Paper page - UrbanGround: From Local Perception to Spatial Agency in a Real-Scale City

Source: https://huggingface.co/papers/2608.27456 Published on Aug 27

#3 Paper of the day Authors:

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

Abstract

UrbanGround evaluates whether multimodal language model agents can sustain reliable navigation and spatial reasoning in a realistic 3D city replica, revealing that local perceptual skills fail to compose into extended goal-directed behavior.

Multimodal large language models(MLLMs) can interpret a street view, but urban agency depends on whether such local evidence remains useful after the agent starts to move. In this paper, we investigate how far currentMLLM agentscan turn local urban perception into reliable action in a complicated real-scale city. We proposeUrbanGround, the first sandbox to make this question testable in a physically constrained replica of Hong Kong built from territory-wide3D geospatial data.UrbanGroundsupportsclosed-loop interactionfrom afirst-person viewand provides an interactive map for navigation. Agents can directly enter the 3D city and explore from afirst-person view. Our analysis follows the growth of the spatial problem through three research questions. We first test whether an agent can ground a local scene well enough to answer spatial questions after active observation. Then we ask whether that grounding supports navigation as destinations become farther away and less explicit. Finally, we examine whether the resulting behavior survives changes in route availability and pedestrian motion. ContemporaryMLLM agentsusually show useful atomic abilities in visual recognition and short-rangespatial reasoning, while orientation andpedestrian-aware movementremain unreliable. Their central failure emerges over extended exploration, where local abilities do not compose into sustainedgoal-directed behaviorand errors accumulate without effective correction. We hopeUrbanGroundwill support broader study of how far currentMLLM agentscan explore reliably in complex, open-ended urban environments.

View arXiv pageView PDFProject pageGitHub17Add to collection

Get this paper in your agent:

hf papers read 2608\.27456

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2608.27456 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2608.27456 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2608.27456 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.