Adnan KhanAdnanProduct & AI
San Francisco Bay Area
Get in touch

Source note3 min read

UrbanDS: A Graph-Guided LLM Multi-Agent System for Data-Intensive Urban Tasks

Source

arXiv

arxiv.org/abs/2607.26724v1

Read the source

Source

Plain-Language Brief

The paper presents UrbanDS, a multi-agent system that uses a graph to organize and analyze large-scale and heterogeneous urban data. The system includes a Data Profiling Agent, a Relation Agent, a Planner Agent, multiple Execution Agents, and a Report Agent. The authors claim that UrbanDS outperforms existing data science agents on data-intensive tasks and has been successfully deployed in real-world urban applications.

What the Document Says

  • The paper introduces UrbanDS, a graph-guided LLM multi-agent system for data-intensive urban tasks.
  • UrbanDS constructs a unified dataset graph to organize reusable dataset skills and relationships among datasets.
  • A Data Profiling Agent constructs a skill for each dataset in the graph.
  • A Relation Agent identifies relationships among datasets and integrates these relationships into the dataset graph.
  • A Planner Agent retrieves task-relevant datasets from the graph and generates execution plans.
  • Multiple Execution Agents perform data processing and analysis, with their progress and intermediate results shared through a common memory.
  • A Report Agent synthesizes experimental logs into a report that can be refined based on user feedback.
  • UrbanDS has been shown to outperform existing data science agents on data-intensive tasks.
  • The system has been deployed on the urban operations platform of Dongxihu District, Wuhan.
  • The authors have constructed UrbanDS-Bench, an urban data science benchmark covering representative data analysis and modeling tasks.

Key Concepts Mentioned

  • Large Language Model (LLM)
  • Multi-Agent System
  • Urban Data
  • Data Science Tasks
  • Graph
  • Data Profiling Agent
  • Relation Agent
  • Planner Agent
  • Execution Agents
  • Report Agent

Claims Made by the Document

  • UrbanDS outperforms existing data science agents on data-intensive tasks.
  • UrbanDS has been successfully deployed in real-world urban applications.

Evidence Used

  • Experiments on both general and urban benchmarks
  • Deployment on the urban operations platform of Dongxihu District, Wuhan

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Large language model (LLM) agents have been widely applied in automating data science tasks. However, existing methods typically rely on a limited set of provided datasets, and they face challenges in data-intensive scenarios that require discovering and leveraging relevant information from large-scale and heterogeneous data repositories. Urban tasks are representative examples of such scenarios, as urban data are not only large-scale and multi-sourced, but also exhibit complex spatial, temporal, and semantic relationships. To address these challenges, we propose UrbanDS, a graph-guided LLM multi-agent system for data-intensive urban tasks. We first construct a unified dataset graph to organize reusable dataset skills and the relationships among datasets. Specifically, we develop a Data Profiling Agent that constructs a skill for each dataset. Moreover, a Relation Agent identifies relationships among datasets and integrates these relationships into the dataset graph. At runtime, a Planner Agent retrieves task-relevant datasets from the graph and generates execution plans. Multiple Execution Agents then perform data processing and analysis, while their execution progress and intermediate results are shared through a common memory. Finally, a Report Agent synthesizes the experimental logs into a report, which can be further refined based on user feedback. To systematically evaluate the capability of agents in handling data-intensive urban scenarios, we further construct UrbanDS-Bench, an urban data science benchmark covering representative data analysis and modeling tasks. Experiments on both general and urban benchmarks demonstrate that UrbanDS consistently outperforms existing data science agents on data-intensive tasks. Furthermore, UrbanDS has been deployed on the urban operations platform of Dongxihu District, Wuhan, demonstrating its effectiveness in real-world urban applications.