Pieter Abbeel

Cross-Hand Latent Representation for Vision-Language-Action Models

跨手的潜在表示用于视觉-语言-动作模型

Guangqi Jiang, Yutong Liang, Jianglong Ye, Jia-Yang Huang, Changwei Jing, Rocky Duan, Pieter Abbeel, Xiaolong Wang, Xueyan Zou

AI总结 XL-VLA通过统一的潜在动作空间实现跨手灵巧操作的可扩展学习，提升视觉-语言-动作模型的训练效率和性能。

Comments Website: https://xl-vla.github.io