TBD-VLA: 時間ブロック拡散を用いた視覚言語行動モデル
TBD-VLA: Temporal Block Diffusion Vision Language Action Model
離散トークン型の視覚言語行動モデル(VLA)において、行動系列を時間ブロックに分割し、ブロック内ではマスク拡散、ブロック間では自己回帰生成を行うことで、時間的一貫性と推論速度を両立する新しいフレームワークを提案した。シミュレーションと実機操作タスクで既存手法を上回る性能を示した。
著者: Sung-Wook Lee, Xuhui Kang, Yen-Ling Kuo
分類: cs.CV, cs.RO
原文アブストラクト
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/