外观
论文arXiv 2610.0599423 天前
如何（以及如何不）在 VLA 后训练中使用数据增强
原文链接：How (and How Not) to Use Data Augmentation in VLA Post-Training（arXiv 2610.05994 · 发布日期：2026-10-05）
How (and How Not) to Use Data Augmentation in VLA Post-Training
系统研究图像增强用于VLA强化学习后训练，验证其在LIBERO-Plus分布外泛化上的效果。
核心要点
- VLA在真实机器人任务中表现强，但面对大幅视觉分布外变化仍泛化不足。
- 研究图像增强对强化学习后训练的影响，提出仅增强critic、actor保持干净输入。
- 在π_{0.5}和GR00T N1.5上，LIBERO-Plus分布外成功率分别提升7.8和10.0点。
- 比较多种增强类型与强度，并给出改进VLA后训练泛化的实践建议。
验证范围与局限：验证范围为π_{0.5}和GR00T N1.5在LIBERO-Plus上的分布外成功率。
英文摘要 (Abstract)
Vision-language-action (VLA) models currently demonstrate strong performance in a wide range of real-world robotics tasks. However, they often still lack the generalization ability to handle large visual out-of-distribution shifts. Post-training of VLAs with reinforcement learning (RL) has been shown to benefit robustness, but significant room for improvement remains. In this work, we systematically study the effect of image augmentation on VLA post-training. We find that it is crucial to augment only the critic module during RL updates, while leaving the actor's input clean during both rollouts and updates. For $π_{0.5}$ and GR00T N1.5 this raises out-of-distribution success on LIBERO-Plus by $7.8$ and $10.0$ points respectively, while augmenting the actor collapses training entirely. We investigate a range of augmentation types and strengths, and provide practical recommendations for improving generalization in VLA post-training.
作者：Bram Grooten, Joaquin Vanschoren
来源
本页事实对应的公开页面，以原文为准。