Xiaohan Mao

language grounding (2) multi-modal learning (2) question answering (1) 3d scene understanding (1) visual grounding (1) embodied ai (1) group theory (1) deep learning (1) symmetry learning (1) 3d perception (1) causal intervention (1) transformation learning (1) semantic occupancy (1) compositional zero-shot learning (1) attribute-object composition (1) language annotation (1) 3d scene dataset (1) embodied perception (1) 3d visual grounding (1) large language model (1)