Author name: Rui Qian, Xin Yin, Chuanhang Deng, Zhiyuan Peng, Jian Xiong, Wei Zhai, Dejing Dou

arXiv:2510.03853v4 Announce Type: replace
Abstract: We present UGround, a \textbf{U}nified visual \textbf{Ground}ing paradigm that dynamically selects intermediate layers across \textbf{U}nrolled transformers as “mask as prompt,” diverging from the …