Astra：通用交互世界模型与自回归去噪
Astra: General Interactive World Model with Autoregressive Denoising
AI总结：
Astra提出了一种通用交互世界模型，通过自回归去噪和动作感知适配器实现长周期视频预测与多样化交互。
摘要：
最近的扩散变压器进展使视频生成模型能够从文本或图像生成高质量的视频片段。然而，能够从过去观察和行动中预测长周期未来的世界模型仍然处于探索阶段，特别是在通用场景和各种行动形式中。为填补这一差距，我们引入了Astra，一种交互式通用世界模型，能够为多样化的场景（例如自动驾驶、机器人抓取）生成真实的未来。我们提出了一种自回归去噪架构，并利用时间因果注意机制来聚合过去的观察并支持流式输出。我们使用噪声增强的历史记忆来避免过度依赖过去帧，以平衡响应速度与时间一致性。为了实现精确的动作控制，我们引入了动作感知的适配器，它直接将动作信号注入去噪过程中。我们进一步开发了动作专家的混合模型，动态路由异构的动作模态，从而在探索、操作和摄像头控制等多样化的现实任务中增强通用性。Astra实现了交互式、一致且通用的长期视频预测，并支持各种形式的交互。在多个数据集上的实验表明，Astra在保真度、长程预测和动作对齐方面优于现有最先进的世界模型。
英文摘要：
Recent advances in diffusion transformers have empowered video generation models to generate high-quality video clips from texts or images. However, world models with the ability to predict long-horizon futures from past observations and actions remain underexplored, especially for general-purpose scenarios and various forms of actions. To bridge this gap, we introduce Astra, an interactive general world model that generates real-world futures for diverse scenarios (e.g., autonomous driving, robot grasping) with precise action interactions (e.g., camera motion, robot action). We propose an autoregressive denoising architecture and use temporal causal attention to aggregate past observations and support streaming outputs. We use a noise-augmented history memory to avoid over-reliance on past frames to balance responsiveness with temporal coherence. For precise action control, we introduce an action-aware adapter that directly injects action signals into the denoising process. We further develop a mixture of action experts that dynamically route heterogeneous action modalities, enhancing versatility across diverse real-world tasks such as exploration, manipulation, and camera control. Astra achieves interactive, consistent, and general long-term video prediction and supports various forms of interactions. Experiments across multiple datasets demonstrate the improvements of Astra in fidelity, long-range prediction, and action alignment over existing state-of-the-art world models.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
