Angel X Chang

Associate Professor@Simon Fraser University·加拿大·OpenReview

研究方向

text to scene generation · text to shape generation · 3d scene understanding · multimodal reasoning · grounded language · embodied ai · referring expressions

7.0

Do 3D Large Language Models Really Understand 3D Spatial Relationships?

ICLR 2026Poster

4.7

MLFM: Multi-Layered Feature Maps for Richer Language Understanding in Zero-Shot Semantic Navigation

合作者 (20)

Angel X Chang

Do 3D Large Language Models Really Understand 3D Spatial Relationships?

MLFM: Multi-Layered Feature Maps for Richer Language Understanding in Zero-Shot Semantic Navigation

VISOR: VIsual Spatial Object Reasoning for Language-driven Object Navigation

SINGAPO: Single Image Controlled Generation of Articulated Parts in Objects

Duoduo CLIP: Efficient 3D Understanding with Multi-View Images

CLIBD: Bridging Vision and Genomics for Biodiversity Monitoring at Scale