用缓存复用历史动作，让视觉语言动作模型推理快10倍以上。
ActionCache: Training-Free Acceleration for Vision-Language-Action Models with Action Caching and Refinement
- 构建外部缓存，用多模态键值对存储历史中间动作。
- 在仿真和真实场景中实现最高40.17倍的推理加速，成功率不降。
- 无需训练，即插即用，适合实时机器人控制应用。
视觉-语言-动作（VLA）模型为通用机器人操作提供了新思路。基于流匹配的VLA模型因其能生成精确平滑的动作序列并捕捉多模态分布而表现优异。然而，动作头中的迭代去噪过程成为主要计算瓶颈，阻碍实时部署。为此，我们提出ActionCache，一种即插即用的外部缓存机制，通过有选择地重用过去的中间动作来热启动生成，显著降低推理延迟。具体而言，ActionCache以紧凑的多模态键存储中间动作，可在不同任务或不同回合中检索相似上下文下的动作。实验结果表明，在仿真和真实环境中，ActionCache在低延迟条件下仍保持高任务成功率，使代表性流式VLA模型π_{0.5}和GR00T-N1.6的动作头推理加速分别达10.44倍和40.17倍。
原文摘要 · Abstract (English)
Vision-Language-Action (VLA) models have emerged as a promising approach for generalizable robotic manipulations. In particular, flow-matching-based VLA models have shown remarkable success due to their capability to generate precise and smooth action sequences and capture multimodal distributions. However, the iterative denoising process in the action head acts as a major computational bottleneck, posing a critical challenge for real-time deployment. To address this challenge, we propose ActionCache, a plug-and-play external cache that opportunistically reuses past intermediate actions to warm-start generations from the vicinity of target actions, drastically reducing the inference latency. Specifically, ActionCache stores the intermediate actions with compact multimodal keys, which enables retrieval from similar past contexts across different episodes or even different tasks. Experimental results in simulation and real-world environments demonstrate that ActionCache maintains high task success rates in a low-latency regime, achieving action head inference acceleration of up to $10.44\times$ and $40.17\times$ for representative flow-based VLA, $π_{0.5}$ and GR00T-N1.6, respectively.
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。