Published multiple papers on offline, meta, and open-ended RL, diffusion models, and plasticity. Some of the papers include 'A Clean Slate for Offline Reinforcement Learning', 'Token-Sparse Diffusion Transformers', and more, presented at top conferences such as NeurIPS, ICLR, etc.
Research Experience
Currently a student researcher at Google DeepMind, working on Genie. Previously, completed a research internship at Wayve, working in the World Models Team on video generation for autonomous vehicles, as well as SWE internships at Amazon, Arm, and Cubica.
Education
DPhil student at the University of Oxford, supervised by Jakob Foerster and Shimon Whiteson.
Background
Fourth-year DPhil student at the University of Oxford, with research interests in how VLMs can improve RL and how RL can improve VLMs.