Skip to content

论文arXiv 2610.1394319 天前

用于 VLA 适应的噪声空间学习 ​

原文链接:Noise Space Learning for VLA Adaptation(arXiv 2610.13943 · 发布日期:2026-10-09)

Noise Space Learning for VLA Adaptation

提出 NSL 在固定预训练 VLA 下学习上下文条件噪声分布,并在 LIBERO、RoboTwin 及四项真机操作任务上验证。

核心要点 ​

  • 预训练 VLA 具备广泛操作先验,但面向特定任务适应时受机器人交互、计算资源和安全约束影响。
  • NSL 面向基于流匹配的 VLA,通过学习上下文条件噪声分布实现适应,同时保持预训练 VLA 固定。
  • 该接口可容纳不同学习算法,论文分别以 PPO、SAC 和 AWR 实例化,并给出信息论分析。
  • 实验在 LIBERO、RoboTwin 和四项真机操作任务上研究 NSL 变体,并总结不同资源与部署条件下的选择建议。

验证范围与局限:验证范围包括 LIBERO、RoboTwin 和四项真实机器人操作任务,均处于交互、计算与执行约束下。

英文摘要 (Abstract) ​

Pretrained vision-language-action (VLA) models encode broad manipulation priors, yet adapting them to specific tasks requires balancing limited robot interaction, computational resources, and safety constraints. Noise space learning (NSL) adapts flow-matching-based VLAs by learning a context-conditioned distribution over noise while keeping the pretrained VLA fixed. Building on prior noise-steering methods, we formulate NSL as an interface for VLA adaptation that accommodates different learning algorithms and instantiate it with PPO, SAC, and AWR. Theoretically, we provide an information-theoretic analysis of how NSL enables efficient adaptation. Empirically, we study these NSL algorithms on LIBERO, RoboTwin, and four real-robot manipulation tasks under interaction, computation, and execution constraints. Across these settings, we develop a suite of effective NSL variants and derive practical recommendations for choosing among them under different resource and deployment regimes.

作者:Junjie Lu, Chuheng Zhang, Yucen Wang, Yuhua Jiang, Xinyao Qin, Min Xu, Li Zhao

来源

本页事实对应的公开页面,以原文为准。

  1. arXiv 2610.13943:Noise Space Learning for VLA Adaptation访问 2026-10-21