Hello! I'm a software engineer at Epic on the Curiosity team, building generative foundation models on medical event sequences for clinical outcome prediction. I did my undergrad in CS and math at UMass Amherst, along with a combined masters in CS as a Bay State Fellow. I am fortunate to have been advised throughout by Professor Jeff M. Phillips of University of Utah. I'm broadly interested in understanding what modern ML models learn and why they generalize. I've worked on LLM interpretability, robust statistics, and the evaluation of language models.

[ github] [ linkedin] [ scholar] [ cullenyanderson@gmail.com]

Publications

• From Token Probabilities to Semantic Constraints: Towards Declarative Probabilistic Evaluation of Language Models
  Kyle Richardson, Cullen Anderson, Pranav Balakrishnan, Takuto Ban, Daksha Ladia, Ankita Gupta, Marisa Hudspeth
  EMNLP Main 2026 🏆 Outstanding Paper Award
• Robust High-Dimensional Mean Estimation With Low Data Size, an Empirical Study
  Cullen Anderson, Jeff M. Phillips
  TMLR 2025

Preprints

• Understanding and Mitigating Dataset Corruption in LLM Steering
  Cullen Anderson, Narmeen Oozeer, Foad Namjoo, Remy Ogasawara, Amirali Abdullah, Jeff M. Phillips
  Under submission
• Orthogonal and Parametrized Multi-Attribute Steering of LLMs
  Remy Ogasawara, Cullen Anderson, Narmeen Oozeer, Foad Namjoo, Amirali Abdullah, Jens Kristian Refsgaard Schou, Jeff M. Phillips
  Under submission
• Judging by the Cover: Cleaning LLM Truthfulness Benchmarks to Avoid Surface-Level Feature Leakage
  Foad Namjoo, Remy Ogasawara, Amirali Abdullah, Cullen Anderson, Narmeen Oozeer, Jeff M. Phillips
  Under submission