Foad Namjoo

prof_c.jpg

I’m a Computer Science Ph.D. student at the Kahlert School of Computing (University of Utah), advised by Jeff Phillips in the UtahDB Lab. I work on interpretable/controllable LLMs, ML, high-dimensional data geometry, and reliable wearable data systems.

What I build

  • Activation steering for LLMs — residual-stream injections (layer/α sweeps) to control verbosity & tone in models like Llama 3.1 and Qwen-2.5, plus guardrailed evaluation (LLM judges + JSON schema) to reduce hallucinations. I also created several structured trait datasets for conciseness and positivity as part of a research collaboration with Martian.

  • dKS: Efficient & Stable Multi-Dimensional Kolmogorov–Smirnov Test — dKS extends the Kolmogorov–Smirnov distance to higher dimensions with a unit-invariant, ε-accurate IPM that avoids the instability of mdKS and Peacock methods. The method runs in near-linear O(n log n) time in 2D and scales efficiently to 3–4 dimensions using grid-based rounding and CDF comparisons. This provides a fast, reliable two-sample test for statistics, ML, and AI applications.
    arXiv:2504.11299

  • MotionPI — an NIH-funded, privacy-first wearable sensing platform integrating a Flutter-based smartphone app with BLE wristbands and a secure Node.js + MongoDB backend. MotionPI was developed in the Kahlert School of Computing in collaboration with the Huntsman Cancer Institute, the Department of Health & Kinesiology Sciences, the College of Social & Behavioral Science, and wearable hardware and firmware teams at The Ohio State University.

    → Full MotionPI Project Page

Recent

  • Presentation: Presented our SmartSP 2025 paper (Designing a Secure and Resilient Distributed Smartphone Participant Data Collection System) on December 1, 2025 at EAI SmartSP conference.
  • Poster: Presented dKS research at the Utah AI Summit 2025 (University of Utah) — June 18, 2025.
  • Designing a Secure and Resilient Distributed Smartphone Participant Data Collection System. EAI SmartSP 2025 (accepted).
    Publications page · PDF
  • Efficient and Stable Multi-Dimensional KS Distance. arXiv 2025.
    arXiv · PDF
  • HCD-Net for Hyperspectral Change Detection. Remote Sensing 2024.
    See all