2pix-based architectural generators, bridging generative research and practical applications.

This study explores the role of artificial intelligence (AI) in the conceptual design phase of interior design education, focusing on AI's potential to help students visualise and refine creative ideas. Conducted within a design studio course, the research integrates text-to-image generators, particularly Midjourney to support students' design processes. Implemented in the fourth week of a 14-week course, a structured workshop introduced students to Midjourney, with surveys conducted both at this stage and during the final submission to capture changes in student perspectives. Using a two-phase case study involving a workshop, surveys, and interviews among senior undergraduate students in the bachelor's program of the Interior Architecture and Environmental Design Department, the study assesses the impact of AI prompts, from simple keywords to detailed narratives, on concept development and project outcomes. Findings indicate that AI broadens design possibilities, facilitates iterative ideation, and improves conceptual precision through high-fidelity visualizations. While students view AI as a valuable addition to their creative process, they also express concerns about ethics and the need to balance AI's benefits with preserving design authenticity. This research contributes to the broader discussion on AI's role in design, advocating for a balanced integration that respects both technological potential and human creativity.

As society confronts increasingly complex demands and the growing need for carbon-neutral architecture, AI-driven design methodologies are evolving rapidly. However, the lack of a unified integration platform in the design process continues to hinder AI's integration into real-world workflows. To address this challenge, we introduce ArchiWeb, a web-based platform specifically built to support AI-driven processes in early-stage architectural design. ArchiWeb transforms architectural representation and problem formulation by utilizing lightweight data protocols and a modular algorithmic network within an interactive web environment. Through its cloud-native, open-architecture framework, ArchiWeb enables deeper integration of AI technologies while accelerating the accumulation, sharing, and reuse of design knowledge across projects and disciplines. Ultimately, ArchiWeb aims to drive architectural design toward greater intelligence, efficiency, and sustainability—supporting the transition to data-informed, computationally enabled, and environmentally responsible design practices.

This paper proposes a deep learning-based intelligent modeling framework for generating 3D architectural models from manual sketches, addressing the domain gap in 2D-to-3D transformation. By integrating architectural domain knowledge—specifically the phased, selective, and cyclic characteristics of the design process—the framework ensures a structured and iterative generative approach. The framework consists of a 2D design phase, where image retrieval, Stable Diffusion, and CycleGAN facilitate conceptual exploration, multi-scheme generation, and depth map extraction, and a 3D design phase, where Pixel2Mesh generates 3D forms, refined through Grasshopper-based parametric optimization. Empirical evaluation demonstrates that the framework effectively preserves structural fidelity while allowing for generative variations. Structural similarity and geometric accuracy metrics validate its performance, confirming its ability to balance AI-driven massing generation with architectural precision. A Mars habitat case study, conducted in an academic research setting, serves as a controlled experiment to assess adaptability. While demonstrating the framework's potential for AI-assisted architectural generation, the study also highlights the need for broader validation across diverse architectural typologies. This research bridges traditional and AI-driven design methodologies by integrating computer vision and generative models into architectural workflows. The proposed framework contributes to architectural design by introducing a cross-disciplinary approach that enhances the efficiency, quality, and innovation of design processes.

The effectiveness of machine learning (ML) models for architectural applications relies on high-quality datasets balanced with advancements in model architecture and computational capacity. Current methods for evaluating architectural floor plan datasets typically depend on explicit semantic annotations, which limit their effectiveness and scalability when annotations are unavailable or inconsistent. To address this limitation, this research develops an isovist-based latent representation approach to quantitatively measure typicality and diversity within architectural datasets without relying on semantic labels. We introduce Isovist Latent Norm Typicality, a metric that leverages the statistical structure of latent representations derived from isovist morphological features using a variational autoencoder (VAE). This metric quantifies typicality by analyzing distributional shifts in latent representations between individual floor plans and a reference dataset using a modified Wasserstein distance. Experimental results demonstrate the approach's ability to distinguish typical from atypical floor plan configurations, capturing the morphological features that complement conventional metrics. Comparative analysis indicates that our method provides insights into spatial organization, correlating with conventional properties such as programmatic diversity and spatial openness. By quantifying typicality through purely morphological features, the proposed methodology facilitates dataset curation prior to costly semantic annotation, enhancing dataset quality and enabling scalability to more extensive and diverse architectural datasets.

In modern architectural design, as complexity increases and diverse demands emerge, reconstructing 3D spaces has become a crucial method. However, existing methods remain limited to small-scale scenarios and exhibit poor reconstruction accuracy when applied to building-scale environments, resulting in unstable mesh quality and reduced design productivity. Furthermore, the lack of real-time, interactive editing tools prolongs design iteration cycles and impedes workflow efficiency. To address this issue, we propose the following contributions:

(1) We construct ArchiNet++, an architectural dataset that includes 710,180 multi-view images, 5200 SketchUp models, and corresponding camera parameters from the conceptual design phase of architectural projects.

(2) We introduce Drag2Build++, an interactive 3D mesh reconstruction framework featuring drag-based editing and three core innovations: a differentiable geometry module for fine-grained deformation, a 2D-3D rendering bridge for supervision, and a GAN-based refinement module for photorealistic texture synthesis.

(3) Comprehensive experiments demonstrate that our model excels in generating highquality 3D meshes and enables rapid mesh editing via drag-based interactions. Furthermore, by incorporating textured mesh generation into this interactive workflow, it improves both efficiency and modeling flexibility.

We hope this combination can contribute to a more intuitive modeling process and offer a practical tool set that supports the digital transformation efforts within architectural design.

This paper presents a new generative artificial intelligence (AI) approach for creating modular skeletal frameworks, using vernacular bamboo stilt houses as examples to investigate an innovative methodological perspective. By transforming building skeletons to connected graphs, our method uses Variational Graph Autoencoders (VGAE) and Graph Sample and Aggregate (GraphSAGE) to generate 3D modular components based on spatial constraints set by users, such as axis grids and chosen room areas. The graph representation encodes structural elements as edges and their connections as nodes, maintaining critical dimensional constraints and spatial relationships. Using data from bamboo stilt houses built without architects, we make a specialized dataset of geometric skeletons for model training. Experimental results demonstrate the effectiveness of our approach in capturing the distribution of featured elements in building frameworks and in generating structurally sound designs, with GraphSAGE showing better performance compared to alternative methods. The probabilistic edge prediction approach allows for a collaborative human-AI design process, empowering designers while utilizing computational capabilities. The inherent flexibility of the graph-based representation makes it adaptable to a wide range of materials and scales.

Pre-sale and second-hand housing transaction modes dominate China's real estate market. However, many existing studies tend to treat the housing market as a homogeneous entity, overlooking the heterogeneity in core influencing factors across different transaction types. Thoroughly understanding the factors affecting various housing types can assist policy-makers in formulating differentiated regulatory decisions through environmental intervention. Therefore, this study utilized multi-source big data and compared the performance of multiple machine learning models to evaluate the relative importance and nonlinear effects of building-level, neighborhood-level, and street-level built environment factors on pre-sale and second-hand housing prices. The empirical study of Chengdu, China revealed that distance to city center was the most significant explanatory factor influencing pre-sale and second-hand housing prices among all factors. Significant differences existed between neighborhood-level and street-level built environment factors' nonlinear and threshold effects on pre-sale and second-hand housing prices. Notably, subway accessibility showed a U-shaped impact on pre-sale housing prices. To the best of our knowledge, our study is one of the early studies systematically investigating the influencing differences between pre-sale housing prices and second-hand housing prices, providing robust evidence for regulating housing prices through environmental interventions and offering critical references for policymakers and market participants.

Deep reinforcement learning (DRL) remains underexplored within architectural robotics, particularly in relation to self-learning of architectural design principles and design-aware robotic fabrication. To address this gap, we applied established DRL methods to enable robot arms to autonomously learn design rules in a pilot block wall assembly-design scenario. Recognizing the complexity inherent in such learning tasks, the problem was strategically decomposed into two sub-tasks: (i) target reaching (T1), modeled within a continuous action space, and (ii) sequential planning (T2), formulated within a discrete action space. For T1, we evaluated major DRL algorithms—Proximal Policy Optimization (PPO), Advantage Actor-Critic (A2C), Deep Deterministic Policy Gradient, Twin Delayed Deep Deterministic Policy Gradient, and Soft Actor-Critic (SAC), and PPO, A2C, and Double Deep Q-Network (DDQN) were tested for T2. Performance was assessed based on training efficacy, reliability, and two novel metrics: degree index and variation index. Our results revealed that SAC was the best for T1, whereas DDQN excelled in T2. Notably, DDQN exhibited strong learning adaptability, yielding diverse final layouts in response to varying initial conditions.

The nature—culture divide, a longstanding conceptual separation between human beings and the natural environment, is increasingly challenged by the pressing need to address climate change. This urgency calls for design approaches that can synthesise social and sustainable aspects, creating environmental-user-centric solutions. Our study aimed to bridge this divide by exploring the integration of digital and human crafts, with a focus on wood upcycling furniture as a case study. It investigates the flow of design information, creating an interactive feedback loop between physical and digital domains. To ensure the workflow aligns with stakeholder needs, the study engages professionals interdisciplinarily, including designers, informaticists, and engineers, to collectively test and reflect on the process. The proposed pipeline was then compared with the collaborative pipeline that emerged, incorporating stakeholder perspective to refine the system design. The resulting workflow embraced 3D scanning, AIdriven design generation, VR user scenario simulation, and AR-assisted physical fabrication. The digital and physical furniture prototypes suggest new avenues for design informatics by synthesising objective mathematical decisions with subjective semiotic inputs. By exploring the integration of human and machine crafts in the co-creation process, the reflections contribute to sustainable urban and community construction (SDG 11), revealing potentials for scalability in architectural production.

China's rapid urbanization presents significant challenges for rural construction and resource management, often prioritizing economic gains over climate adaptability and energy efficiency. This study focuses on traditional Huizhou houses, integrating energy consumption and comfort analysis into the early design stages. Initial simulations using the Universal Thermal Climate Index (UTCI) established a baseline model for comparison. Through the Wallacei_X plugin, optimized designs achieved a 19.88% reduction in energy use intensity (EUI) and a 9.37% improvement in summer outdoor comfort (UTCI_H) compared to the baseline. Further analysis along the Pareto frontier using Scikit-learn demonstrated high predictive accuracy with XGBoost (F₁ scores: 0.80 for 4-side houses, 0.78 for 3-side houses). To enhance interpretability, SHapley Additive exPlanations (SHAP) analysis explored nonlinear relationships between design variables and building performance, while coupling analysis examined the spatial relationships between houses and their environmental impact. In the final validation, the proposed workflow effectively linked building performance prediction with design optimization, achieving a 26% performance improvement over the original site plan. This integrated approach enables rapid performance evaluations, reduces costs, and provides practical design references. It highlights the potential of combining genetic algorithms and machine learning to drive sustainable rural development.

As global urbanization accelerates, the spatial vitality of historical urban areas has become a critical issue in urban regeneration and sustainable development. Some existing spatial vitality evaluation frameworks fail to integrate multiple dimensions effectively, limiting their capacity to capture the dynamic complexity of these areas comprehensively. This study utilizes multi-source big data and deep learning technologies to propose a new multidimensional evaluation system for spatial vitality, improving existing models and systematically analyzing distribution patterns and formation mechanisms. The research results show that: (1) The spatial vitality of Changsha's historical urban area exhibits a distinct “core-periphery” pattern. Core commercial zones show high vitality due to functional concentration, whereas peripheral areas have weaker vitality because of lower physical space quality and limited functional diversity. (2) Through correlation and principal component analyses, five key factors influencing spatial vitality were identified: emotional perception, visual aesthetics, spatial attractiveness, Functionality and Structure, and traffic conditions. (3) Bivariate spatial autocorrelation analysis further revealed spatial clustering effects between spatial vitality and its key factors, emphasizing the potential for enhancing functional diversity and optimizing road traffic conditions in core areas. The study's findings offer scientific guidance for urban regeneration and policy-making, particularly in optimizing spatial layouts, enhancing vitality, and fostering the coordinated development of cultural heritage protection, providing valuable insights for other developing countries.

The history of architectural and urban practices reflects humanity's enduring quest to comprehend and shape its environment, often through the lens of unifying meta-narratives. This paper critiques the tendency to seek cohesive frameworks, drawing from Graham Harman's speculative realism and Bruno Latour's “Principle of Irreduction,” which challenge hierarchical structures in understanding reality. These perspectives underscore the irreducibility and multiplicity of existence, advocating for a paradigm shift that resists determinism and embraces open-endedness. In this context, Adrian Bejan's constructal law offers a compelling alternative for interpreting architectural and urban forms. Constructal theory conceptualizes form and design as evolutionary responses to flow systems, framing architecture as an active participant in the dynamic interplay of environmental, social, and temporal forces. This perspective encourages a reevaluation of architectural practices not as definitive solutions but as iterative processes that engage with complexity and contingency. By integrating constructal theory with contemporary philosophical critiques, this article proposes a poly-narrative of architecture and urbanism that aligns with the fluidity and multiplicity of modern existence. It argues for a departure from static frameworks toward adaptive methodologies that acknowledge the interconnectedness of actors, scales, and temporalities. Ultimately, this approach reframes design as a dialogic process, fostering resilience and innovation in confronting the uncertainties of a rapidly evolving world.

The urban spatial structure reflects a city's development history, cultural heritage, and socio-economic conditions. A rational urban spatial structure is crucial for urban development. This study focuses on the main urban area of Nanjing, analyzing POI and nighttime light data from 2016 to 2020. Utilizing kernel density estimation and coupling coordination models, it explores the temporal and spatial evolution characteristics of Nanjing's urban spatial structure. Geographic detectors are employed to assess the impact of various factors on this structure. The findings indicate that: (1) Nanjing's urban spatial structure displays a pattern of central aggregation and peripheral expansion, with high brightness concentrated in the urban center and a significant increase in peripheral brightness, signaling initial success in establishing urban subcenters; (2) The coupling relationship between nighttime light brightness and POI density has strengthened, suggesting improved coordination of the urban spatial structure; (3) The evolution of Nanjing's urban spatial structure results from the combined effects of multiple factors, including economic level, population distribution, transportation conditions, and policy planning.

This study investigates the architectural interventions of Francesco Venezia in the Belice Valley after the 1968 earthquake, with a particular focus on the c and the Open-Air Theatre in Salemi. Employing a qualitative, design-driven methodology, the research integrates formal spatial analysis with interpretative frameworks from spatial theory, cultural memory studies, and phenomenological approaches to architectural experience. Primary sources, including on-site surveys, original drawings, and project documentation, are complemented by critical essays and historical accounts. The analysis centers on the theme of spatio-temporal continuity, examining how Venezia's works engage with memory, ruins, and the fragmented identity of place. The findings reveal that Venezia's design process anchored in reinterpretation rather than reconstruction produces anomalous monuments that reestablish a sense of historical depth while resisting conventional forms of memorialization. His architecture articulates a dialectical relationship between absence and presence, solidifying a new spatial narrative in a landscape marked by trauma and displacement. This paper presents a globally applicable design paradigm for handling cultural memory, identity, and continuity in the architecture of crisis and recovery by suggesting a substitute for traditional post-disaster restoration.

Accurate evaluation on residential energy demand is crucial for sustainable energy systems and urban development. Bottom-up approaches are reliable to capture the building energy characteristics. However, the existing bottom-up approaches require large volumes of high-quality residential building data, which are often inaccessible in developing countries like China. This study proposes a bottom-up approach based on prototype residential units to assess energy challenges in urban residential sector of China. By integrating data collection, variable selection, and K-prototype clustering analysis, the method identifies several residential prototypes that can be used for predictions on energy dynamic variation in residential sector. The proposed method is applied in Guangzhou, a major city in southern China, during heat wave events as a case study. The results indicate that both daytime and nighttime cooling loads in the residential sector are significant and should not be overlooked; peak hourly energy demand typically occurs at 7:00 a.m. and 9:00 p.m. The proposed approach provides a scalable framework for forecasting energy demand, supporting policy and urban planning to reduce consumption while enhancing resilience to extreme weather and understanding of energy challenges in China's urban residential sector.