Peter Hase

Ph.D. Student
University of North Carolina
Connect with Peter

Peter is a fifth-year Ph.D. student in the UNC-NLP lab at the University of North Carolina at Chapel Hill, where he is advised by Mohit Bansal. His work at UNC is supported by a Google PhD Fellowship and previously by a Royster Fellowship. Before this, he graduated with a bachelor’s degree from Duke University, where his thesis advisor was Cynthia Rudin. At Duke, he was supported by a Trinity Scholarship.

Peter’s research interests center on interpretable machine learning and natural language processing. Below are some of the main problems he worked on:

  • Mechanistic Interpretability
  • Natural Language Explanations
  • Model Editing
  • Scalable Oversight
  • Supervised and Decomposable Reasoning
  • XAI Methods & Evaluation

Broadly, he is interested in explaining model behavior and improving model safety. Peter sees language models as a good object of study since we lack complete explanations for their behavior and human language provides a rich means of interaction with models. He finds work on clarifying concepts and developing strong evaluation procedures especially valuable.

Podcast Episodes