Xiaolong Wang

Xiaolong Wang is an Associate Professor in the Department of Electrical and Computer Engineering at the University of California, San Diego. His research focuses on exploiting the structure in data for learning visual representations, particularly the spatial-temporal structure in videos and its connection to 3D and semantic structures. His work includes self-supervised learning and modeling data structure for human activity analysis, scene affordance reasoning, and object interaction, with applications in robotics. Wang joined UC San Diego as an assistant professor in 2020. He obtained his Ph.D. from The Robotics Institute at Carnegie Mellon University and has collaborated with research labs including Berkeley AI Research, Facebook AI Research, and Allen Institute for Artificial Intelligence.

Xiaolong Wang is an Associate Professor at UC San Diego, specializing in computer vision, machine learning, and robotics. His research focuses on advancing visual representation learning, robotic learning, and generalization under distribution shifts. Wang's work includes developing non-local neural networks, unsupervised visual representations from videos, and test-time training with self-supervision. He has also contributed to open x-embodiment datasets, robotic learning models, and zero-shot recognition using semantic embeddings. His research emphasizes the integration of vision and language, as seen in projects like GroupViT and SpatialRGPT. Wang's publications explore generative modeling, spatial reasoning, and control systems for robotics, with a focus on real-world applications and robustness.