AR-VLA: True Autoregressive Action Expert for Vision-Language-Action Models

TL;DR

AR-VLA introduces a novel autoregressive action model for vision-language-action tasks, enabling context-aware, smooth, and scalable robotic policy generation with superior history awareness.

Contribution

It presents a standalone autoregressive Action Expert that maintains history, addresses temporal frequency mismatch, and integrates seamlessly with perception backbones.

Findings

Outperforms reactive VLAs in smoothness and history awareness.

Effectively replaces chunk-based action heads in manipulation tasks.

Maintains or exceeds state-of-the-art task success rates.

Abstract

We propose a standalone autoregressive (AR) Action Expert that generates actions as a continuous causal sequence while conditioning on refreshable vision-language prefixes. In contrast to existing Vision-Language-Action (VLA) models and diffusion policies that reset temporal context with each new observation and predict actions reactively, our Action Expert maintains its own history through a long-lived memory and is inherently context-aware. This structure addresses the frequency mismatch between fast control and slow reasoning, enabling efficient independent pretraining of kinematic syntax and modular integration with heavy perception backbones, naturally ensuring spatio-temporally consistent action generation across frames. To synchronize these asynchronous hybrid V-L-A modalities, we utilize a re-anchoring mechanism that mathematically accounts for perception staleness during both training and inference. Experiments on simulated and real-robot manipulation tasks demonstrate that the proposed method can effectively replace traditional chunk-based action heads for both specialist and generalist policies. AR-VLA exhibits superior history awareness and substantially smoother action trajectories while maintaining or exceeding the task success rates of state-of-the-art reactive VLAs. Overall, our work introduces a scalable, context-aware action generation schema that provides a robust structural foundation for training effective robotic policies. Code and Videos available at https://arvla.insait.ai