Sunjun Hwang

About Me

I am an undergraduate researcher in Computer Science at Yonsei University, primarily focused on quantum computing and AI security. My research explores adversarial robustness, quantum-inspired learning, and hardware-aware AI systems, spanning both theoretical analysis and system-level implementation. I aim to develop secure and efficient AI models by bridging quantum mechanics, machine learning, and real-world deployment constraints.

Research Interests View Details

Quantum Computing

Quantum algorithms, NISQ-era circuits, variational quantum circuits (VQC)

AI Security

Adversarial robustness, federated learning security, post-hoc defense

AI Semiconductors

AI accelerators, neuromorphic computing

Quantum–ML Integration

Quantum machine learning, hybrid quantum-classical systems

Carla Autonomous Driving

High-fidelity driving simulation, sensor configuration, scenario generation

Education

Mar 2022 - Aug 2026

Bachelor of Science

Yonsei University, School of Computer Science

Relevant coursework:
Math & Science: Calculus (I, II), General Physics (I, II), General Chemistry (I, II), Discrete Mathematics, Linear Algebra, Numerical Partial Differential Equations, Graph Theory
CS Core: Computational Thinking, Computer Programming, Java Programming, Data Structures, Algorithms, Object Oriented Programming, System Programming, Operating System, Microprocessor, Software Engineering, Opensource and Linux Systems
Data & Signal: Database, Data Science, Datamining, Signal and Systems, Modeling and Simulation
AI & Security: Artificial Intelligence, Natural Language Processing, AI Mathematics, AI Security, Information Security Applications, Cryptography
Quantum: Quantum Mechanics (I, II)

GPA: 4.03 / 4.5

Sep 2026 - Apr 2027

Language Study Abroad

Vancouver, Canada

Intensive English language program focusing on academic English, professional communication, and cross-cultural collaboration. Enhancing global communication skills for international research and academic engagement.

Experience

Dec 2024 - Jun 2026

Undergraduate Researcher

RAISE Lab, Yonsei University

Research Assistant in Reliable Artificial Intelligence and System Engineering & Quantum Computing. Focused on innovative applications of reliable AI and quantum computing to enhance computational reliability and efficiency. Collaborated with researchers to design algorithms addressing AI reliability, interpretability, and system optimization. Conducted experiments integrating reliability frameworks into AI models, improving system robustness and trustworthiness. Engaged in interdisciplinary projects combining quantum mechanics principles with computational engineering approaches. Presented findings at lab seminars, contributing to discussions on advancing reliable computing technologies.

Mar 2025 - Jan 2026

CTO / AI & Web Developer

Youth Lab

Led technical strategy and development as CTO. Built and deployed AI-powered features and full-stack web applications. Managed the development pipeline, coordinated cross-functional collaboration, and made key architectural decisions for product development.

Jan 2026 - Feb 2026

Technical Advisor / AI Developer

Futurevel

Provided technical consulting on AI integration and system architecture. Advised on machine learning model deployment strategies and contributed to AI feature development for company products.

Jan 2025 - Feb 2025, Jan 2026 - Feb 2026

AI Developer & Team Leader — LG Aimers

LG AI Research Institute

Participated in two cohorts (6th and 8th) of the LG Aimers program. Led a team to develop AI models for real-world data analysis, including infertility patient data for early detection and LLM compression (EXAONE). Gained hands-on experience in end-to-end ML pipelines from data preprocessing to model evaluation.

Sep 2024 - Jun 2026

Teaching Assistant & Tutor

Yonsei University

Assisted professors and supported students across multiple courses over four semesters:
Java Programming (Fall 2024), Calculus & Vector Analysis I (Spring 2025), Engineering Mathematics I (Spring 2025), Artificial Intelligence (Fall 2025), Artificial Intelligence Mathematics (Spring 2026).
Conducted tutoring sessions, graded assignments, and provided one-on-one guidance to help students deepen their understanding.

Publications View All

Functional Recovery of Deep Neural Networks via Logit-Based External Calibration

Sunjun Hwang, Jihyuk Ha.

KICS 2026 (Accepted, Jan 16, 2026)

This paper presents techniques for recovering software errors and quantitatively analyzes the extent to which recovery is possible.

Adversarial Robustness Analysis of Deep Learning-Based Automatic Modulation Classification in Wireless Communication

Sunjun Hwang, Eunho Choi, Dohyun Hwang.

IEEE ICAIIC 2026 (Accepted, Dec 21, 2025)

This paper investigates the adversarial robustness of deep learning–based automatic modulation classification systems in wireless communication environments. Various attack scenarios and robustness evaluation metrics are analyzed to assess model reliability under adversarial perturbations.

Design and Implementation of an FPGA-Based Real-Time Voice Risk Detection System

Sunjun Hwang, Seunghui Ye.

KCS 2026 (Accepted, Dec 15, 2025)

This work presents the design and implementation of a real-time voice risk detection system on FPGA hardware, focusing on low-latency signal processing and robust emergency voice and scream detection in practical environments.

Quantum-Secured Hybrid Communication System for Tactical Military Networks: Implementation and Performance Analysis of BB84 Protocol Based on PennyLane

Sunjun Hwang.

Journal of the Korean Institute of Communications and Information Sciences (JKICS), 2026 (Accepted, Dec 01, 2025)

This paper proposes a quantum-secured hybrid communication system for tactical military networks. The BB84 quantum key distribution protocol is implemented using PennyLane, and its performance is analyzed in realistic communication scenarios.

Quantum Noise-based Adversarial Attack on Diffusion Models and Analysis of Defense Mechanisms

Sunjun Hwang.

KIIT-JICS 2026 (Accepted, Nov 2025)

This study explores a novel adversarial attack framework on diffusion models using quantum noise characteristics and evaluates defense mechanisms against such quantum-inspired perturbations.

Logit-based Knowledge Distillation for Heterogeneous Medical Image Federated Learning

Sunjun Hwang, Wooseok Wang, Jaehoon Lee.

Proceedings of KIIT Conference, 2025 (Accepted, Nov 03, 2025)

This paper proposes a logit-based knowledge distillation approach to improve performance in heterogeneous federated learning environments for medical image analysis while preserving data privacy.

Post-hoc Defense with Knowledge Distillation in Federated Learning: An Empirical Study against FGSM and PGD Attacks

Sunjun Hwang,Hongjoon Jun, Wooseok Wang, Jaehoon Lee.

Proceedings of KICS Conference, 2025 (Accepted, Oct 21, 2025)

This work presents a post-hoc defense strategy using knowledge distillation to enhance adversarial robustness in federated learning systems against FGSM and PGD attacks.

Classification of Pneumonia in Chest X-rays Using a Hybrid Neural Network Based on a 3-Qubit Quantum Circuit

Sunjun Hwang.

KSII Conference, 2025 (Accepted, Sep 23, 2025)

This paper introduces a hybrid quantum–classical neural network incorporating a 3-qubit quantum circuit for pneumonia classification from chest X-ray images.

Performance Comparison of 8 Deep Learning Models for Seismic Signal Denoising

Sunjun Hwang, Sehee Park, Kangmin Ko, Jiyun Baik

Proceedings of KIIT Conference, 2025 (Published, Sep 2025)

The paper conducts a systematic comparison of eight deep learning models—BiLSTM, Denoising Autoencoder, FFTformer, Informer, PatchTST, TCN, UNet1D, and WaveNet—for seismic signal denoising under identical experimental conditions.

A Study on Robustness Enhancement and Multi-Adversarial Attacks in Vision Transformer-based Image Classification Models

Sunjun Hwang, Hongjoon Jun, Sunje Kuem

Proceedings of KIIT Conference, 2025 (Published, May 2025)

This paper analyzes adversarial robustness in ViT-B32 models under FGSM, PGD, and CW attacks, demonstrating that multi-adversarial training significantly improves robustness while maintaining high clean accuracy.

Current Research View All

Autonomous Driving + Federated Learning (Heterogeneous Models) Active

2025 - Present

Training heterogeneous models via federated learning in the same driving environment. Analyzing knowledge transfer between models and performance convergence.

Federated Learning Heterogeneous Models Autonomous Driving

Fault Injection Research Active

2026 - Present

Research on software fault injection in deep neural networks and functional recovery via logit-based external calibration techniques.

Fault Injection DNN Recovery Logit Calibration Reliability

Quantum AI Active

2025 - Present

Quantum-classical hybrid neural networks, quantum noise-based adversarial attacks, and classification models utilizing Variational Quantum Circuits (VQC).

Quantum ML VQC Hybrid Network PennyLane

Autonomous Driving + LLM Active

2026 - Present

Leveraging Large Language Models (LLMs) for autonomous driving decision-making and scenario understanding. Includes research on OpenEMMA, an open-source end-to-end multimodal model for autonomous driving that integrates vision-language reasoning for real-time driving decisions.

LLM Autonomous Driving Decision Making OpenEMMA Multimodal

Cache-inspired Federated Learning Active

2026 - Present

A novel federated learning framework inspired by cache memory hierarchy. Borrowing locality and replacement policy concepts from cache systems to optimize client selection, model aggregation, and communication efficiency in FL.

Federated Learning Cache Memory Client Selection Aggregation

Homogeneous FL for TCP Autonomous Driving (Different Environments) Active

2026 - Present

Homogeneous federated learning research using TCP autonomous driving models across different driving environments. Training the same TCP model in diverse scenarios via FL to analyze robustness and performance convergence under environment heterogeneity.

Federated Learning TCP Autonomous Driving Carla Homogeneous FL

TDA for Medical Data Analysis Active

2026 - Present

Applying Topological Data Analysis (TDA) to BRCA breast cancer datasets to discover novel variables and hidden topological structures. Focused on identifying new biomarkers and feature patterns in breast cancer genomic data that conventional statistical methods fail to capture.

TDA Medical Data Topological Analysis Healthcare Cancer Data Biosignal

TDA-based Detection of Generative AI Attacks Planning

2026

Detecting adversarial attacks generated by generative AI models using Topological Data Analysis (TDA). Leveraging TDA to capture hidden topological features and structural anomalies that conventional detection methods cannot identify.

TDA Generative AI Adversarial Attack Detection Security

Completed Projects

AI-Based Dental Hygiene Hand Posture Training App Completed

2025 (with Dept. of Dental Hygiene)

A mobile application developed in collaboration with the Department of Dental Hygiene that uses AI-based hand pose estimation to evaluate whether the user's hand posture is correct during dental hygiene practice. Includes practice time tracking and feedback on posture accuracy.

Hand Pose Estimation Mobile App Dental Hygiene AI

AI-Based Aquaponics Monitoring & Growth Prediction Completed

2025 (with COREX)

Camera-based plant growth prediction and real-time AI monitoring of pH, temperature, and other environmental factors in aquaponics systems, with automated adjustment upon anomaly detection.

Computer Vision IoT Real-time Monitoring

Mathematical Analysis of Quantum GNN Completed

2025

Theoretical and mathematical analysis of Quantum Graph Neural Networks, investigating expressiveness, convergence properties, and computational advantages.

Quantum GNN Mathematical Analysis

Side Channel Attack Detection Completed

Mar 2025 - Jun 2025

Comparative analysis of AI models to identify the most effective architecture for detecting side-channel attacks in hardware systems. The MLP model’s efficiency and robustness underlined the importance of model-data alignment over architectural complexity.

PyTorch Scikit-learn PCA Security

OCR + LLM-Based Medical Document Automation Completed

Apr 2025 - Nov 2025

Automated processing and structuring of medical documents by combining OCR text extraction with LLM-based understanding and summarization.

Naver OCR OpenAI API NLP Ubuntu

Voice Data-Based Risk Situation Prediction Completed

Sep 2024 - Dec 2024

Detecting and predicting risk situations from voice data using audio feature extraction and deep learning classification. Follow-up research led to FPGA-based implementation (KCS 2026).

PyTorch Librosa Google Speech-to-Text API

LLM-Based Search Query Optimization Completed

Apr 2024 - Jun 2024

Optimizing search queries using LLMs to improve search accuracy and user experience through intent understanding and query reformulation.

LLM NLP Python

Electricity Usage Prediction Completed

2024

Predicting electricity consumption by integrating weather data and historical usage patterns using machine learning models.

Time Series Prediction Weather Data

Technical Skills

Languages: C, C++, Java, Python, Rust, Assembly ML/DL: PyTorch, TensorFlow, Keras, scikit-learn Quantum: Qiskit, Cirq, PennyLane Systems: CUDA, FPGA, Linux

Awards & Honors

Feb 2026

LG Aimers 8th Cohort — EXAONE Model Lightweight LLM Compression

LG AI Research Institute / Ministry of Employment and Labor

Completed LG Aimers 8th program (Jan–Feb 2026). Hackathon topic: EXAONE Model Lightweight LLM Compression.

Dec 2025

Outstanding Paper Award from the Korea Information Technology Society

KIIT

Jul 2025

Outstanding Paper Award from the Korea Information Technology Society

KIIT

2024

LG Aimers 6th Cohort — Data Intelligence (Pregnancy Success Prediction)

LG AI Research Institute / Ministry of Employment and Labor

Spring 2024

Academic Excellence Award

Yonsei University

Fall 2023

Academic Excellence Award

Yonsei University

Ongoing

Volunteering and Social Contributions

The Korean Red Cross

Certifications & Licenses

Mar 2026

SBR - Investigator/Researcher

CITI Program

Credential ID: 75784444 · Expires: Mar 2028

Feb 2026

LG Aimers 8th — LLM Compression

LG AI Research

Jan 2026

Certificate of Completion — LaTeX for Everyone and Everything

Udemy

Credential ID: UC-629e53fd-f56d-4f93-8aa7-43f4e9128ebb

Sep 2025

Certificate of Completion — Graduate Student

Schumpeter

May 2025

Certificate of Completion of the Artificial Intelligence Convergence Technology Expert Training Program

Korea Artificial-Intelligence Convergence Technology Society

Jan 2025

LG Aimers 6th — Data Intelligence

LG AI Research

2024

Software Engineer Literacy Seminar

The Korean Institute of Convergence Signal Processing

Professional Memberships

IEEE KICS (Korean Institute of Communications and Information Sciences) KIIT (Korea Information Technology Society)

Languages

Korean: Native English: Professional Working Proficiency