AtomVLA: 通过预测性潜在世界模型实现机器人操作的可扩展后训练
AtomVLA: Scalable Post-Training for Robotic Manipulation via Predictive Latent World Models
AI总结：
AtomVLA通过预测性潜在世界模型实现机器人操作的可扩展后训练，提升长时间任务的鲁棒性与效率。
摘要：
视觉-语言-动作（VLA）模型在通用化机器人操作中展现出显著潜力。在VLA模型中执行复杂多步骤行为可以通过稳健的指令 grounding 来提高，这是有效控制的关键组件。然而，当前范式主要依赖于监督微调期间的粗略、高层任务指令。这个指令 grounding 的差距导致模型没有明确的中间指导，从而在长时间跨度任务中产生严重的叠加错误。因此，弥合这个指令差距并为VLA模型提供可扩展的后训练是紧迫的。为了解决这个问题，我们提出了\method，这是第一个子任务感知的VLA框架，集成了可扩展的离线后训练流水线。我们的框架利用大型语言模型将高层演示分解为细粒度的原子子任务。这种方法利用预训练的预测性世界模型，在潜在空间中对候选动作块进行评分，以符合子任务目标，从而缓解错误累积，同时显著提高长时间跨度的鲁棒性。此外，这种方法使高效率的组相对策略优化成为可能，而无需与在物理机器人上进行在线 rollout 相关的高昂成本。广泛的模拟验证表明，我们的AtomVLA在扰动下保持了强大的鲁棒性。当与基础基线模型进行评估时，在LIBERO基准上实现了97.0%的平均成功率，在LIBERO-PRO基准上实现了48.0%的成功率。最后，使用Galaxea R1 Lite平台在现实世界中进行的实验证实了其在多样任务中的广泛应用，尤其是长时间跨度的任务。所有数据集、检查点和代码将在本工作被接受后发布到公共领域，以供未来研究使用。
英文摘要：
Vision-Language-Action (VLA) models demonstrate remarkable potential for generalizable robotic manipulation. The execution of complex multi-step behaviors in VLA models can be improved by robust instruction grounding, a critical component for effective control. However, current paradigms predominantly rely on coarse, high-level task instructions during supervised fine-tuning. This instruction grounding gap leaves models without explicit intermediate guidance, leading to severe compounding errors in long-horizon tasks. Therefore, bridging this instruction gap and providing scalable post-training for VLA models is urgent. To tackle this problem, we propose \method, the first subtask-aware VLA framework integrated with a scalable offline post-training pipeline. Our framework leverages a large language model to decompose high-level demonstrations into fine-grained atomic subtasks. This approach utilizes a pretrained predictive world model to score candidate action chunks against subtask goals in the latent space, mitigating error accumulation while significantly improving long-horizon robustness. Furthermore, this approach enables highly efficient Group Relative Policy Optimization without the prohibitive expenses associated with online rollouts on physical robots. Extensive simulations validate that our AtomVLA maintains strong robustness under perturbations. When evaluated against fundamental baseline models, it achieves an average success rate of 97.0\% on the LIBERO benchmark and 48.0\% on the LIBERO-PRO benchmark. Finally, experiments conducted in the real world using the Galaxea R1 Lite platform confirm its broad applicability across diverse tasks, especially long-horizon tasks. All datasets, checkpoints, and code will be released to the public domain following the acceptance of this work for future research.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
