MVP-LAM: Learning Action-Centric Latent Action via Cross-Viewpoint Reconstruction

作者：Jung Min Lee, Dohyeok Lee, Seokhun Ju, Taehyun Cho, Jin Woo Koo, Li Zhao, Sangwoo Hong, Jungwoo Lee · 单位：Seoul National University, Seoul, South Korea Microsoft Research Asia, Beijing, China, Konkuk University, Seoul · 会议/期刊：arXiv · 日期：2026-02-03 · 来源：Low-Level Learning-Based Action Modelling / Input Modelling / 2D Vision Language Action Models with Latent Learning