cv

Curriculum vitae and publications for Hyungwook Choi.

Contact Information

Name Hyungwook Choi
Professional Title PhD Student, Visual Information Processing Lab, Seoul National University
Email hwchooi224@gmail.com
Website https://gogiants1.github.io

Professional Summary

I am a PhD student at the Visual Information Processing Lab at Seoul National University. My research interests include generative modeling, multi-modality, multi-modal large language models, and data selection for MLLMs.

Experience

  • Mar 2025 - Apr 2026

    Research Engineer Intern / Industry-Academia Collaboration
    NAVER Cloud
    Vision Understanding / HyperCLOVA X
    • Conducted model-based evaluation and labeling of large-scale proprietary MLLM SFT data to identify and filter low-quality samples.
    • Performed ablation studies on a 3B parameter model to analyze the impact of model components and data variations.
    • Completed a six-month AI research engineering internship, followed by industry-academia collaboration through April 2026.
  • Jul 2022 - Apr 2023

    Co-founder, Full-Stack Developer
    ColaboAI
    Fitness and online music collaboration services
    • Designed and developed a full-stack application for finding fitness mates.
    • Developed an online music collaboration service using React and Django.
  • Mar 2022 - Jun 2022

    Software Engineering Intern
    FriendliAI
    PeriFlow MLOps product
    • Implemented an Angular web dashboard with data visualization for the PeriFlow MLOps product.
    • Built a linting plugin library to improve team code quality.
  • Dec 2021 - Feb 2022

    Software Engineering Intern
    NAVER
    AutoML MVP product
    • Developed an AutoML MVP product using React.js and Golang.
    • Studied AutoML theory to support product prototyping.
  • Jul 2021 - Aug 2021

    Software Engineering Intern
    Paprika Data Lab
    Cada mobile application
    • Developed and released a React Native image-swipe feature for reviewing crowd-sourced data, tested with 23,000 users.
    • Built a payment page using React.js and Firebase.

Education

  • Mar 2026 - Present

    Ph.D. Student
    Seoul National University
    Computer Science and Engineering
    • Advisor: Prof. Joonseok Lee
    • Researching multi-modal large language models, generative modeling, and data selection for MLLMs.
  • Mar 2024 - Feb 2026

    M.S.
    Seoul National University
    Computer Science and Engineering
    • Advisor: Prof. Joonseok Lee
    • Research interests: multimodal large language models and learning to rank in computer vision.
  • Mar 2016 - Feb 2023

    B.S.
    Seoul National University
    Computer Science and Engineering
    • Minor: Civil and Environmental Engineering

Academic Interests

Research: Generative modeling, Multi-modality, Multi-modal large language models, Data selection for MLLMs

Publications

Projects

  • Scene Text Aware Style Transfer Diffusion Model

    Research project, Mar 2024 - Jun 2024

    • Led a project on combining scene text editing with style transfer.
    • Developed a Latent Diffusion Model-based approach that integrates style transfer and scene text generation.
    • Analyzed and refined cross-attention behavior in the LDM U-Net to improve text accuracy and reduce style entanglement.
  • Social Fitness App

    React Native and FastAPI mobile app, Dec 2022 -

    • Designed and developed a cross-platform mobile app for finding fitness mates.
    • Built real-time chat with Redis-backed messaging and WebSocket communication.
  • Hyper Local Social Media

    React web service, Feb 2023 -

    • Planned and implemented a personalized memory sharing platform on a map with Google Maps integration and note writing.
    • Won 1st place in an on-campus web development club hackathon.
  • Cafeteria Menu Notifier App in SNU

    FastAPI and Python server, Sep 2022 -

    • Maintained and extended a food review and cafeteria menu crawling server.
  • Analysis of Jeonse Loans Data in Seoul's Gwanak District

    Data analysis project, Jun 2022

    • Collected Jeonse price data through web scraping and analyzed it with multiple regression, hedonic price modeling, and machine learning.
    • Found a 25% Jeonse price difference associated with loan eligibility.
  • Cover Song Sharing Service

    React and Django web service, Sep 2021 - Dec 2021

    • Planned and implemented a music editing-focused web service.
    • Practiced test-driven development with Python unittest and React Testing Library.
  • Hands-free Recipe-reading App

    React Native mobile app, Mar 2021 - Jun 2021

    • Built a mobile app that reads recipes aloud using Text To Speech and the YouTube API.
    • Implemented recognition and speech output for recipe text in images.