Research Scientist | Personalization, Alignment, and Large-Scale Learning

Jeong Min Lee Hello, I am Jeong Min Lee, a Research Scientist at Meta working on large-scale personalization and ranking systems. My work has focused on learning robust objectives from noisy implicit feedback, aligning optimization targets with longer-horizon user value, and improving decision systems under bias and uncertainty at internet scale.

I am currently especially interested in post-training for reasoning and agentic systems, including reward modeling, evaluation, memory, and iterative self-improvement.

Previously, I received my Ph.D. in Computer Science from the University of Pittsburgh, advised by Dr. Milos Hauskrecht. Across both academic and industry settings, a recurring theme in my work has been building adaptive learning systems from noisy, heterogeneous signals.

Current interests: post-training for reasoning and agents; personalization and memory; reward/preference modeling from noisy feedback; alignment to long-horizon value; scalable evaluation of learning systems.

Updates

Selected Work

  • MBD: A Model-Based Debiasing Framework Across User, Content, and Model Dimensions
    arXiv 2026
    A model-based debiasing framework across multiple sources of bias in large-scale recommendation systems.
    (PDF)
  • Personalized Event Prediction for Electronic Health Records
    Artificial Intelligence in Medicine (Journal) 2023
    Personalized adaptation methods for clinical event prediction from complex healthcare time-series.
    (PDF) (ScienceDirect)
  • Learning to Adapt Clinical Sequences with Residual Mixture of Experts
    AIME 2022
    Residual mixture-of-experts methods for adaptive sequence learning under heterogeneity.
    (PDF) (Poster) (Code)
  • Transformers4Rec: Bridging the Gap between NLP and Sequential / Session-Based Recommendation
    RecSys 2021
    Transformer-based modeling for sequential and session-based recommendation.
    (PDF) (Code)

Publications

  • Personalized Event Prediction for Electronic Health Records
    Jeong Min Lee and Milos Hauskrecht
    Artificial Intelligence in Medicine (Journal)
    (Paper) (ScienceDirect)
  • Neural Event Prediction for Clinical Event Time-Series
    Jeong Min Lee
    Ph.D. Thesis, University of Pittsburgh 2022
    (Thesis) (Slides)
  • Learning to Adapt Clinical Sequences with Residual Mixture of Experts
    Jeong Min Lee and Milos Hauskrecht
    20th International Conference on Artificial Intelligence in Medicine (AIME) 2022
    (Paper) (Poster) (Code)
  • Effectiveness of Transformers on Session-Based Recommendation
    Sara Rabhi, Ronay Ak, Gabriel de Souza Pereira Moreira, Jeong Min Lee, and Even Oldridge
    16th Women in Machine Learning Workshop (WiML) in NeurIPS 2021
    (Poster Presentation)
  • Transformers4Rec: Bridging the Gap between NLP and Sequential / Session-Based Recommendation
    Gabriel de Souza Pereira Moreira, Sara Rabhi, Jeong Min Lee, Ronay Ak, and Even Oldridge
    15th ACM Conference on Recommender Systems (RecSys) 2021
    (Paper) (BibTeX)
  • Neural Clinical Event Sequence Prediction through Personalized Online Adaptive Learning
    Jeong Min Lee and Milos Hauskrecht
    19th International Conference on Artificial Intelligence in Medicine (AIME) 2021Best Paper Nominee
    (Paper) (BibTeX) (Code)
  • Modeling Multivariate Clinical Event Time-series with Recurrent Temporal Mechanisms
    Jeong Min Lee and Milos Hauskrecht
    Artificial Intelligence in Medicine (Journal)
    (Paper) (BibTeX)
  • MTM: Multi-scale Temporal Memory for Clinical Event Time-Series Prediction
    Jeong Min Lee and Milos Hauskrecht
    18th International Conference on Artificial Intelligence in Medicine (AIME) 2020
    (Paper) (Presentation) (BibTeX)
  • Clinical Event Time-series Prediction with Periodic Events
    Jeong Min Lee and Milos Hauskrecht
    The Thirty-Third International FLAIRS Conference 2020
    (Paper) (BibTeX)
  • Recent context-based LSTM for Clinical Event Time-series Prediction
    Jeong Min Lee and Milos Hauskrecht
    17th Conference on Artificial Intelligence in Medicine (AIME) 2019Best Paper Award
    (Paper) (BibTeX)
  • Diagnosis Code Prediction from Electronic Health Records as Multilabel Text Classification: A Survey
    Jeong Min Lee and Aldrian Obaja Muis
    Neural Network for Natural Language Processing (NLP) at CMU - Fall 2017 Course Project
    (Paper)
  • Modeling Patient Mortality from Clinical Text by Combining Topic Modeling and Ontological Feature Learning with Group Regularization
    Jeong Min Lee, Charmgil Hong, and Milos Hauskrecht
    Machine Learning for Signal Processing at CMU - Fall 2016 Course Project
    (Paper) (Presentation)
  • Towards Next Generation Health Data Exploration: A Data Cube-based Investigation into Population Statistics for Tobacco
    James P. McCusker, Deborah L. McGuinness, Jeong Min Lee, Chavon Thomas, Paul Courtney, Zaria Tatalovich, Noshir S. Contractor, Glen D. Morgan, and Abdul R. Shaikh
    In Proceedings of Hawaii International Conference on System Sciences (HICSS) 2013Best Paper Award
    (Paper) (BibTeX)
  • Climate Change, Disaster and Sentiment Analysis over Social Media Mining
    Jeong Min Lee, James McCusker, and Deborah L. McGuinness
    American Geophysical Union (AGU) Fall Meeting 2012, Linked Data for Earth and Space Science Posters Session
    (Poster)
  • Public Health Surveillance Using Global Health Explorer
    James P. McCusker, Jeong Min Lee, Thomas Chavon, and Deborah L. McGuinness
    In Proceedings of Joint Workshop on Semantic Technologies Applied to Biomedical Informatics and Individualized Medicine (SATBI+SWIM), co-located with the 11th International Semantic Web Conference (ISWC) 2012
    (Paper) (BibTeX)

Work Experience

  • Meta / Instagram Menlo Park, CA · Dec 2020 - Present
    Research Scientist working on internet-scale ranking and personalization systems, including objective design from noisy implicit feedback, user-history representation, uncertainty-aware modeling, long-horizon value alignment, and debiasing.
  • NVIDIA Remote · May 2020 - Aug 2020
    Research Intern focused on transformer-based sequence modeling and recommendation; contributed to Transformers4Rec.
  • Facebook Seattle, WA · 2018 & 2019
    Machine Learning PhD Intern on Pages Sciences, working on content embeddings, representation learning, and sequence models for recommendation.
  • Earlier Roles
    Software engineering and research work in healthcare, public health, and data systems, including EMR work in Malawi and research at RPI.

Notes

Contact

Please reach out via LinkedIn, Google Scholar, or GitHub.