profpic5.png

Sam Young

Physics PhD @ Stanford // youngsam [at] stanford.edu

I’m a third year PhD student in the Physics Department at Stanford University, interested in building intelligent systems that can understand and reason about data from particle physics experiments. I am advised by Kazu Terao in the neutrino group at SLAC National Accelerator Laboratory.

I’m broadly interested in AI for science, foundation models, 2D/3D computer vision, and inverse problems.

For my PhD, I’ve been doing:

  • Representation learning and foundation models for scientific imaging data
  • 3D computer vision for particle physics
  • Differentiable surrogate models for calibrating simulation

Previously, I worked with Julia Gonski of the SLAC ATLAS group on applying machine learning techniques for anomaly detection in the ATLAS experiment at CERN, and with Giorgio Gratta of the Stanford neutrino group on detector R&D for the nEXO experiment. Before that, I started my research journey at the University of Pennsylvania working with Josh Klein on novel neutrino detector instrumentation for THEIA, a proposed hybrid optical neutrino detector. I graduated from Penn in 2023 with a bachelor’s and master’s in physics.

News!

2025 Gave a talk on recent work on building a foundation model for neutrino physics at the NPML 2025. Slides
2025 Gave a talk on foundation models for neutrino physics at the Machine Learning for Fundamental Physics (ML4FP) summer school. Slides
2025 Gave a talk on on foundation models for scientific imaging (LArTPC data) at the APS Global Summit.
2024 I was awarded the HAI Graduate Fellowship!
2024 Gave a talk on differentiable surrogate modeling of optical propagation in LArTPCs at APS April.
2024 Won Stanford’s CS 229 Machine Learning’s Best Project Award for my rotation work on pileup synthesis and anomaly detection for the ATLAS experiment.

Papers

2025

  1. Panda: Self-distillation of Reusable Sensor-level Representations for High Energy Physics
    tl;dr: We applied DINO-like self-distillation to 3D images of particle interactions.
    Samuel Young, and Kazuhiro Terao
    arXiv, 2025
  2. 10 Million Particle Events: Enabling Foundation Models for Sparse 3D Inverse Problems
    tl;dr: A proposal for a new dataset containing 10 million images of particle interactions with realistic detector response and multiple modalities.
    Omar Alterkait, Sam Young, Ka Vang Tsang, Junjie Xia, Carolyn H Smith, Taritree Wongjirad, and Kazuhiro Terao
    NeurIPS 2025 AI4Science Workshop (Spotlight), 2025
  3. Particle Trajectory Representation Learning with Masked Point Modeling
    tl;dr: We proposed a masked point modeling approach for learning representations of particle trajectories from LArTPC images.
    Sam Young, Yeon-jae Jwa, and Kazuhiro Terao
    arXiv, 2025