Max Torop

headshot.png

I am a sixth year PhD candidate in Prof. Jennifer Dy’s Machine Learning Lab at Northeastern University. I work on representation engineering for LLMs and interpretable ML. I also collaborate with scientists at MSKCC on ML for dermatology. I was a research intern at Apple during Summer 2024, working on methods for selecting SFT/RLHF training data for LLMs.

Before coming to Northeastern I completed a Masters in CS at WUSTL, where I developed deep learning methods for MRI data as a member of Prof. Ulugbek Kamilov’s Computational Imaging Group. I recieved a BS in Data Science from the University of Rochester in 2018.

Outside of research I love to watch anime and read. Some favorites are One Piece, Fullmetal Alchemist, Dororo, One Hundred Years of Solitude, The Stormlight Archive, Never Let Me Go, East of Eden and Hyperion.

News

Sep 18, 2025 Our paper DISCO: Disentangled Communication Steering for Large Language Models has been accepted to NeurIPS 2025!
Aug 18, 2025 I will be attending the National Cancer Institute (NCI) sponsored ITCR meeting in Rockville, MD.
Jan 21, 2025 Our work Axiomatic Explainer Globalness via Optimal Transport has been accepted to AISTATS 2025.
Oct 24, 2024 Gave a talk on our work SmoothHess to Prof. Doshi-Velez’s DtAK lab at Harvard.
Apr 22, 2024 I will be joining Apple as an ML Research Intern this Summer!

Selected Publications

2025

  1. DISCO.png
    DISCO: Disentangled Communication Steering for Large Language Models
    Max Torop, Aria Masoomi, Masih Eskandar, and 1 more author
    In The Thirty-ninth Annual Conference on Neural Information Processing Systems, 2025
  2. WGlob.png
    Axiomatic Explainer Globalness via Optimal Transport
    Davin Hill, Joshua Bone, Aria Masoomi, and 2 more authors
    In International Conference on Artificial Intelligence and Statistics, 2025

2024

  1. GPEC_Preview.png
    Boundary-Aware Uncertainty for Feature Attribution Explainers
    Davin Hill, Aria Masoomi, Max Torop, and 2 more authors
    In Proceedings of The 27th International Conference on Artificial Intelligence and Statistics, 02–04 may 2024

2023

  1. SmoothHess_Website.png
    SmoothHess: ReLU Network Feature Interactions via Stein’s Lemma
    Max Torop, Aria Masoomi, Davin Hill, and 3 more authors
    In Advances in Neural Information Processing Systems, 02–04 may 2023

2020

  1. RoAR_Preview.png
    Deep learning using a biophysical model for robust and accelerated reconstruction of quantitative, artifact-free and denoised images
    Max Torop, Satya VVN Kothapalli, Yu Sun, and 4 more authors
    Magnetic resonance in medicine, 02–04 may 2020