I am a fifth-year physics PhD student at Cornell University, advised by Professor Chris Earls. Broadly speaking, I am interested in using scientific methods to explain, improve, and safeguard the thinking process of AI, and in applying insights from AI to better model physical systems.
Prior to Cornell, I received my B.S. in applied physics from Caltech, and my B.A. in physics and film studies from Wesleyan University.
A curated selection, not a full publication list.
When an LLM answers, which prompt tokens drove that decision? Jacobian Scopes attribute predictions—logits, full distributions, and uncertainty—to inputs via fast vector-Jacobian products, with strong performance on attribution benchmarks.
How do LLMs learn from in-context data? Using Intensive Principal Component Analysis, we visualize their density-estimation trajectories and find an inference process akin to kernel regression, with a kernel whose width and shape adapt as evidence accumulates.
What separates abstract thoughts from concrete memories? Treating memories as fixed points, we cast abstraction as a hierarchy of slow manifolds and build a diffusion RNN that extracts them from data—with an analytic theory linking data distributions to the learned manifolds.
Can a pretrained LLM make sense of an unseen synthetic language—say, symbols from a random Markov process? LLaMA 2 models in-context learn a range of stochastic systems, physical and symbolic, revealing an in-context neural scaling law.
Hyperbolic volume grows exponentially with radius—like the branching of a tree—making it a natural host for hierarchical data. Shadow Cones embed partial orders by casting nested shadows: a child lies in its parent’s cone, yielding a fast algorithm for embedding concept graphs.
