CV

Curriculum Vitae of Woocheol Shin.

Contact Information

Name Woocheol Shin
Professional Title M.S. Student
Email woofe@kaist.ac.kr
Phone (+82) 010-4340-2018

Professional Summary

I am broadly interested in building generative policies for sequential decision-making toward human-desired goals in the real world. My current research focuses on steering large generative models, especially diffusion and flow-based models, toward desired outputs using reinforcement learning and reward-guided optimization.

Education

Research Experience

  • 2024 - 2025

    Daejeon, South Korea

    Undergraduate Research Intern
    System Intelligence Lab, KAIST
    Advisor: Prof. Jinkyoo Park. Focus: Generative model alignment, Diffusion/Flow models, Reinforcement learning.
  • 2022 - 2024

    Ulsan, South Korea

    Undergraduate Researcher, Seminar Organizer
    Data Mining Lab, UNIST
    Advisor: Prof. Junghoon Kim. Focus: Network stability, Cohesive subgraph discovery.

Industrial Projects

  • 2025 - Present

    Researcher
    Collaborative Project with ETRI
    Topic: Reinforcement Learning-based Job Scheduling Policy Learning

Teaching Experience

  • 2025 - 2025

    Teaching Assistant
    KAIST
    IE437: Data-Driven Decision Making and Control

Publications

  • 2026
    Automated Kernel Discovery Towards Understanding High-dimensional Bayesian Optimization
    ICML 2026 Workshop
  • 2026
    Aligning Few-Step Generative Models by Amortizing Sample-based Variational Inference
    ICML 2026 Workshop
  • 2026
    Diffusion Fine-Tuning via Reparameterized Policy Gradient of the Soft Q-Function
    ICLR 2026
  • 2026
    Diffusion Alignment as Variational Expectation-Maximization
    ICLR 2026
  • 2026
    Urban Traffic Network Layout Optimization with Guided Discrete Diffusion Models
    WSDM 2026
  • 2025
    Beyond Trivial Edges: A Fractional Approach to Cohesive Subgraph Detection in Hypergraphs
    Knowledge-Based Systems

Awards

  • 2025
    Excellence Award, ISE & GSDS Research Day Poster Competition
    KAIST

    For the poster: Diffusion Finetuning via Reparameterized Policy Gradient of the Soft Q-function

  • 2024
    Dean's List
    UNIST CSE

    Awarded to top 4 students out of 224

  • 2024
    Superior Semester Grades
    UNIST

    Spring 2023 (4.3/4.3), Fall 2023 (4.24/4.3), Spring 2024 (4.3/4.3)

Scholarships

  • 2026
    Master's Excellence Scholarship in Science & Engineering
    Korea Student Aid Foundation
  •
    KAIST Scholarship
    KAIST

    Whole semesters

  • Academic Performance Scholarship
    UNIST

    Whole semesters

Patent

  • 2025
    Data Analysis System and Method for Detecting Cohesive Subgraph in Hypergraphs

    Korean Patent Application No. 10-2025-0107601. Applicants: UNIST; Industry-Academic Cooperation Foundation, Kongju National University. Inventors: Junghoon Kim, Hyewon Kim, Woocheol Shin, Dahee Kim, Hyunji Jeong.

Projects

  • UIRP, UNIST (Undergraduate Interdisciplinary Research Project)

    Topic: Generalised Anchoring Methods for Maximising User Engagement in Complex Social Networks

  • Creative Independent Project, UNIST AIGS

    Topic: A Budget Minimisation to Reinforcing Social Engagement

  • AICP, UNIST (UNIST AI Challengers Program)

    Topic: Bipartite Cohesive Subgraph Discovery in Networks

Other Experiences

  • 2022 - 2023

    Instructor
    NAVER CONNECT 1784 School Revolution Program, NAVER
    Teaching AI, SW, and Autonomous Vehicles
  • 2022 - 2022

    Instructor
    ARAMCO Coding School, ARAMCO
    Teaching Programming, iOS app development
  • 2019 - 2019

    Instructor
    Education Donation Activities by University Students, KOFAC
    Teaching Programming

References

  • Prof. Jinkyoo Park

    Associate Professor, Department of Industrial and Systems Engineering, KAIST.
    Adjunct Associate Professor, Graduate School of Data Science and Graduate School of AI, KAIST.
    Email: jinkyoo.park@kaist.ac.kr

  • Prof. Junghoon Kim

    Assistant Professor, Department of Computer Science and Engineering, UNIST.
    Email: junghoon.kim@unist.ac.kr