Oct 2025: Released WorldGym, a world model for evaluating robot policies (e.g., OpenVLA, Octo)
Sep 2025: Organizing a workshop on world models in robotics at CoRL 2025
Aug 2025: Paper 'VideoAgent: Self-Improving Video Generation' selected as spotlight at RL Conference 2025 Workshop on Reinforcement Learning Beyond Rewards
Jul 2025: Presented on evaluating robot policies in a world model at ICML Workshop on Building Physically Plausible World Models
May 2024: Paper 'UniSim: Learning Interactive Real-World Simulators' received Outstanding Paper Award at ICLR 2024
Background
Assistant Professor of Computer Science at NYU Courant
Staff Research Scientist at Google DeepMind
Researches in machine learning with a focus on reinforcement learning and generative modeling
Recently interested in problems at the intersection of foundation models and decision making, such as learning world models and agents