Astra: General Interactive World Model with Autoregressive Denoising

TL;DR

Astra is a versatile world model capable of generating long-term, interactive video predictions across diverse real-world scenarios by integrating autoregressive denoising, action-aware control, and dynamic routing of action modalities.

Contribution

The paper introduces Astra, a novel interactive world model with an autoregressive denoising architecture, action-aware adaptation, and dynamic action routing for diverse real-world applications.

Findings

Outperforms existing models in fidelity and long-range prediction.

Supports various interactions like exploration, manipulation, and camera control.

Demonstrates effectiveness across multiple datasets.

Abstract

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.