Improving Robustness of Vision-Language-Action Models by Restoring Corrupted Visual Inputs

作者：Daniel Yezid Guarnizo Orjuela, Leonardo Scappatura, Veronica Di Gennaro Riccardo Andrea Izzo, Gianluca Bardaro, Matteo Matteucci · 单位：Department of Electronics, Information, and Bioengineering · 会议/期刊：arXiv · 日期：2026-02-01 · 来源：Low-Level Learning-Based Action Modelling / Input Modelling / 2D Vision Language Action Models with Robustness / Attack