用分块+专家选择，让机器人视频生成更连贯持久
LongScape: Advancing Long-Horizon Embodied World Models with Context-Aware MoE
- 按动作语义分块视频，每块独立生成更合理
- 跨块自回归+块内扩散，生成100步以上仍稳定
- 适合需要长时序动作生成的机器人仿真研究
基于视频的世界模型在生成高质量具身操作数据方面潜力巨大。然而，现有视频生成方法难以实现稳定的长时序生成：经典扩散模型常出现时间不一致和视觉漂移，而自回归方法则牺牲了视觉细节。为此，我们提出LongScape，一种混合框架，自适应结合块内扩散去噪与块间自回归因果生成。核心创新是动作引导的变长分块机制，根据机器人动作的语义上下文划分视频，确保每块代表一个完整连贯的动作，支持灵活生成多样化动态。我们进一步设计上下文感知的多专家（CMoE）框架，生成时自适应激活各块的专用专家，保障高视觉质量并实现块间无缝衔接。大量实验表明，该方法可在长时间滚动中保持稳定一致的生成效果。代码已开源：https://github.com/tsinghua-fib-lab/Longscape。
原文摘要 · Abstract (English)
Video-based world models hold significant potential for generating high-quality embodied manipulation data. However, current video generation methods struggle to achieve stable long-horizon generation: classical diffusion-based approaches often suffer from temporal inconsistency and visual drift over multiple rollouts, while autoregressive methods tend to compromise on visual detail. To solve this, we introduce LongScape, a hybrid framework that adaptively combines intra-chunk diffusion denoising with inter-chunk autoregressive causal generation. Our core innovation is an action-guided, variable-length chunking mechanism that partitions video based on the semantic context of robotic actions. This ensures each chunk represents a complete, coherent action, enabling the model to flexibly generate diverse dynamics. We further introduce a Context-aware Mixture-of-Experts (CMoE) framework that adaptively activates specialized experts for each chunk during generation, guaranteeing high visual quality and seamless chunk transitions. Extensive experimental results demonstrate that our method achieves stable and consistent long-horizon generation over extended rollouts. Our code is available at: https://github.com/tsinghua-fib-lab/Longscape.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。