I am the Co-founder of Assured Robot Intelligence (ARI). Our mission is to build super-intelligent humanoid robots at scale to deliver abundant physical labor to humanity. If you are interested in joining our mission, please send us an email.
Before ARI, I was a researcher at NVIDIA, where I built foundation models. Here is a list of my work at NVIDIA.
I am also an associate professor at UC San Diego. I was a postdoctoral fellow at UC Berkeley with Alexei Efros and Trevor Darrell. I received a Ph.D. in robotics from the Carnegie Mellon University, at where I worked with Abhinav Gupta.
Research
I am interested in representation learning through videos and physical robotic interaction data. We explore various means of supervision signals from the data itself, language, and common sense knowledge. We leverage these comprehensive representations to facilitate the learning of robot skills, with the goal of generalizing the robot to interact effectively with a wide range of objects and environments in the real physical world.
News
I gave a talk in the UW Robotics Colloquium on Modeling Humans for Humanoid Robots (2025, March).
I gave a Keynote talk in ICPR 2024 on From Perception to Embodied AI: Modeling Humans for Humanoid Robots (2024, Dec).
I gave a talk in the Stanford Robotics Seminar on Modeling Humans for Humanoid Robots (2024, Nov).
I gave a talk in the Whole-body Control Workshop in CoRL 2024 on Generalizing Humanoid Robots (2024, Nov).
I gave a talk in the Dexterous Manipulation Workshop in CoRL 2024 on Dexterous Manipulation with Real and Sim Data (2024, Nov).
I gave talks in the seminars at TTIC, Cornell Tech, NYU, Umass Amherst on Learning Humanoid Robots (2024, Aug-Oct).
I gave a talk in the Tactile Sensing Workshop in RSS 2024 on Learning In-hand Object Rotation (2024, July).
I gave a talk in the Implicit Neural Representation for Vision Workshop in CVPR 2024 on Feature Fields for Manipulation (2024, June).
I gave a talk in the Multimodalities for 3D Scenes Workshop in CVPR 2024 on Generalizable 3D Spatial Perception and Control (2024, June).
I gave talks in the Transformers for Vision (T4V), and Prompting in Vision Workshops in CVPR 2024 on Learning to (Learn at Test Time) (2024, June).
I gave talks in the Computer Vision in the Wild, and Visual Perception via Learning in an Open World Workshops in CVPR 2024 on Spatial Perception and Control in the Wild (2024, June).
I gave talks in the RoboNerF: 1st Workshop On Neural Fields In Robotics Workshop in ICRA 2024 on Learning Generalizable Feature Fields for Mobile Manipulation (2024, May).
I gave talks in the Agile Robotics: From Perception to Dynamic Action Workshop in ICRA 2024 on Learning Humanoid Robots (2024, May).
I gave talks in the 5NRP Workshop, the Intel's Rising Star Tech talk, and USC Robotics Seminars on Learning Humanoid Robots (2024, March).
I gave a talk in the TTI/Vanguard: [next] Workshop on Human-Centric Robot Learning (2023, Dec).
I gave a talk in the UPenn GRASP seminar on Generalizable Geometric Robot Learning (2023, Nov).
I gave a talk in the UC San Diego Contextual Robotics Institute's forum (2023, Nov).
I gave a talk in the CoRL 2023 Deployable Workshop on Generalizable Geometric Robot Learning (2023, Oct).
I gave a talk in the NSF SAIL Workshop on Human-Centric Robot Learning (2023, Oct).
Our RSS paper Rotating without Seeing is covered in Scientific American.
I gave a talk in the Workshop on 3D Scene Understanding for Vision, Graphics, and Robotics in CVPR 2023 on 3D Scene Understanding for Locomotion Control.
I gave a talk in the Workshop on 3D Vision and Robotics in CVPR 2023 on Geometric Robot Learning for Generalizable Dexterous Manipulation.
I gave a talk in the MIT Embodied Intelligence Seminar, Stanford SVL Seminar, and UC Berkeley on Geometric Robot Learning for Generalizable Skills Acquisition.
I gave a talk in the Workshop on Neural Fields across Fields in ICLR 2023, on Generalization in Neural Radiance Fields. Also see our recent release on Generalizable NeRFs: TUVF, ActorsNeRF, FeatureNeRF, MonoNeRF.
I am co-organizing the Workshop on Learning Dexterous Manipulation in RSS 2023.
I am co-organizing the Workshop on 4D Hand Object Interaction in CVPR 2023.
I gave a talk on Learning Rich and Generalizable Representation for Visual Motor Control at Nuro, Covariant, UCSD Visual Computing Retreat, Brown Visual Computing Seminar. Here is the recorded video.
I am co-organizing the Tutorial on Building and Working in Environments for Embodied AI in CVPR 2022. Here is the code base for the tutorial.
I gave a talk in the Workshop on Generalizable Policy Learning in the Physical World in ICLR 2022, on Generalizing Dexterous Manipulation by Learning from Humans.
I gave a talk in the Tutorial on Large Scale Holistic Video Understanding in ICCV 2021, on Learning to Perceive Videos for Embodiment.
I gave a talk in the Large-scale Video Object Segmentation Challenge Workshop in CVPR 2021, on Self-Supervised Representation Learning with Videos.
I am co-organizing the 3rd Tutorial on Learning Representations via Graph-structured Networks in CVPR 2021. Here is the recorded video.
I am co-organizing the Comprehensive Tutorial on Video Modeling in CVPR 2021.
I gave a talk in Nvidia on Self-Supervised Learning. Here is the recorded vidre is the recorded video.
I am co-organizing the Workshop on Sensing, Understanding and Synthesizing Humaeo.
I gave a talk in the 4D Vision Workshop in ECCV 2020. Hens in ECCV 2020.
I am co-organizing the 2nd Tutorial on Learning Representations via Graph-structured Networks in CVPR 2020. Here is the recorded video.
I am co-organizing the Tutorial on Learning Representations via Graph-structured Networks in CVPR 2019.
I am co-organizing the Workshop on Multi-Modal Learning from Videos in CVPR 2019.
I gave a talk in the Tutorial on GANs in CVPR 2018, here is the recorded video.