Urbex: Agentic Spatial Grounding for City-Scale 3D Scenes

City-scale 3D grounding is commonly formulated as candidate ranking, which requires precomputed instance proposals and becomes inefficient in large outdoor scenes.

By Wang, Dai, Chen +5Oct 13, 2026

Score██████░░░░5.7

Caveats

Preprint; not yet peer reviewed.

VerdictCompetent work. Briefs at most.

Read the original PDF Code

Abstract

City-scale 3D grounding is commonly formulated as candidate ranking, which requires precomputed instance proposals and becomes inefficient in large outdoor scenes. We propose Urbex, an agentic vision-language framework that instead casts grounding as active spatial search. Urbex represents a city-scale 3D scene as an interactive multi-view environment, allowing a VLM agent to query landmarks, zoom into local regions, render oblique views, and commit a grounded bounding box. We optimize this tool-use policy with reinforcement learning on CityRefer, using smooth localization rewards and lightweight shaping rewards for efficient exploration. Experiments show that Urbex achieves the best Hit@1 and IoU on CityRefer, supports candidate-free grounding without ground-truth boxes, and is substantially faster than the strongest city-scale ranking baseline. Zero-shot results on STPLS3D-Refer further indicate improved cross-scene robustness.

Yu Wang, Rui Dai, Wei Chen, Yi Wang, Bo Dang, Kaikui Liu, Xiangxiang Chu, Yansheng Li

The editor's rubric

Heuristic review

Dimension	Level	Weight	What that level means
Leverage	████░ 4	24%	A general-purpose tool used across several fields (Adam, ResNet, LoRA, next-generation sequencing).
Magnitude	██░░░ 2	18%	Solid incremental gain on a meaningful problem.
Evidence	██░░░ 2	14%	Limited: single setting, weak baselines, or an observational association presented as causal.
Novelty	██░░░ 2	16%	A new combination of known ideas.
Trajectory	███░░ 3	18%	A clear path to scale.
Stakes	██░░░ 2	10%	Benefits a professional community (practitioners, clinicians, engineers).

Editor’s rationale

Heuristic triage from title and abstract text only, not a reading of the paper. Cues found: method (we propose); breadth (zero/few-shot); verification (code released); scale (scalable, efficient); stakes (general AI).

How the score was computed

rank-2026-10-07

Score██████░░░░5.7

Score = 10 × (65% × adjusted merit / 10 + 25% × attention + 10% × freshness)

Merit: 4.4 / 10; Weighted rubric, evidence-gated.
Adjusted merit: 4.2 / 10; Shrunk toward the desk prior by editor confidence (40%).
Attention: 89%; Citations, upvotes, points, mentions.
Freshness: 75%; Half-life decay since publication.

Hugging Face upvotes103 (reference 25, via hf-daily, Oct 14, 2026, 13:49 UTC)
GitHub stars3 (reference 250, via hf-daily, Oct 14, 2026, 13:49 UTC)

The record

Reviewed by heuristic-v6 on Oct 14, 2026, 13:49 UTC. Paper type: method.
Categories: cs.CV, cs.AI
TOP, No.2 in the Artificial Intelligence edition of October 14, 2026.