Metraipolis › archive › dc_fc9ee8fd5417
ZimaBlue: Evolving Generalizable World Action Models through Scalable Video Pre-training — arXiv:2609.00188
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.00188
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T07:10:22.043Z
- SHA-256 of the archived text
066d15bcfb9a85184023e79102f9645d73cfc8db43f62e46082178c9821aede0- Stable URL of this page
https://metraipolis.com/documento/dc_fc9ee8fd5417
How to cite
ZimaBlue: Evolving Generalizable World Action Models through Scalable Video Pre-training — arXiv:2609.00188 — arXiv (en), https://arxiv.org/abs/2609.00188 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fc9ee8fd5417 (Metraipolis archive dc_fc9ee8fd5417, sha256 066d15bcfb9a85184023e79102f9645d73cfc8db43f62e46082178c9821aede0)
BibTeX
@misc{dcfc9ee8fd5417,
title = {ZimaBlue: Evolving Generalizable World Action Models through Scalable Video Pre-training — arXiv:2609.00188},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.00188},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fc9ee8fd5417; sha256 066d15bcfb9a85184023e79102f9645d73cfc8db43f62e46082178c9821aede0},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.00188. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Robotic manipulation faces a fundamental scaling challenge: robust generalization demands broad physical experience, yet action-labeled robot trajectories are expensive to collect and inherently limited in diversity. Egocentric videos offer a far more scalable source of embodied experience, capturing object interactions, contact dynamics, tool use, and long-horizon behaviors across diverse environments. The central challenge is how to convert this abundant but action-free experience into effective robot control. We introduce ZimaBlue, a scalable framework for learning generalizable World Action Models (WAMs) from large-scale video. ZimaBlue follows a three-stage training curriculum: it first performs causal embodied video pre-training on large-scale human and robot egocentric videos, then grounds the learned visual dynamics in heterogeneous robot trajectories through video-action mid-training with a unified action representation, and finally specializes the model to a target robot for deployment. To make generative WAMs practical for real-time control, ZimaBluefurther adopts an asynchronous Slow-Fast dual-system architecture, where a high-capacity Slow world model provides generalizable spatiotemporal representations and a lightweight Fast branch enables 30 Hz action prediction on NVIDIA RTX 4090. On real-robot zero-shot evaluations, scaling from target-robot data alone to over 120,000 hours of embodied video improves success from 36.1% to 77.8%. ZimaBlue further delivers strong performance across multiple benchmarks, with particularly pronounced gains on unseen tasks. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-02 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.00188 — 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 2026-10-17T07:00:58.890Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).