# InternW0：用于高效真实世界交互的基础物理世界模型

> 原文链接：https://robotics.hellobike.com/papers/2026-09-23-2609-27656
> 频道：papers | 更新日期：2026-09-23 | 标签：arxiv, cs.RO
> 核心摘要：InternW0通过全模态接口、异步多频处理和局部物理建模，联合学习视觉动态与机器人控制，并在仿真基准和真实科学任务中验证。

# InternW0：用于高效真实世界交互的基础物理世界模型

::: v-pre
> 原文链接：[InternW0: A Foundational Physical World Model for Efficient Real-World Interactions](https://arxiv.org/abs/2609.27656)（arXiv 2609.27656 · 发布日期：2026-09-23）

**InternW0: A Foundational Physical World Model for Efficient Real-World Interactions**

InternW0通过全模态接口、异步多频处理和局部物理建模，联合学习视觉动态与机器人控制，并在仿真基准和真实科学任务中验证。

## 核心要点

- 动机在于物理智能的预测需在世界持续变化时保持可执行，而不只是预测演化。
- 方法采用非对称视频-动作架构与flow matching，联合学习未来视觉动态和连续机器人控制。
- 复用逐层K/V并通过观测条件上下文路由适配新状态，接口和软提示支持异构具身。
- 约7200小时异构机器人与自我中心数据训练，评估覆盖仿真基准和真实科学任务。

**验证范围与局限**：验证范围包括仿真基准和真实科学任务：15阶段金属有机框架合成流程，以及5阶段接触和力感知灵巧移液操作。

## 英文摘要 (Abstract)

Physical intelligence requires more than predicting how the world may evolve: predictions must remain actionable as the world continues to change. We introduce InternW0, the first instantiation of the InternW physical world model series from Shanghai AI Laboratory, built around omnimodal interfaces, asynchronous multi-frequency processing, and local physical modeling under partial observations and external influences. InternW0 jointly learns future visual dynamics and continuous robot control through an asymmetric video--action architecture with flow matching. A high-capacity video expert provides longer-horizon predictive context, while a lightweight action expert operates at a faster timescale. Instead of regenerating the future for every action update, InternW0 reuses layerwise K/V and adapts it to newly observed states through observation-conditioned context routing. Domain-specific interfaces and soft prompts support heterogeneous embodiments, while contact-aware post-training incorporates force and tactile signals for contact-rich manipulation. We train InternW0 on approximately 7,200 hours of heterogeneous robot and egocentric data, including EgoLab, a 275-hour real-laboratory egocentric dataset. Evaluation spans simulation benchmarks and real-world scientific tasks, including a 15-stage metal--organic framework synthesis workflow and 5-stage contact- and force-aware dexterous manipulation for general-purpose quantitative pipetting. These results advance scalable, asynchronous, and science-native physical world models for universal and efficient real-world interactions.

作者：Jisong Cai, Yao Mu, Ganlin Yang, Zhe Cao, Zhangzheng Tu, Xing Gao, Kailin Li, Xinyu Zhan, Lixin Yang, Yangkun Zhu
:::

## 原始参考来源 (Sources)

- [arXiv 2609.27656：InternW0: A Foundational Physical World Model for Efficient Real-World Interactions](https://arxiv.org/abs/2609.27656)