用块扩散模型提升动作生成速度与连贯性
TBD-VLA: Temporal Block Diffusion Vision Language Action Model
- 将动作序列分块，块内用掩码离散扩散，块间保持自回归
- 推理速度比传统方法快，动作轨迹时间连贯性更强
- 适合需要实时动作生成的机器人任务
离散视觉-语言-动作（VLA）模型通常将动作生成建模为在离散动作空间上的逐个词元预测，每个词元自回归地依赖于先前上下文。虽然有效，但该范式导致高推理延迟，并忽略动作轨迹中固有的时间结构。近期工作引入并行解码以提升效率，但缺乏对词元依赖关系的显式建模。本文提出TBD-VLA，一种基于块扩散的离散词元VLA框架，将动作序列划分为时间块，在每块内进行掩码离散扩散，同时保持块间的自回归生成。该设计统一了时间自回归与并行解码，实现强时间连贯性与更高推理速度。此外，显式的时间建模支持异步执行动作块（如实时分块），通过时间插补实现。TBD-VLA在仿真和真实世界操作任务中显著优于以往VLA方法，为快速、时间感知的离散VLA模型提供了可扩展路径。
原文摘要 · Abstract (English)
Discrete Vision-Language-Action (VLA) models typically formulate action generation as next-token prediction over discretized action spaces, conditioning each token autoregressively on prior context. While effective, this paradigm incurs high inference latency and largely ignores the temporal structure inherent in action trajectories. Recent efforts introduce parallel decoding to improve efficiency, enabling faster inference, but lack explicit mechanisms for modeling token dependencies. We introduce TBD-VLA, a discrete token-based VLA framework that incorporates block diffusion to enable temporal action generation. We partition action sequences into temporal blocks and perform masked discrete diffusion within each block, while maintaining autoregressive generation across blocks. This design unifies temporal autoregression and parallel action decoding, achieving both strong temporal coherence and improved inference speed. In addition, the explicit temporal modeling enables asynchronous execution of action chunks (e.g., Real-Time Chunking) via temporal in-painting. TBD-VLA significantly outperforms prior VLA approaches in both simulation and real-world manipulation tasks, offering a scalable path toward fast, temporally aware, discrete VLA models. Project webpage: https://tbd-vla.github.io/
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。