In-Context World Modeling for Robotic Control
Summary
This paper introduces In-Context World Modeling (ICWM), a framework that enables robot policies to infer system variables from self-generated interactions, allowing adaptation to novel configurations without parameter updates by treating system identification as an in-context adaptation problem. It outperforms standard VLA baselines on novel camera viewpoints in simulation and real-world experiments.
View Cached Full Text
Cached at: 06/26/26, 06:05 AM
Paper page - In-Context World Modeling for Robotic Control
Source: https://huggingface.co/papers/2606.26025
Abstract
ICWM enables robot policies to infer system variables from self-generated interactions, allowing adaptation to novel configurations without parameter updates by treating system identification as an in-context adaptation problem.
Modern Vision-Language-Action (VLA) models often fail to generalize to novel setups, such as altered camera viewpoints or robot morphologies, because they are typically conditioned only on current observations and language instructions. By ignoring the underlying system configuration as a variable, these models implicitly assume a fixed execution context encountered during training, necessitating data-intensive fine-tuning for any new environment. In this work, we introduce In-ContextWorld Modeling(ICWM), a framework that treatssystem identificationas anin-context adaptationproblem. ICWM enablesrobot policiesto autonomously infer essential system variables from a short history of self-generated,task-agnostic interactions. Unlike traditional In-Context Learning that uses demonstrations to specify what task to perform, ICWM leverages the context window to understand how the system operates. By processing these interactions before task execution, the model implicitly captures the world dynamics of the current system, enabling adaptation tonovel configurationswithoutparameter updates. Extensive experiments insimulationand onreal-world robot platformsdemonstrate that ICWM significantly outperforms standard VLA baselines on novel camera viewpoints.
View arXiv pageView PDFAdd to collection
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2606.26025 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2606.26025 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2606.26025 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
World in World: Explore the World with World Models
The paper presents World in World, a training-free interface that enables flexible camera and time control in frozen autoregressive video world models by using correspondence-guided queries and evidence-wise attention guidance.
WCM: A World Critic Model for Vision-Language-Action Reinforcement Learning
Introduces WCM, a World Critic Model that jointly predicts future latent states and estimates values to improve temporal modeling for Vision-Language-Action reinforcement learning, achieving state-of-the-art results across robotic manipulation benchmarks.
GigaWorld-1: A Roadmap to Build World Models for Robot Policy Evaluation
This paper systematically studies world models for robotic policy evaluation, introduces the WMBench benchmark and GigaWorld-1 model, and shows that long-horizon rollout consistency is more critical than short-term visual realism.
World Model for Robot Learning: A Comprehensive Survey
This comprehensive survey reviews the literature on world models for robot learning, covering their roles in policy learning, planning, and simulation. It highlights key paradigms, benchmarks, and future directions for predictive modeling in embodied agents.
Predicting Consequences and Reinforcing Navigation Policies with Latent World Models
This paper proposes a Latent World Model (LWM) for robot navigation that predicts action-conditioned latent feature compatibility, enabling policy learning from unlabeled video data and reinforcement learning without additional environment interaction, outperforming existing methods.