City Editing: Hierarchical Agentic Execution for Dependency-Aware Urban Geospatial Modification
cs.MA, cs.AI
Submitted: 2026-02-22
Updated: 2026-09-09
Comments: Accepted by ACM SIGSPATIAL 2026
Code: https://github.com/anonymous-share-review/CEAE
License: http://creativecommons.org/licenses/by/4.0/
The gist: Urban renewal requires incremental modifications to existing geospatial plans, yet manually updating complex layouts under spatial constraints is labor-intensive and error-prone.
Terminology
Abstract
Urban renewal requires incremental modifications to existing geospatial plans, yet manually updating complex layouts under spatial constraints is labor-intensive and error-prone. To tackle this, we propose CEAE, a hierarchical agentic framework that formulates urban renewal as machine-executable GeoJSON editing from natural-language instructions. CEAE decomposes instructions into hierarchical geometric intents, executing edits from coarse to fine while preserving spatial consistency through a self-reflective execution-validation loop. Experimental results show that CEAE outperforms baselines in execution validity, robustness, and geometric accuracy.
Sources
- GPT-4 Technical Report
- Qwen3 Technical Report
- LLaMA: Open and Efficient Foundation Language Models
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Agent AI: Surveying the Horizons of Multimodal Interaction
- Generative AI Meets Future Cities: Towards an Era of Autonomous Urban Intelligence
- HuggingFace's Transformers: State-of-the-art Natural Language Processing
- Large Language Model for Participatory Urban Planning
Related papers
- Highway Congestion Reduction through Reinforcement Learning Based Eulerian Headway Control
- You Only Align Once: Propagating Cooperative Behaviors in Multi-Agent Systems through Seed Agents
- Deny Without Disabling: Authorization-Paired Evaluation and Control for Multi-Agent Systems
- MA-SAPO: Multi-Agent Reasoning for Score-Aware Prompt Optimization
- PeroMAS: A Multi-agent System of Perovskite Material Discovery
- StitchCUDA: An Automated Multi-Agents End-to-End GPU Programing Framework with Rubric-based Agentic Reinforcement Learning