Metraipolis › archive › dc_545df39f456b
WorldAgent: Verification-Guided Agentic Physical World Construction — arXiv:2609.33208
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.33208
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T20:56:32.324Z
- SHA-256 of the archived text
e80315a5c98f172a96eeaaef5511ab953165bb585e5db383e99c16976ece3e46- Stable URL of this page
https://metraipolis.com/documento/dc_545df39f456b
How to cite
WorldAgent: Verification-Guided Agentic Physical World Construction — arXiv:2609.33208 — arXiv (en), https://arxiv.org/abs/2609.33208 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_545df39f456b (Metraipolis archive dc_545df39f456b, sha256 e80315a5c98f172a96eeaaef5511ab953165bb585e5db383e99c16976ece3e46)
BibTeX
@misc{dc545df39f456b,
title = {WorldAgent: Verification-Guided Agentic Physical World Construction — arXiv:2609.33208},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.33208},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_545df39f456b; sha256 e80315a5c98f172a96eeaaef5511ab953165bb585e5db383e99c16976ece3e46},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33208. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Constructing complex physical worlds from language requires coordinating extensive 3D environments, detailed structures and objects at different spatial scales, and interacting physical processes under both stated goals and implicit physical constraints. We present WorldAgent, an agentic framework for verification-guided physical world construction from a single natural-language prompt, without iterative user debugging. A world construction layer expands the prompt into a structured world specification and uses physical knowledge to build scenes and run numerical simulations. After every step, a verification layer inspects scene geometry and simulation states alongside rendered views. Failed checks guide automatic revisions to the specification and re-execution of the affected steps. Accepted worlds pass the required checks and remain editable for further inspection and resimulation. We introduce AgenticSimBench, on which WorldAgent achieves the best scores among the evaluated agent-based methods on five of seven metrics. In a 26-participant user study, it receives the highest mean ratings across all four criteria. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI) ; Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.33208 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-07T20:29:01.176Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).