Dongwon Shin
Dongwon Shin

Security Researcher

About Me

I am a Ph.D. student @KAIST WSP Lab under the supervision of Sooel Son. My primary research interests lie in various attacks and defenses against websites providing privacy-sensitive services. I am also interested in analyzing diverse security issues related to large language models. If you have any questions, feel free to reach out to me via email (dongwon.shin@kaist.ac.kr).

Education

Mar. 2024 - Ph.D. student, Graduate School of Information Security, KAIST (Advisor: Sooel Son)
Mar. 2022 - Feb. 2024 Master's student, Graduate School of Information Security, KAIST (Advisor: Sooel Son)
Mar. 2018 - Feb. 2022 B.S., Cyber Security, Ajou University (Advisor: Taeshik Shon)

Publications

(*: equal contribution)
  1. Private Investigator: Extracting Personally Identifiable Information from Large Language Models Using Optimized Prompts
    Seongho Keum, Dongwon Shin, Leo Marchyok, Sanghyun Hong, and Sooel Son.
    34th USENIX Security Symposium (USENIX Security 2025).
    [paper] [bib] [code]

  2. You Only Perturb Once: Bypassing (Robust) Ad-Blockers Using Universal Adversarial Perturbations
    Dongwon Shin, Suyoung Lee, Sanghyun Hong, and Sooel Son.
    40th Annual Computer Security Applications Conference (ACSAC 2024).
    [paper] [bib] [slides] [code]

  3. RICC: Robust Collective Classification of Sybil Accounts
    Dongwon Shin*, Suyoung Lee*, and Sooel Son.
    The Web Conference 2023: Security, Privacy, and Trust Research Track (WWW 2023).
    [paper] [bib] [slides] [code]

Research Experiences

Web Security & Privacy Lab, KAIST
Ph.D. student, (Mar. 2022 - present)
Web Security, Machine Learning

Information Communication Security Lab, Ajou University
Undergraduate research intern, (Sep. 2020 - Aug. 2021)
Digital Forensics