Preprint
Oct 2026
eRLT: Efficient VLA Reinforcement Learning via Action-Relevant Token Routing
Vision-Language-Action (VLA) models provide strong behavioral priors for robotic manipulation, yet efficiently adapting them to downstream tasks remains challenging. Recent work addresses this challenge by adapting frozen VLAs through online reinforcement learning (RL), whose sample efficiency depends on the quality of...
De-Hao Huang, Jian-Bang Liu, Jian-Pan Gao et al.
· 0 citations