新模型仅需一次示范就能快速学会新技能，适配新环境和机器人。
MoS-VLA: A Vision-Language-Action Model with One-Shot Skill Adaptation
- 将操控策略拆解为一组基础技能的线性组合，构建结构化技能空间。
- 单次示范后通过轻量优化求解技能表示，无需梯度更新，速度快。
- 在5个未见数据集上表现优于基线，实机与仿真任务均成功部署。
视觉-语言-动作（VLA）模型在大规模机器人数据集上训练，有望实现跨领域、跨机器人的通用鲁棒控制。然而，现有方法在新环境、新机器人或新任务中常无法直接使用。我们提出混合技能视觉-语言-动作模型（MoS-VLA），将机器人操作策略表示为一组有限学习基函数的线性组合。预训练阶段，MoS-VLA联合学习来自Open X-Embodiment项目多个数据集的基函数，形成结构化技能空间。测试时，仅需一个专家示范即可完成新任务适应。通过求解轻量凸优化问题，最小化L1动作误差，推断对应技能表示，无需梯度更新。该免梯度适应方法开销极小，可快速生成新技能。实验表明，MoS-VLA在五个未见数据集上均实现更低的动作预测误差，在模拟与真实机器人任务中均取得成功，而预训练模型则完全失败。
原文摘要 · Abstract (English)
Vision-Language-Action (VLA) models trained on large robot datasets promise general-purpose, robust control across diverse domains and embodiments. However, existing approaches often fail out-of-the-box when deployed in novel environments, embodiments, or tasks. We introduce Mixture of Skills VLA (MoS-VLA), a framework that represents robot manipulation policies as linear combinations of a finite set of learned basis functions. During pretraining, MoS-VLA jointly learns these basis functions across datasets from the Open X-Embodiment project, producing a structured skill space. At test time, adapting to a new task requires only a single expert demonstration. The corresponding skill representation is then inferred via a lightweight convex optimization problem that minimizes the L1 action error, without requiring gradient updates. This gradient-free adaptation incurs minimal overhead while enabling rapid instantiation of new skills. Empirically, MoS-VLA achieves lower action-prediction error on five out of five unseen datasets and succeeds in both simulation and real-robot tasks where a pretrained VLA model fails outright. Project page: mos-vla.github.io/
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。