A new arXiv paper shows that for vision-language-action models like π0.5 and GR00T N1.5, applying image augmentation only to the critic during reinforcement learning post-training boosts out-of-distribution success…

#Robotics #VLA #ReinforcementLearning #Robustness
https://arxiv.org/abs/2610.05994

#xAILabBamberg #BaCAI #HightechAgendaBayern #MedicalImageAnalysis #DomainGeneralization #Robustness #Reliability
@uni-bamberg.de @miccaisociety.bsky.social

Enhance the Safety in Reinforcement Learning by ADRC Lagrangian Methods

Mingxu Zhang, Huicheng Zhang, Jiaming Ji, Yaodong Yang, Ying Sun

Action editor: Steffen Udluft

https://openreview.net/forum?id=3IbuT8uzYS

#lagrangian #reinforcement #robustness