Sanghyun Park

A student full of curiosity

ducks

Hello, I’m Sanghyun Park!

I am currently an M.S. student in the Department of School of Computing at KAIST, interested in fuzzing, and overall software security.

Education

  • M.S. in School of Computing, KAIST, 2023 - Present
    • Thesis (in progress): Systematic 1-Day Bug Reproduction with Large Language Model
    • Advisor: Prof. Sang Kil Cha
  • B.S. in Computer Science and Engineering, Chungnam National University(CNU), 2017 - 2023
    • GPA: 4.21/4.5 (Rank 2/102 in CSE, Major GPA: 4.27/4.5)
    • Advisor: Prof. Hoon Choi

Projects

  • LLM1dFuzz
    • Proposed a systematic 1-day bug reproduction tool using GPT-4 Turbo.
    • Improved directed fuzzing performance by 37.5% across 15 programs using LLM-generated seeds.
  • Fuzzing Survey
    • Maintaining an up-to-date genealogy database of over 361+ fuzzers.
  • ChatFive
    • Developed a system utilizing multiple LLM agents to enhance user experience in personality tests.
    • Achieved 77% accuracy in key traits, with 85% of users preferring the system over the Big Five test.
  • Edge-Cloud Migration of Intruder Detection System
    • Designed a communication protocol for edge-cloud migration to accelerate response in critical situations.
    • Implemented custom APIs using FastAPI, enabling seamless updates and integration for developers.
  • DasMin
    • Architected a high-performance distributed IoT data storage system utilizing MinIO.
    • Improved GET request performance by 347% and PUT request by 155% across 4 nodes.
  • Shot Tracker
    • Designed a novel shot group measurement IoT system using LoRa and YOLO V5.
    • Achieved 91.8% mean accuracy in indoor shot group measurements.

Experience

  • Graduate Researcher, KAIST Software Security Lab, Mar 2023 - Present

  • Research Intern, CNU Mobile Distributed Computing Lab, Mar 2022 - Feb 2023

  • Visiting Scholar, Purdue University, Dec 2021 - Feb 2022

  • Research Intern, CNU Software Analysis and Testing Lab, Jul 2021 - Dec 2021

Publications

International Conferences

  • Systematic Bug Reproduction with Large Language Model
    • Sanghyun Park, Haeun Lee, Sang Kil Cha
    • In proceedings of the ESORICS Workshop on Security and AI (SECAI), 2024, [pdf]
  • ChatFive: Enhancing User Experience in Likert Scale Personality Test through Interactive Conversation with LLM Agents
    • Jungjae Lee, Yubin Choi, Minhyuk Song, Sanghyun Park
    • In proceedings of the ACM Conference on Conversational User Interfaces (CUI), 2024, [pdf]
  • Feasibility of Measuring Shot Group Using LoRa Technology and YOLO V5
    • Sanghyun Park, Dongheon Lee, Jisoo Choi, Dohyeon Ko, Minji Lee, et al.
    • In proceedings of the IEEE Sensors Applications Symposium (SAS), 2022, [pdf]

Domestic Conferences

  • Design and Implementation of Communication Protocol for Edge-Cloud based Power Plant Surveillance System
    • Yewon Kim, Sanghyun Park, Honggi Oh, Jongik Kim, Hoon Choi
    • 2022 KICS Fall Conference, 2022
  • A Study on Distributed Data Storing Edge Node and Data Pipeline for Efficient IoT Data Acquisition
    • Honggi Oh, Sanghyun Park, Yewon Kim, Jongik Kim, Hoon Choi
    • 2022 KICS Fall Conference, 2022

Selected Awards

  • 2022 CNU Engineer of the Year Award

  • 2022 SW Festival Outstanding Project Award
    • Corporate Award
  • 2022 CNU SW/AI Project Fair
    • 1st Prize
  • 2021 CNU Creative SW Festival Club Performance Presentation
    • 2nd Prize

Others

  • Mandatory Military Service
    • Social Service Agent, Republic of Korea Army (2019-2021)
  • Skills
    • C, Python, Bash, SQL, Docker, Git
  • Languages
    • Korean(Native), English(Fluent)
    • OPIc Intermediate High - (2H4346729438)
  • Certificates
    • SQL Developer - (SQLD-041006166)
  • Academic Club
    • S.P.G (System Programming Group) at CNU
    • President of the Club in 2021