{
  "id": 3273,
  "url": "https://arxiv.org/abs/2606.02753v1",
  "title": "MetaWorld: Scaling Multi-Agent Video World Model from Single-view Video Data",
  "summary": "Video world models are a foundational generative technology for embodied AI and the Metaverse, yet existing approaches are inherently limited to a single agent observing from a single perspective. Extending these models to multi-agent settings introduces two critical challenges: data scarcity (coordinated multi-view recordings are prohibitively expensive to collect for general open-domain scenarios) and world state alignment (independently generated video streams cannot ensure that shared physic",
  "authors": "Teng Hu, Mingchun Lu, Yating Wang, Jiangning Zhang, Jinkun Hao, Ye Pan et al.",
  "category": "research",
  "topics": "safety-alignment,agents-autonomy",
  "orgs": null,
  "regions": null,
  "published_at": "2026-06-01T18:20:20.000Z",
  "fetched_at": "2026-07-14T16:30:09.959Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/3273",
  "original_url": "https://arxiv.org/abs/2606.02753v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}