Experience

Research, independent study, internships, leadership, and project work from my CV.

Research experience

Research project · Prof. Kangwook Lee’s lab · UW–Madison

October 2024 – January 2025 · Faculty mentor: Kangwook Lee (ECE)

  • Created synthetic chain-of-thought reasoning data to train a Process Reward Model (PRM).
  • Implemented auto-labelling to produce ground-truth labels for fine-tuning.

Independent study · Speech transcription · UIUC

August 2023 – January 2024 · Faculty mentor: Lawrence Angrave (CS)

  • Analyzed Whisper on STEM lecture audio.
  • Evaluated fine-tuning on a specific speaker’s voice with limited training data.

NSF REU internship · Penn State University

May 2023 – August 2023 · Faculty mentor: Dongwon Lee (College of IST)

  • Led a 10-week machine learning project with a professor and PhD student mentor.
  • Studied plagiarism from training data in large language model outputs.
  • Found plagiarism in up to ~15% of samples from GPT-NeoX; reduced it to roughly 2–5% via in-context learning and decoding adjustments.

Independent study · Collaborative editing · UIUC

December 2022 – May 2023 · Faculty mentor: Lawrence Angrave (CS)

  • Surveyed real-time collaborative editing: Operational Transformation (OT) and CRDTs.
  • Recommended ShareDB (OT) for new features in ClassTranscribe and helped coordinate work with Project Code.

Publications

  • VersaPRM: Multi-Domain Process Reward Model via Synthetic Reasoning Data — T. Zeng, S. Zhang, S. Wu, C. Classen, D. Chae, E. Ewer, M. Lee, H. Kim, W. Kang, J. Kunde, Y. Fan, J. Kim, H. Koo, K. Ramchandran, D. Papailiopoulos, K. Lee. ICML, 2025.
  • In-batch Ensemble Drafting: Robust Speculative Decoding for LVLMs — M. Lee, W. Kang, B. Ahn, C. Classen, M. Yan, H. Koo, K. Lee. ICLR 2025 SCOPE Workshop.
  • Balancing Reality with Originality: How Prompting Affects Plagiarism in Language Models — C. Classen, J. Lee, D. Lee.

Presentations

  • “An Investigation of Collaborative Editing Implementations” — poster, UIUC Undergraduate Research Symposium, Urbana, IL, 2023.

Leadership & extracurriculars

Wisconsin AI Safety Initiative (WAISI) · Exec board member

September 2025 – present

  • Lead weekly meetings on current AI safety research.
  • Help design curriculum for an intro AI safety program for undergraduates.

Wisconsin AI Safety Initiative (WAISI) · Tech scholar

September 2024 – present

  • Participate in weekly discussions on AI safety research topics.

Project Code · Vice president

June 2023 – May 2024

  • Interviewed project manager candidates and ran training.
  • Mentored project managers so teams stayed on track.
  • Grew projects per semester with Professor Lawrence Angrave.
  • Helped organize career events (e.g. resume reviews, recruiter talks).

Project Code · Project manager

August 2022 – May 2023

  • Led two semester-long ML projects (12–15 students each, many new to ML).
  • Produced videos, slides, and demos on core concepts.
  • Guided members through training their own DCGAN models in PyTorch.