ABot-M0.5：统一移动与操控世界动作模型

摘要

移动操作是实现通用型机器人的关键能力，但对于当前的具身学习方法而言仍极具挑战性。视觉-语言-动作（VLA）策略通常是被动的，缺乏显式的世界建模，而现有的世界动作模型（WAM）在移动操作的结构对齐方面仍存在明显不足：它们以粗糙的视频片段为单位运行，将导航与操作动作纠缠建模，并在与自回归推理不匹配的监督下训练逆动力学模型。因此，这些模型往往缺失细粒度的接触动力学信息，面临动作分布冲突问题，并在长时程执行中累积误差。我们提出ABot-M0.5，该新WAM基于以下洞察构建：移动操作需要在三个层次实现对齐——时间粒度、动作空间以及训练与测试一致性。为对齐时间粒度，我们引入中间潜在动作，捕捉局部视觉状态转移，并作为视频潜在表示与具身特定控制之间的桥梁动作空间。为对齐动作空间，我们设计了一种双层混合Transformer架构，将模态表征与异构动作子空间（如基座移动与手臂操纵）进行解耦。为对齐推理条件，我们提出“梦想强制”训练策略，在模型预测的视频上逐步训练逆动力学模型，从而提升自回归预测中的训练-测试对齐性与鲁棒性。在具有挑战性的移动操作与细粒度操作基准测试上的实验表明，ABot-M0.5在长时程任务成功率与细粒度控制精度上均达到最先进水平。这些结果突显了粒度对齐、动作解耦与推理一致性在世界-动作建模中的关键重要性。

English

Mobile manipulation is a key capability for general-purpose robots, yet remains challenging for current embodied learning methods. VLA policies are typically reactive and lack explicit world modeling, while existing World Action Models (WAMs) are still poorly aligned with the structure of mobile manipulation: they operate on coarse video chunks, model entangled navigation-manipulation actions, and train inverse dynamics under supervision that does not match autoregressive inference. As a result, they often miss fine-grained contact dynamics, suffer from action-distribution conflicts, and accumulate errors over long-horizon rollouts. We propose ABot-M0.5, a new WAM built on the insight that mobile manipulation requires alignment at three levels: temporal granularity, action space, and train-test consistency. To align temporal granularity, we introduce intermediate latent actions that capture local visual state transitions and serve as an bridging action space between video latents and embodiment-specific controls. To align action space, we design a dual-level Mixture-of-Transformers architecture that disentangles both modality representations and heterogeneous action subspaces such as base movement and arm manipulation. To align inference conditions, we propose the dream-forcing training strategy that progressively trains inverse dynamics on model-predicted videos, improving train-test alignment and robustness during autoregressive prediction. Experiments on challenging mobile and fine-grained manipulation benchmarks demonstrate that ABot-M0.5 achieves state-of-the-art performance in both long-horizon task success and finegrained control accuracy. These results highlight the critical importance of granularity-aligned, action-disentangled, and inference-consistent world-action modeling.