Andrey Gromov
Staff AI Research Scientist at Meta Superintelligence Labs
About me
I am a Staff AI Research Scientist at Meta Superintelligence Labs. In a previous chapter, I was a professor of theoretical condensed matter physics at Brown University and the University of Maryland.
My work spans large language models, data, reasoning, efficiency and interpretability. I am more inclined to understand AI foundations using simplified toy models and phenomenology. More recently I became very interested in using AI to accelerate natural sciences.
AI
The central theme of my AI research is understanding how intelligence emerges from compute, data and sampling.
- Science of LLMs: initialization, scaling laws, post-training, evaluation, interpretability, and sampling.
- Science of data: toy models, artificial data, model collapse, and curricula.
- Science of reasoning: inference and pre-training trade-offs.
- Science of efficiency: pruning, quantization, distillation, and efficient architectures.
- AI for theoretical sciences.
Theoretical physics
The central theme of my physics research is emergence in classical and quantum systems.
- Topological phases of matter and fractional quantum Hall effect
- Fractons and topological glass
- Active matter
- Correlated systems in curved space
Vitae
Staff AI Research Scientist, Meta Superintelligence Labs, 2024-present- Senior Research Scientist, Meta FAIR, 2022-2024
Assistant Professor, University of Maryland, College Park, 2022-2024
Assistant Professor, Brown University, 2019-2022
Postdoctoral Fellow, UC Berkeley, 2018-2019
Kadanoff Fellow, University of Chicago, 2015-2018
Ph.D. in Physics, Stony Brook University, 2009-2015
B.Sc. in Physics, Saint Petersburg State University, 2005-2009
Selected papers