Recurrent-Depth VLA: Implicit Test-Time Compute Scaling of Vision-Language-Action Models via Latent Iterative Reasoning

作者：Jalal Naghiyev, Haoquan Fang, Wei-Chuan Tsai, Jiafei Duan, Dieter Fox, Ranjay Krishna · 单位：Stanford University, Technical University of Munich, University of Washington, Allen Institute for Artificial Intelligence · 会议/期刊：arXiv · 日期：2026-02-08 · 来源：Low-Level Learning-Based Action Modelling / Input Modelling / 2D Vision Language Action Models with Auxiliary Tasks - Text Goal Extraction