Event ·Official first-hand

UrbanAgent:跨系统城市任务的工具增强智能体框架

UrbanAgent是一个面向跨系统城市任务的工具增强智能体框架,将大语言模型的推理能力与代码执行、API调用及Model Context Protocol工具集结合。研究团队同时推出了Urban-Eval基准,用于评估跨系统城市请求的任务结果与执行质量。实验显示UrbanAgent任务成功率达71%,比最强基线高出10个百分点,该优势在GPT-5-mini、Gemini-2.5-flash、DeepSeek-V4-flash和Qwen3-235B-A22B上保持一致。

Current read

想解决城市服务碎片化问题?这个框架把LLM和代码、API、MCP接起来,跑通跨系统任务,成功率比最强基线高10个点。

8 sources39° AI heatLast updated 2026/8/4 02:04:44

Evidence

8 sources
Primary sourcearXiv: DeepSeek
View source
Related source 5歸藏(guizang.ai)
View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.