A Tutorial on Physical AI from a Multimedia Perspective: Part II Cosmos Foundation Model
ACM Transactions on Multimedia Computing, Communications, and Applications · 06.11.2026
DOI: 10.1145/3862336 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM Transactions on Multimedia Computing, Communications, and Applications, online erschienen am 06.11.2026. Verfasst von Yang Liu und 2 weiteren Autor:innen. Die Arbeit nennt 33 Literaturangaben.
Abstract (Originalsprache)
Physical AI aims to enable systems to perceive, reason, and act in the unstructured physical world, yet it faces bottlenecks of scarce real-world data and high deployment risks. Traditional hand-engineered simulators provide controllable environments yet struggle with scalable coverage of diverse scenes and long-tail scenarios. Cosmos addresses these challenges as a data-driven world foundation model platform, unifying generation- verification loop to form a closed-loop data flywheel for Physical AI development. This tutorial systematically introduces Cosmos's multi-modal design, core components, and practical deployment, offering a comprehensive guide for leveraging learned world models in physical system training and evaluation.
Zitieren
APA
Liu, Y., Dong, H., & Saddik, A. E. (2026). A Tutorial on Physical AI from a Multimedia Perspective: Part II Cosmos Foundation Model. ACM Transactions on Multimedia Computing, Communications, and Applications. https://doi.org/10.1145/3862336
Vancouver
Liu Y, Dong H, Saddik AE. A Tutorial on Physical AI from a Multimedia Perspective: Part II Cosmos Foundation Model. ACM Transactions on Multimedia Computing, Communications, and Applications. 2026. doi:10.1145/3862336
BibTeX
@article{Liu2026,
author = {Liu, Yang and Dong, Haiwei and Saddik, Abdulmotaleb El},
title = {A Tutorial on Physical AI from a Multimedia Perspective: Part II Cosmos Foundation Model},
journal = {ACM Transactions on Multimedia Computing, Communications, and Applications},
year = {2026},
doi = {10.1145/3862336}
}
EUVIMED