Hyunsik Jeon

Hyunsik Jeon

Senior Machine Learning Engineer at Roblox

I am a Senior Machine Learning Engineer at Roblox. My research focuses on recommender systems, large language models, and multimodal learning, with an emphasis on understanding and reasoning about user needs.

Previously, I was a Senior Applied Scientist at Microsoft and a postdoctoral researcher at UC San Diego CSE with Prof. Julian McAuley. I received my Ph.D. and M.Sc. from Seoul National University, advised by Prof. U Kang, and my B.Sc. from Hanyang University.

CV Google Scholar DBLP LinkedIn GitHub
Portrait of Hyunsik Jeon

Research

My research traces a path from behavioral patterns to understanding and reasoning in recommender systems. I began with collaborative filtering under sparsity, cold-start recommendation, and calibrated ranking. I later incorporated commonsense knowledge, natural-language dialogue, and visual content into recommendation. I am now interested in adaptive systems that reason about items and user needs while remaining practical for real-world deployment.

Recommender Systems Large Language Models Multimodal Learning Efficient AI

Public datasets

  • Visually-aware conversational recommendation

    A multimodal conversational recommendation dataset with item images.

    paper dataset & code
  • Action recommendation for smart homes

    A dataset for context-aware smart-home action recommendation.

    paper dataset

Publications

Self-EvolveRec: Self-Evolving Recommender Systems with LLM-based Directional Feedback

Sein Kim, Sangwu Park, HongSeok Kang, Wonjoong Kim, Jimin Seo, Yeonjun In, Kanghoon Yoon, Hyunsik Jeon, and Chanyoung Park

NeurIPS 2026

arXiv code

Adapting Large Vision-Language Models to Visually-Aware Conversational Recommendation

Hyunsik Jeon, Satoshi Koide, Yu Wang, Zhankui He, and Julian McAuley

KDD 2025, Toronto, Canada

paper slides poster code bib

Calibration-Disentangled Learning and Relevance-Prioritized Reranking for Calibrated Sequential Recommendation

Hyunsik Jeon, Se-eun Yoon, and Julian McAuley

CIKM 2024, Boise, USA

paper slides code bib

Neighborhood-Based Collaborative Filtering for Conversational Recommendation

Zhouhang Xie*, Junda Wu*, Hyunsik Jeon*, Zhankui He, Harald Steck, Rahul Jha, Dawen Liang, Nathan Kallus, and Julian McAuley

RecSys 2024, Short Paper, Bari, Italy

paper poster code bib

Cold-start Bundle Recommendation via Popularity-based Coalescence and Curriculum Heating

Hyunsik Jeon, Jong-eun Lee, Jeongin Yun, and U Kang

The Web Conference 2024, Oral Presentation, Singapore

paper slides poster code bib

Imagery as Inquiry: Exploring a Multimodal Dataset for Conversational Recommendation

Se-eun Yoon, Hyunsik Jeon, and Julian McAuley

arXiv, 2024

paper

Aggregately Diversified Bundle Recommendation via Popularity Debiasing and Configuration-aware Reranking

Hyunsik Jeon, Jongjin Kim, Jaeri Lee, Jong-eun Lee, and U Kang

PAKDD 2023, Osaka, Japan

paper slides code bib

Diversely Regularized Matrix Factorization for Accurate and Aggregately Diversified Recommendation

Jongjin Kim, Hyunsik Jeon, Jaeri Lee, and U Kang

PAKDD 2023, Osaka, Japan

paper arXiv code bib

Accurate Bundle Matching and Generation via Multitask Learning with Partially Shared Parameters

Hyunsik Jeon, Jun-Gi Jang, Taehun Kim, and U Kang

PLOS ONE, 2023

paper arXiv code bib

Accurate Action Recommendation for Smart Home via Two-Level Encoders and Commonsense Knowledge

Hyunsik Jeon, Jongjin Kim, Hoyoung Yoon, Jaeri Lee, and U Kang

CIKM 2022, Atlanta, USA · SIGIR Student Travel Grant

paper slides code dataset bib

Accurate Node Feature Estimation with Structured Variational Graph Autoencoder

Jaemin Yoo, Hyunsik Jeon, Jinhong Jung, and U Kang

KDD 2022, Washington, D.C.

paper slides code bib

PGT: News Recommendation Coalescing Personal and Global Temporal Preferences

Bonhun Koo, Hyunsik Jeon, and U Kang

Knowledge and Information Systems, 2021

paper bib

Multi-EPL: Accurate Multi-Source Domain Adaptation

Seongmin Lee, Hyunsik Jeon, and U Kang

PLOS ONE, 2021

paper code bib

Unsupervised Multi-Source Domain Adaptation with No Observable Source Data

Hyunsik Jeon, Seongmin Lee, and U Kang

PLOS ONE, 2021

paper code bib

Accurate News Recommendation Coalescing Personal and Global Temporal Preferences

Bonhun Koo, Hyunsik Jeon, and U Kang

PAKDD 2020, Best Student Paper, Singapore

paper project bib

Data Context Adaptation for Accurate Recommendation with Additional Information

Hyunsik Jeon, Bonhun Koo, and U Kang

IEEE BigData 2019, Los Angeles, USA

paper slides project bib

Belief Propagation Network for Hard Inductive Semi-supervised Learning

Jaemin Yoo, Hyunsik Jeon, and U Kang

IJCAI 2019, Macao, China

paper slides poster code bib

UniWalk: Explainable and Accurate Recommendation for Rating and Network Data

Haekyu Park, Hyunsik Jeon, Junghwan Kim, Beunguk Ahn, and U Kang

arXiv, 2017

paper project bib

Experience & Education

Experience

  1. Roblox

    Senior Machine Learning Engineer

  2. Microsoft

    Senior Applied Scientist · Redmond, WA

  3. University of California San Diego

    Postdoctoral Researcher, Computer Science and Engineering

  4. Hyperconnect

    Research Intern, Machine Learning Team

Education

  1. Seoul National University

    Ph.D., Computer Science and Engineering

  2. Seoul National University

    M.Sc., Computer Science and Engineering

  3. Hanyang University

    B.Sc., Computer Science and Engineering

Awards & Grants