Shailen K. Smith

Hi! I’m a second-year Computer Science PhD Student at Dartmouth College advised by Professor Adam Breuer. I am interested broadly in AI privacy, robustness, and interpretability. I completed my undergrad in May 2024 from Stony Brook University in Mathematics.

With the Breuer lab, my current research explores the effect of adversarial robustness on model inversion attacks (MIAs) for deep facial classification models. Our recent paper, “Reducing information dependency does not cause training data privacy. Adversarially non-robust features do”, was recently accepted by ICLR 2026 (I will be attending, come say hi!).

At Stony Brook, I researched human-aware autoencoding LLMs under Professor H. Andrew Schwartz, with mentorship from Adithya V. Ganesan and Nikita Soni.

In summer 2023, under Professor Neil Heffernan and PhD student Morgan Lee at Worcester Polytechnic Institute (WPI), I trained a Bayesian Knowledge Tracing model to predict student mastery of class content for use in the ASSISTments educational technology platform. In summer 2022, I researched in algorithmic fairness at WPI under Professor Elke Rundensteiner and PhD student Kathleen Cachel.

Outside of research, I have a passion for math education and environmentalism. During my junior and senior years at Stony Brook, I was Lead Tutor at the Academic Success and Tutoring Center, and Treasurer for the Stony Brook Environmental Club.

In my free time, I enjoy chess, Scrabble, tennis, piano, funk music, and rock climbing.