Patrick Wierzbicki

profile_pic.jpeg

Hello! I am a research intern at the Explainable and Controllable AI Lab, working with Dr. Zining Zhu. Prior to joining the Explainable and Controllable AI Lab, I received my Bachelor’s Degree in Computer Science at Ramapo College of New Jersey.

My research broadly lies in Artificial Intelligence, Information Retrieval, and Natural Language Processing. I aim to develop Language Models(LMs) that can utilize vast swathes of information to assist humans in their day-to-day lives.

Towards this end, my current research interests include:

  • Retrieval-augmented Generation
  • Semiparametric and Nonparametric Retrieval-based LMs
  • Long-Context Language Modeling
  • Self-Correction in LMs

Preprints

  1. Interpolation versus Self-Recovery: Does the kNN-LM Retriever Introduce Exposure Bias?
    Patrick Wierzbicki, and Zining Zhu
    Submitted to NAACL 2025, Dec 2024