ProxiDex: Learning Dynamics-Guided Proximity Policy for Dexterous Manipulation
ProxiDex is presented, a dynamics-guided proximity policy framework that treats hand-object proximity as an interaction state for dexterous manipulation, and adaptively reweights proximity tokens across manipulation phases and uses dynamics-consistency supervision to guide policy inference, stabilizing action generatio...
Yu-Shan Bai, Bo-Yu Zheng, Zhi-Yang Mao et al.
· 0 citations