Kairos: 面向物理AI的原生世界模型栈

摘要

世界模型正在从被动的视觉生成器，转变为具身智能的基础性、可运行基础设施：它必须能够自然地从异构经验中获取世界知识，在长时程上维持持久状态，并在真实的部署约束下高效执行。为此，我们提出Kairos，一个围绕这些需求设计的原生世界模型技术栈。（1）Kairos通过首创的**原生预训练范式**来学习世界，该范式由**跨具身数据课程表**驱动，将开放世界视频、人类行为数据和机器人交互组织成一条渐进式发展路径。（2）Kairos通过**原生统一架构**（配备**混合线性时间注意力**）维持世界——该架构同时实现统一的世界理解、生成与预测：滑动窗口注意力捕捉局部动态，扩张滑动窗口捕捉中程依赖，而门控线性注意力维持持久的全局记忆。我们建立了形式化的理论界限，证明这种时间分解严格限制了误差累积，从数学上保证了状态在扩展时域上的传播。（3）Kairos通过集成**部署感知系统协同设计**来运行世界，支持在服务器级和消费级硬件上生成低延迟的轨迹，服务于真实世界的观测-行动-反馈循环。在具身世界模型、长时程基准和行动策略基准上的实验表明，Kairos在达到顶尖性能的同时，展现出强大的效率-能力权衡优势。综合这些结果，Kairos定位为未来自演化物理智能的一个有机运行基础。

English

World models are transitioning from passive visual generators to foundational, operational infrastructure for Physical AI: they must natively acquire world knowledge from heterogeneous experience, maintain persistent states over long horizons, and execute efficiently within real deployment constraints. We introduce Kairos, a native world model stack designed around these requirements. (1) Kairos learns the world by pioneering a Native Pre-training Paradigm governed by a Cross-Embodiment Data Curriculum, which organizes open-world videos, human behavioral data, and robot interactions into a progressive developmental pathway. (2) Kairos maintains the world by unified world understanding, generation, and prediction within a Native Unified Architecture equipped with Hybrid Linear Temporal Attention, where sliding-window attention captures local dynamics, dilated sliding windows capture mid-range dependencies, and gated linear attention maintains persistent global memory. We establish formal theoretical bounds demonstrating that this temporal factorization strictly limits error accumulation, mathematically guaranteeing state propagation across extended horizons. (3) Kairos runs the world by incorporating a Deployment-Aware System Co-Design to support low-latency rollout generation on server and consumer-grade hardware for real-world observation-action-feedback loops. Experiments on embodied world-model, long-horizon, and action-policy benchmarks show that Kairos achieves top level performance while offering a strong efficiency-capability trade-off. Together, these results position Kairos as a cohesive operational foundation for future self-evolving physical intelligence.