Henry Smith

Henry-Smith-headshot.jpeg

Sequoia Hall

Palo Alto, CA 94305

My name is Henry (he/him), and I am a fourth-year PhD student in the Department of Statistics at Stanford University, where I am fortunate to be advised by Professors Scott Linderman and Brian Trippe. My graduate studies are graciously supported by the Knight-Hennessy fellowship and the National Science Foundation Graduate Research Fellowship (GRFP).

I am broadly interested in applied probability and probabilistic machine learning (probabilistic ML), with applications to neuroscience and computational protein design. I am especially excited about continuous-time latent variable models, such as diffusion and flow models.

Find my CV here. You can contact me at [last-name]hd@stanford.edu.

News

Jul 6, 2026 My paper “Calibrating Generative Models to Distributional Constraints” with Nathaniel Diamant and Brian Trippe was accepted at ICML.
Jun 15, 2026 A preprint of my paper “Closing the Approximation Gap in Simulation-free Latent SDEs” with Brian Trippe and Scott Linderman is available on arXiv.
Jun 1, 2026 I will be interning over the summer at the Flatiron Institute, where I will be advised by Minhuan Li, Mark Goldstein, and Pilar Cossio.
Oct 11, 2025 A preprint of my paper “Calibrating Generative Models” with Nathaniel Diamant and Brian Trippe is available on arXiv. Check out our code as well.
Sep 17, 2025 My paper “SING: SDE Inference with Natural Gradients” with Amber Hu and Scott Linderman was accepted at NeurIPS.