发表机构
Moxin Technology（魔心科技）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究旨在提升世界模型实用性与推理效率。提出MoWorld，基于3D原生数据引擎，有课程跨帧预训练等策略，能在NPU上达50 FPS实时交互，平均推理成本低，为世界模型大规模应用提供基础并展示多种应用。
AI 中文摘要
世界模型的未来不仅取决于扩展模型能力，还取决于扩展实用性和推理效率。高帧率推理可实现现实世界自主系统中的响应式感知、规划和控制。为此，我们提出了MoWorld，这是一种经济高效且高性能的快速世界模型，具有跨越数据生成、预训练、蒸馏和高效推理的端到端框架，无需高端GPU即可实现高达50 FPS的具有电影视觉质量的实时交互。为实现大规模现实世界部署，MoWorld在整个开发流程中联合优化模型能力和成本。具体而言，与主要依赖大规模视频语料库的现有方法不同，MoWorld基于从大规模3D视觉和生成建模流程积累的可扩展3D原生数据引擎构建，能够在不同的现实世界和合成环境中高效构建几何一致的训练数据。在此基础上，提出了用于稳定且可扩展的世界模型学习的课程跨帧预训练策略、用于降低扩散训练成本的高效去噪步长蒸馏算法以及用于低成本实时部署的混合精度并行推理框架。MoWorld是首个基于神经处理单元（NPU）构建的实时交互式世界模型，在这类设备中可实现高达50 FPS，为大规模实际部署提供了基础。综合评估表明MoWorld取得了领先性能，其平均推理成本仅为现有世界模型的30%-50%，为世界模型的大规模实际应用提供了实践基础。我们还展示了MoWorld的多种应用。
英文摘要
The future of World Models depends not only on scaling model capability, but also on scaling practicality and inference efficiency. High-frame-rate inference enables responsive perception, planning, and control in real-world autonomous systems. To this end, we present MoWorld, a cost-effective yet high-performance Flash World Model with an end-to-end framework spanning data generation, pre-training, distillation, and efficient inference, enabling up to 50 FPS real-time interaction with cinematic visual quality without the need of high-end GPUs. To enable large-scale real-world deployment, MoWorld jointly optimizes model capability and cost throughout the entire development pipeline. Specifically, unlike existing approaches that primarily rely on large-scale video corpora, MoWorld is built upon a scalable 3D-native data engine accumulated from our large-scale 3D vision and generative modeling pipeline, enabling the efficient construction of geometrically consistent training data across diverse real-world and synthetic environments. Based on this foundation, a curriculum cross-frame pre-training strategy for stable and scalable World Model learning, an efficient denoising-step distillation algorithm to reduce diffusion training cost, and a mixed-precision parallel inference framework for low-cost real-time deployment. MoWorld is the first real-time interactive World Model built on the Neural Processing Unit (NPU) and can achieves up to 50 FPS in such the devices, enabling practical and efficient deployment at scale. Comprehensive evaluations demonstrate that MoWorld achieves leading performance; notably, its average inference cost is only 30\%-50\% of that of existing World Models, providing a practical foundation for large-scale real-world applications of World Models. We also demonstrate diverse applications of MoWorld.
CommentsProject Page: https://moxin-tech.github.io/moworld/
