City Editing: Hierarchical Agentic Execution for Dependency-Aware Urban Geospatial Modification

TL;DR

This paper introduces a hierarchical agentic framework for efficient, robust, and consistent urban plan modifications using structured geospatial data and natural language instructions, improving urban renewal processes.

Contribution

It presents a novel hierarchical agentic system that decomposes natural language edits into geometric intents and coordinates multi-level urban plan modifications.

Findings

Significant efficiency improvements in urban plan editing.

Enhanced robustness and correctness in modifications.

Maintained spatial validity and consistency during edits.

Abstract

As cities evolve over time, challenges such as traffic congestion and functional imbalance increasingly necessitate urban renewal through efficient modification of existing plans, rather than complete re-planning. In practice, even minor urban changes require substantial manual effort to redraw geospatial layouts, slowing the iterative planning and decision-making procedure. Motivated by recent advances in agentic systems and multimodal reasoning, we formulate urban renewal as a machine-executable task that iteratively modifies existing urban plans represented in structured geospatial formats. More specifically, we represent urban layouts using GeoJSON and decompose natural-language editing instructions into hierarchical geometric intents spanning polygon-, line-, and point-level operations. To coordinate interdependent edits across spatial elements and abstraction levels, we propose a hierarchical agentic framework that jointly performs multi-level planning and execution with explicit propagation of intermediate spatial constraints. We further introduce an iterative execution-validation mechanism that mitigates error accumulation and enforces global spatial consistency during multi-step editing. Extensive experiments across diverse urban editing scenarios demonstrate significant improvements in efficiency, robustness, correctness, and spatial validity over existing baselines.