Recently, 3D vision-based diffusion policies have shown strong capability in learning complex robotic manipulation skills. However, a common architectural mismatch exists in these models: a tiny yet efficient point-cloud encoder is often paired with a massive decoder. Given a compact scene representation, we argue that this may lead to substantial parameter waste in the decoder. Motivated by this observation, we propose PocketDP3, a pocket-scale 3D diffusion policy that replaces the heavy conditional U-Net decoder used in prior methods with a lightweight Diffusion Mixer (DiM) built on MLP-Mixer blocks. This architecture enables efficient fusion across temporal and channel dimensions, significantly reducing model size. Notably, without any additional consistency distillation techniques, our method supports two-step inference without sacrificing performance, improving practicality for real-time deployment. Across three simulation benchmarks--RoboTwin2.0, Adroit, and MetaWorld--PocketDP3 achieves state-of-the-art performance with fewer than 1% of the parameters of prior methods, while also accelerating inference. Real-world experiments further demonstrate the practicality and transferability of our method in real-world settings. Code will be released.

翻译：近年来，基于三维视觉的扩散策略在习得复杂机器人操作技能方面展现出强大能力。然而，这些模型普遍存在一个架构上的不匹配问题：一个微小而高效的点云编码器往往与一个庞大的解码器配对。考虑到紧凑的场景表示，我们认为这可能导致解码器存在显著的参数浪费。受此观察启发，我们提出了PocketDP3，一种掌上规模的三维扩散策略。该方法将先前方法中使用的重型条件U-Net解码器替换为基于MLP-Mixer模块构建的轻量级扩散混合器（DiM）。该架构能够高效融合时间和通道维度，显著减小模型规模。值得注意的是，在无需任何额外一致性蒸馏技术的情况下，我们的方法支持两步推理且不牺牲性能，从而提升了实时部署的实用性。在三个仿真基准测试——RoboTwin2.0、Adroit和MetaWorld——中，PocketDP3以少于先前方法1%的参数实现了最先进的性能，同时加速了推理。真实世界实验进一步证明了我们方法在实际场景中的实用性和可迁移性。代码将予以发布。