{"slug": "world-in-context-learning-for-agentic-robot-manipulation", "title": "World In-Context Learning for Agentic Robot Manipulation", "summary": "A paper submitted to arXiv on 9 Nov 2026 introduces World In-Context Learning (WICL), a paradigm in which a frozen model rehearses a task in a simulator and writes the feedback into its context rather than its weights, implemented as RoboScript. On all 50 tasks of RoboTwin 2.0, RoboScript reached 83.4% mean success across clean and randomized settings with one demonstration per task and no gradient updates, exceeding the strongest trained action policy by over 20 points, and it transferred to a physical dual-arm robot through cross-embodiment memory adaptation. The code is released on GitHub under ArcheOne-AI/RoboScript.", "body_md": "# Computer Science > Computer Vision and Pattern Recognition\n\n  [Submitted on 9 Nov 2026]\n\n# Title:World In-Context Learning for Agentic Robot Manipulation\n\n[View PDF](http://arxiv.org/pdf/2611.12885v1)\n\nAbstract:Trained action policies write manipulation skill into network weights, tying what is learned to the distribution it is learned from. In-context learning (ICL) avoids this coupling but has not matched trained policies, because a demonstration recorded in another scene carries no feedback from the world the robot currently faces. We observe that what an agent learns in context is semantic, recording task steps and object roles rather than visual appearance, so experience gathered in one world transfers to another. Based on this insight, we propose World In-Context Learning (WICL), a paradigm in which a frozen model starts from a demonstration, rehearses the task in a world such as a simulator, and writes the feedback into its context rather than its weights. We build RoboScript, a modular infrastructure that implements WICL for agentic manipulation. From a single demonstration, RoboScript learns each task through rehearsal in simulation, freezes the learned knowledge as a Task Memory, and deploys the frozen memory on new scenes with data isolation that ensures every deployment input is available on a real robot. On all 50 tasks of RoboTwin 2.0, RoboScript reaches 83.4% mean success across clean and randomized settings with one demonstration per task and no gradient updates, exceeding the strongest trained action policy by over 20 points. The method generalizes across visual distributions, improves through reflection, scales with the foundation model, and transfers to a physical dual-arm robot through cross-embodiment memory adaptation. With a world in the loop, a robot can learn a manipulation task in its context as well as it learns one in its weights. All code is available at [this https URL](https://github.com/ArcheOne-AI/RoboScript)\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/world-in-context-learning-for-agentic-robot-manipulation", "canonical_source": "http://arxiv.org/abs/2611.12885v1", "published_at": "2026-11-10 15:26:52+00:00", "updated_at": "2026-11-10 15:52:57.993317+00:00", "lang": "en", "topics": ["robotics", "machine-learning", "ai-research", "ai-agents"], "entities": ["World In-Context Learning", "RoboScript", "RoboTwin 2.0", "ArcheOne-AI", "arXiv"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/world-in-context-learning-for-agentic-robot-manipulation", "markdown": "https://wpnews.pro/news/world-in-context-learning-for-agentic-robot-manipulation.md", "text": "https://wpnews.pro/news/world-in-context-learning-for-agentic-robot-manipulation.txt", "jsonld": "https://wpnews.pro/news/world-in-context-learning-for-agentic-robot-manipulation.jsonld"}}