About
I’m a Ph.D. candidate in the CILVR lab at NYU Courant, co-advised by Rob Fergus and Lerrel Pinto. My research is supported by a DeepMind Ph.D. Scholarship and an NSF Graduate Research Fellowship.
I’m interested in reasoning, decision-making, and open-ended interaction with large generative models. Recently, I’ve been thinking about:
-
How do we go about building agents that can flexibly explore and interact with the world the way we (humans and animals) do?
-
When is next-token prediction a sufficient pretraining objective for text-based decision-making?
-
What sort of structures, priors, and training/finetuning/prompting paradigms can enable a large pre-trained model to be an elegant proof solver? To act as an effective surrogate for ODEs/PDEs/non-linear dynamics? To serve as a (truly) universal high-level controller in robotics settings?
My work touches on generative modeling and reinforcement learning across modalities (vision, natural language processing, simulators, etc.). I’m also broadly interested in differentiable computing.
I’ve spent time at EPFL LCN, the Applied Science Team at Google Research, and Microsoft Research NYC. I did my undergrad in mathematics at MIT, during which I was exceptionally lucky to be mentored by Kelsey R. Allen, Gigliola Staffilani, Jörn Dunkel, and Raffaele Ferrari.