Visual-language action (VLA) models enable robots to predict actions directly from observations and language instructions, but their performance depends on large-scale, high-quality data and is limited by the scarcity of real-world robot action datasets. To facilitate VLA model learning with abundant unlabeled human videos, Latent Action Models (LAM) learn latent action representations from visual dynamics to provide additional supervision for VLA learning. However, LAM and VLA are typically trained separately, leaving LAM ungrounded during VLA training and VLA models constrained by frozen LAM representations. To address these issues, we propose Latent Action Representation Alignment (LARA), a plug-and-play framework that jointly optimizes LAM and VLA via representation alignment. This enables reciprocal benefits where LAMs learn with action trajectories to avoid spurious visual changes, while VLAs are regularized by forward dynamics learned within LAMs to reduce hallucinations of functionally ineffective trajectories. We demonstrate LARA versatility and effectiveness for pre-training, post-training enhancement of pre-trained VLA models, and LAM refinement, achieving an average of ~10%, ~5%, and ~15% improvement over 3 simulation and 1 meticulously designed real-world robotic manipulation benchmarks.

翻译：视觉-语言-动作（VLA）模型使机器人能够直接从观测和语言指令中预测动作，但其性能依赖于大规模高质量数据，并受限于真实世界机器人动作数据集的稀缺性。为利用丰富的无标注人类视频促进VLA模型学习，潜在动作模型（LAM）从视觉动力学中学习潜在动作表示，为VLA学习提供额外监督。然而，LAM与VLA通常分开训练，导致LAM在VLA训练过程中缺乏基础行为支撑，而VLA模型则受限于冻结的LAM表示。为解决这些问题，我们提出潜在动作表示对齐（LARA）——一种即插即用框架，通过表示对齐联合优化LAM与VLA。该框架实现双向增益：LAM借助动作轨迹学习以避免虚假视觉变化，而VLA则通过LAM学习的前向动力学进行正则化，减少对功能无效轨迹的幻觉预测。我们展示了LARA在预训练、预训练VLA模型的后训练增强以及LAM优化中的通用性与有效性，在3个仿真基准和1个精心设计的真实机器人操作基准中分别实现约10%、5%和15%的平均性能提升。