Vision-Language-Action (VLA) models have demonstrated strong performance in robotic manipulation, yet their closed-loop deployment is hindered by the high latency and compute cost of repeatedly running large vision-language backbones at every timestep. We observe that VLA inference exhibits structured redundancies across temporal, spatial, and depth dimensions, and that most existing efficiency methods ignore action context, despite its central role in embodied tasks. To address this gap, we propose Action-Context-aware Adaptive Computation for VLA models (AC^2-VLA), a unified framework that conditions computation on current visual observations, language instructions, and previous action states. Based on this action-centric context, AC^2-VLA adaptively performs cognition reuse across timesteps, token pruning, and selective execution of model components within a unified mechanism. To train the adaptive policy, we introduce an action-guided self-distillation scheme that preserves the behavior of the dense VLA policy while enabling structured sparsification that transfers across tasks and settings. Extensive experiments on robotic manipulation benchmarks show that AC^2-VLA achieves up to a 1.79\times speedup while reducing FLOPs to 29.4% of the dense baseline, with comparable task success.

翻译：视觉-语言-动作（VLA）模型在机器人操作任务中已展现出强大性能，但其闭环部署受到高延迟和高计算成本的阻碍，原因在于需要在每个时间步重复运行大型视觉-语言主干网络。我们观察到，VLA推理在时间、空间和深度维度上存在结构化冗余，且大多数现有效率优化方法忽视了动作上下文，尽管其在具身任务中扮演核心角色。为弥补这一不足，我们提出了面向VLA模型的动作-上下文感知自适应计算框架（AC^2-VLA），这是一个统一框架，其计算过程以当前视觉观测、语言指令以及先前的动作状态为条件。基于这种以动作为核心的上下文，AC^2-VLA通过统一机制，自适应地执行跨时间步的认知重用、令牌剪枝以及模型组件的选择性执行。为训练该自适应策略，我们引入了一种动作引导的自蒸馏方案，该方案在保持稠密VLA策略行为的同时，实现了可跨任务和场景迁移的结构化稀疏化。在机器人操作基准测试上的大量实验表明，AC^2-VLA在取得可比任务成功率的同时，实现了高达1.79倍的加速，并将FLOPs降至稠密基线模型的29.4%。