{"slug": "robojepa-scaling-robotic-latent-world-models", "title": "RoboJEPA: Scaling Robotic Latent World Models", "summary": "Researchers submitted RoboJEPA, an 8B-parameter latent world model built on the Joint Embedding Predictive Architecture (JEPA) and trained on a dataset spanning 12 robotic embodiments, to arXiv on 7 Oct 2026. The team reports that RoboJEPA's imagination error follows a second-order power law in compute, that downstream robotic planning performance improves predictably with compute, and that the model can be deployed zero-shot as a robotic agent planning toward a single goal image for long-horizon tasks on real hardware. The authors call it the first work to establish scaling laws for multi-embodiment robotic world models trained on real robot data and the largest JEPA predictor model trained to date, and they released all model checkpoints plus training and robot deployment code.", "body_md": "# Computer Science > Artificial Intelligence\n\n  [Submitted on 7 Oct 2026]\n\n# Title:RoboJEPA: Scaling Robotic Latent World Models\n\n[View PDF](http://arxiv.org/pdf/2610.10515v1)\n\n[HTML (experimental)](https://arxiv.org/html/2610.10515v1)\n\nAbstract:Latent world models have shown a remarkable ability to predict future states and to plan in the real world. In practice, however, we lack a principled way to estimate how their capabilities scale with model size, data, and compute, an open problem that slows progress in the field. In this work we present RoboJEPA, a world model based on the Joint Embedding Predictive Architecture (JEPA) and trained on a large-scale dataset spanning 12 robotic embodiments. We show that RoboJEPA's imagination error, the error of its latent rollouts, follows a second-order power law in compute, allowing us to predict model quality well beyond the scale at which the law is fit. We further show that downstream robotic planning performance improves predictably with compute, and that imagination error is strongly correlated with it, making it a reliable proxy for real-robot evaluation. Finally, we demonstrate that latent world models can be deployed zero-shot as robotic agents, planning toward a single goal image to solve tasks requiring long-horizon planning on real hardware. We release all model checkpoints together with our training and robot deployment code. To our knowledge, this is the first work to establish scaling laws for multi-embodiment robotic world models trained on real robot data, and RoboJEPA, at 8B parameters, is the largest JEPA predictor model trained to date.\n    \n\n### Additional Features\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer \n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers \n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps \n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations \n\n*(*[What are Smart Citations?](https://www.scite.ai/))\n# Code, Data and Media Associated with this Article\n\nalphaXiv \n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers \n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub \n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub \n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face \n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast \n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))\n# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower \n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender \n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))\n# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [**Learn more about arXivLabs**](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/robojepa-scaling-robotic-latent-world-models", "canonical_source": "http://arxiv.org/abs/2610.10515v1", "published_at": "2026-10-08 14:26:05+00:00", "updated_at": "2026-10-08 14:49:36.582855+00:00", "lang": "en", "topics": ["robotics", "artificial-intelligence", "machine-learning", "ai-research"], "entities": ["RoboJEPA", "Joint Embedding Predictive Architecture", "JEPA", "arXiv"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/robojepa-scaling-robotic-latent-world-models", "markdown": "https://wpnews.pro/news/robojepa-scaling-robotic-latent-world-models.md", "text": "https://wpnews.pro/news/robojepa-scaling-robotic-latent-world-models.txt", "jsonld": "https://wpnews.pro/news/robojepa-scaling-robotic-latent-world-models.jsonld"}}