GenAI Models Capture Urban Science but Oversimplify Complexity
Zhang, Zhao, Huang, Wang, Ma, Long · physics.soc-ph · 2026-09-05 · 原文
Generative artificial intelligence (GenAI) models are increasingly used for scientific data generation, yet their alignment with empirical knowledge in urban science remains unclear. We therefore ask whether generated urban data reproduce empirical regularities and support repeatable experiments. We introduce AI4US, a framework for evaluating data synthesis and conditional intervention across text and image modalities. Four cases spanning urban systems, within-city structure, neighbourhood vitality and streetscape perception are evaluated against published parameters, observed urban data and human judgements. Generated outputs recovered recognizable scaling and distance-decay patterns and yielded measurable associations between neighbourhood morphology indicators and pedestrian activity, while model judgements showed positive agreement with sampled human choices. Controlled changes to urban conditions produced repeatable output responses across all four cases. However, generated data often compressed empirical numerical coverage, local variation and visual diversity. In an inspectable GenAI model, intermediate activations linearly distinguished some urban relationships, while activ
讲义
讲义·推断 依据「原文」自动生成的结构化摘要（推断），非原文表述；以原文为准。
1. 人话版
Generative artificial intelligence (GenAI) models are increasingly used for scientific data generation, yet their alignment with empirical knowledge in urban science remains unclear.
We therefore ask whether generated urban data reproduce empirical regularities and support repeatable experiments.
2. 领域脉络
本文类目：physics.soc-ph，属于其所在研究脉络的最新进展。
3. 机制拆解
Four cases spanning urban systems, within-city structure, neighbourhood vitality and streetscape perception are evaluated against published parameters, observed urban data and human judgements.
Generated outputs recovered recognizable scaling and distance-decay patterns and yielded measurable associations between neighbourhood morphology indicators and pedestrian activity, while model judgements showed positive agreement with sampled human choices.
4. 证据与数字
We introduce AI4US, a framework for evaluating data synthesis and conditional intervention across text and image modalities.
5. 反例与边界
However, generated data often compressed empirical numerical coverage, local variation and visual diversity.
6. 跨领域连接与意外收获
思考本文机制能否迁移到你正在跟进的问题。
7. 可复用方法
把本文机制与你手头项目对照，找一个两周内能验证的最小实验。
8. 术语表
精读时把不熟的术语记入此处，作为下次回忆的锚点。