City World Model

城市世界模型

Research on perception-aware world models that predict how cities feel, flow, and respond to interventions.

研究感知对齐的城市世界模型——预测城市如何被感受、如何流动，以及如何响应干预。

2026 - present

Static maps and one-shot forecasts are no longer enough for digital planning. Cities are dynamic systems: form, activity, and subjective human experience continually interact. A street redesign, a shade intervention, or a new building changes how people see, feel, and use space.

City World Model develops perception-aware world models for the built environment—representations that predict not only “hard” urban states (mobility, land use, morphology) but also how human perception and comfort shift under scenes and interventions. The aim is a simulatable, evaluable representation of urban dynamics: predictive intelligence, not static layers.

The programme connects three existing strands. Perception and data: street- and window-view human labels and survey platforms (including SP-Survey) supply preference and comfort signals. Spatial backbone: concept retrieval and indexing over street networks (e.g. StreetRAG) ground what the city contains. Reasoning and agents: planning-oriented AI handles constraints, trade-offs, and explainable justification. The world-model layer asks: given an intervention, how will urban state and lived experience evolve?

Current priorities include semantic–topological scene representations; perception-conditioned state transitions and intervention evaluation; and interfaces that can plug into digital twins, simulation, and later embodied urban applications. The work sits alongside agentic urban planning and spatial intelligence of urban comfort—closing the loop from sensing to prediction to decision.

Related empirical and methodological work is listed below.

静态地图与一次性预测已不足以支撑数字规划。城市是动态系统：形态、活动与人类主观体验持续相互作用；一次街道改造、一处遮阴或一栋新建筑，会改变人们如何看见、如何感受、如何使用空间。

城市世界模型（City World Model） 探索面向建成环境的感知对齐世界模型：不仅预测交通或用地等“硬”状态，也建模人类感知与舒适如何随场景与干预而变化。目标是可模拟、可评测、可对接规划决策的城市动态表征——预测性智能，而非静态图层。

研究议程连接三条既有线索。感知与数据：街景/窗景人因标注与调查平台（含 SP-Survey）提供人类偏好与舒适信号；空间骨干：街道网络上的概念检索与空间索引（如 StreetRAG）锚定“城市里有什么”；推理与智能体：规划向 AI 负责约束、权衡与可解释论证。世界模型层回答的是：给定干预，城市状态与体验将如何演化。

当前重点包括：城市场景的语义—拓扑表示；感知感知的状态转移与干预评测；以及与数字孪生、仿真与后续具身/机器人城市应用可对接的接口。工作与智能体城市规划、城市舒适空间智能相互衔接，共同构成从感知到预测再到决策的闭环。

2026

City Landscape In Sight: A Crowdsourced Framework For Unlocking Urban-Scale Window View Perceptions From Real Estate Imagery

Chucai Peng^†, Sijie Yang^†, Ang Liu, Yang Xiang, Zhixiang Zhou, and Filip Biljecki^*

Landscape and Urban Planning, 2026

† Equal contribution. Accepted manuscript.

Abs Bib HTML PDF

City landscapes viewed through home windows influence quality of life, yet perceptions of actual window views at the urban scale remain understudied. This study presents an approach for large-scale mapping of perceptions using 12,334 window view images (WVIs) collected from actual residential properties listed on real estate platforms in Wuhan, China, representing a rarely explored form of urban view imagery that offers advantages over the rendered or simulated window views commonly examined in previous studies. Through a non-immersive virtual reality platform, we collected 27,477 pairwise comparisons across six perceptual dimensions (e.g., Vivid) from 304 participants based on 499 WVIs. A hybrid neural network model was trained to predict human perceptions of all crowdsourced WVIs and map their spatial distribution. Results reveal significant spatial autocorrelation with distinct hot and cold spots across the whole city. Floor level strongly influences human perceptions: while higher floors offer more preferred and extensive window views, lower-floor windows provide residents with quiet and vivid views. An inference model further shows that window view composition matters considerably: high ratios of sky, trees, and low-rise buildings enhance people’s preferences and perceptions of vividness, whereas high ratios of high-rise buildings increase perceptions of monotony and oppression. Importantly, these effects are non-linear: the excessive presence of certain elements can alter their impact on human perception. This work advances urban-scale understanding of residents’ visual experiences and provides evidence-based guidance for human-centric urban planning and real estate to optimise visual landscapes from windows.
@article{peng2026citylandscape, title = {City Landscape In Sight: A Crowdsourced Framework For Unlocking Urban-Scale Window View Perceptions From Real Estate Imagery}, author = {Peng, Chucai† and Yang, Sijie† and Liu, Ang and Xiang, Yang and Zhou, Zhixiang and Biljecki, Filip*}, journal = {Landscape and Urban Planning}, year = {2026}, publisher = {Elsevier}, note = {† Equal contribution. Accepted manuscript.}, }

StreetRAG-Index: Concept-to-Index Retrieval-Augmented Generation Over Urban Street Networks

Sijie Yang, Stephen Law, Zicheng Fan, and Filip Biljecki^*

In Proceedings of the 15th International Space Syntax Symposium, 2026

Bib HTML

@inproceedings{yang2026streetrag,
  title = {StreetRAG-Index: Concept-to-Index Retrieval-Augmented Generation Over Urban Street Networks},
  author = {Yang, Sijie and Law, Stephen and Fan, Zicheng and Biljecki, Filip*},
  booktitle = {Proceedings of the 15th International Space Syntax Symposium},
  address = {Johor Bahru, Malaysia},
  month = jun,
  year = {2026},
}

2025

Reasoning Is All You Need For Urban Planning AI

Sijie Yang, Jiatong Li, and Filip Biljecki^*

In AAAI 2026 (Poster) - AI for Urban Planning (AI4UP), 2025

Abs Bib HTML PDF

AI has proven highly successful at urban planning analysis – learning patterns from data to predict future conditions. The next frontier is AI-assisted decision-making: agents that recommend sites, allocate resources, and evaluate trade-offs while reasoning transparently about constraints and stakeholder values. Recent breakthroughs in reasoning AI – CoT prompting, ReAct, and multi-agent collaboration frameworks – now make this vision achievable. This position paper presents the Agentic Urban Planning AI Framework for reasoning-capable planning agents that integrates three cognitive layers (Perception, Foundation, Reasoning) with six logic components (Analysis, Generation, Verification, Evaluation, Collaboration, Decision) through a multi-agents collaboration framework. We demonstrate why planning decisions require explicit reasoning capabilities that are value-based (applying normative principles), rule-grounded (guaranteeing constraint satisfaction), and explainable (generating transparent justifications) – requirements that statistical learning alone cannot fulfill. We compare reasoning agents with statistical learning, present a comprehensive architecture with benchmark evaluation metrics, and outline critical research challenges. This framework shows how AI agents can augment human planners by systematically exploring solution spaces, verifying regulatory compliance, and deliberating over trade-offs transparently – not replacing human judgment but amplifying it with computational reasoning capabilities.
@inproceedings{yang2025reasoning, title = {Reasoning Is All You Need For Urban Planning AI}, author = {Yang, Sijie and Li, Jiatong and Biljecki, Filip*}, booktitle = {AAAI 2026 (Poster) - AI for Urban Planning (AI4UP)}, year = {2025}, month = nov, }
Urban Comfort Assessment In The Era Of Digital Planning: A Multidimensional, Data-Driven, And AI-Assisted Framework

Sijie Yang, Binyu Lei, and Filip Biljecki^*

In CUPUM 2025 (Oral), 2025

Abs Bib HTML PDF

Ensuring liveability and comfort is one of the fundamental objectives of urban planning. Numerous studies have employed computational methods to assess and quantify factors related to urban comfort such as greenery coverage, thermal comfort, and walkability. However, a clear definition of urban comfort and its comprehensive evaluation framework remain elusive. Our research explores the theoretical interpretations and methodologies for assessing urban comfort within digital planning, emphasising three key dimensions: multidimensional analysis, data support, and AI assistance.
@inproceedings{yang2025urban, title = {Urban Comfort Assessment In The Era Of Digital Planning: A Multidimensional, Data-Driven, And AI-Assisted Framework}, author = {Yang, Sijie and Lei, Binyu and Biljecki, Filip*}, booktitle = {CUPUM 2025 (Oral)}, address = {UCL East, London, UK}, month = jun, year = {2025}, }