Max Hartman

prof_pic.jpg

1308 West Main St.

Urbana, IL 61801

Hi, I’m Max Hartman, an MS student at the UIUC ECE department, where I am fortunate to be advised by Prof. Minh N. Do. Previously, I completed my BS at UIUC in Computer Engineering, advised by Prof. Lav R. Varshney. In Summer 2026, I will be an Applied Research Scientist intern at Meta, working on LLMs for Comment Ranking. Previously, I was a software development engineer intern at Amazon (Summer 2025).

My research interests are centered on AI Safety, Efficiency, and Privacy in Large Language Models and Vision Language Models. My primary goal is to move the field beyond empirical heuristics by establishing theoretically grounded frameworks that explain the underlying mechanisms of model behavior. I enjoy conducting experimental work to motivate algorithmic improvements justified by theoretical results. I often enjoy using tools from Information and Control Theory.

news

Jan 22, 2026 My paper, Federated Nonlinear System Identification, was accepted to The 2026 American Control Conference in New Orleans!
Aug 27, 2025 My paper, Unmasking the functionality of early layers in VLMs, was accepted to the eXCV Workshop at ICCV 2025!
Aug 16, 2025 I started my MS at UIUC! :smile:
Aug 15, 2025 I completed my summer internship at AWS in the AWS LoggingCore Observability department!
May 15, 2025 I graduated from UIUC in Computer Engineering and a math minor with highest honors!

latest posts

selected publications

  1. Skip-It? Theoretical Conditions for Layer Skipping in Vision-Language Models
    Max Hartman, Vidhata Jayaraman, Moulik Choraria, and 2 more authors
    2025
  2. Unmasking the functionality of early layers in VLMs
    Max Hartman, Vidhata Jayaraman, Moulik Choraria, and 2 more authors
    In ICCV 2025 Workshop on Explainable Computer Vision: Quo Vadis? (Non-Proceedings Tracks), 2025