Preprint
Sep 2026
UniMPA: A Unified Memory-Prediction-Action Model via Action-Grounded Transition Modeling
Recent advances in Vision-Language-Action (VLA) models have improved robotic manipulation, yet observation-to-action learning remains limited by a fundamental transition realizability gap, manifested in three tightly coupled problems: (i) Transition ambiguity. Visually similar current observations may correspond to dif...
Wei Li, Rui Shao, Jie He et al.
· 0 citations