360CityArena: A Realistic Virtual Urban Navigation Benchmark for Embodied Agents
Summary
360CityArena is a photorealistic urban navigation benchmark built from 360-degree videos of Tokyo's Akihabara district. It evaluates embodied agents' environment understanding, path reasoning, and spatial reasoning, showing that even strong LMM-based agents like Gemini 2.5 Flash perform far below human level.
View Cached Full Text
Cached at: 08/12/26, 08:18 AM
Paper page - 360CityArena: A Realistic Virtual Urban Navigation Benchmark for Embodied Agents
Source: https://huggingface.co/papers/2608.08814
Abstract
A new photorealistic urban benchmark reveals large performance gaps for embodied agents in city-scale navigation and spatial reasoning.
We present 360CityArena, a benchmark for evaluating the urban exploration capabilities ofembodied agentswithin aphotorealistic environmentconstructed from360-degree videos. Existing outdoor benchmarks either lack sufficient photorealism or complexity, resulting in a considerable gap from real-world urban environments. 360CityArena is built on a realistic reconstruction of the Akihabara district in Tokyo, Japan, using 602 360-degree video segments covering 85 streets, and consists of 175 meticulously human-crafted tasks. It encompasses three task categories: Environment Understanding, Path Reasoning, andSpatial Reasoning, covering fundamental abilities required for urban exploration, such as localization,landmark search,path planning, and relationalspatial reasoning, thereby enabling comprehensive evaluation in realistic urban scenes. Our evaluation using state-of-the-artLMM-based agentsshows that even the strongest model, Gemini 2.5 Flash, performs far below human level (human: 77.3% vs. Gemini 2.5 Flash: 17.1%), revealing substantial challenges that remain incity-scale embodied navigationand reasoning. 360CityArena provides a necessary and challenging testbed for photorealistic urban-district navigation andspatial reasoning.
View arXiv pageView PDFProject pageGitHub5Add to collection
Get this paper in your agent:
hf papers read 2608\.08814
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/2608.08814 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2608.08814 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2608.08814 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.
Similar Articles
WildCity: A Real-World City-Scale Testbed for Rendering, Simulation, and Spatial Intelligence
WildCity introduces a large-scale multimodal dataset for city-scale urban navigation and spatial representation, collected by autonomous fleets. It provides 18 long trajectories and establishes baselines for reconstruction and closed-loop simulation to advance AI systems that can perceive and reason about city-scale environments.
CityRAG: Stepping Into a City via Spatially-Grounded Video Generation
CityRAG introduces a video generative model that produces long, physically grounded, 3D-consistent videos of real-world cities using geo-registered data, enabling realistic navigation and simulation for robotics and autonomous driving.
SpatialWorld: Benchmarking Interactive Spatial Reasoning of Multimodal Agents in Real-World Tasks
SpatialWorld is a unified benchmark for evaluating interactive spatial reasoning in multimodal agents across diverse real-world tasks, revealing that even the strongest models achieve low task success rates.
UrbanAgent: A Tool-Augmented Agent for Cross-System Urban Tasks
UrbanAgent is a tool-augmented agent framework that uses LLMs with code execution, API calls, and MCP to handle cross-system urban requests. The authors also introduce UrbanEval, a benchmark for evaluating task results and execution quality, achieving 71% success rate over baselines.
CityBehavEx: A Scalable and Empirically Validated LLM-Assisted Urban Simulation Platform
CityBehavEx is a scalable LLM-assisted urban simulation platform that combines established human mobility models with fine-tuned cross-encoders to generate realistic, empirically validated mobility patterns for city-sized populations, demonstrating 100,000 agents over 75 days in under one hour on a single consumer GPU.