CV

PhD in Electrical and Computer Engineering

• Thesis: Solving Contemporary Security Problems in Deep Learning
• Advisor: Professor Yunheung Paek

BEng in Electrical and Electronic Engineering

• Degree awarded Aug 2016

Assistant Professor (tenure-track)

Postdoctoral Researcher

• Supervisor: Professor Yunheung Paek

Lecturer

Lecturer

VulScope: Software Vulnerability Detection via Local Context Analysis and Verification

• Hyunjun Kim, Younghan Lee, Woorim Han, Subin Yun, Kang-Won Lee, and Yunheung Paek.
• Empirical Software Engineering. Accepted for publication on 11 May 2026; final bibliographic details pending.
• Role: Co-first author.

Rethinking the Capability of Fine-Tuned Language Models for Automated Vulnerability Repair

• Woorim Han, Yeongjun Kwak, Miseon Yu, Kyeongmin Kim, Younghan Lee, Hyungon Moon, and Yunheung Paek.
• 48th IEEE/ACM International Conference on Software Engineering (ICSE). DOI 10.1145/3744916.3773167.
• Role: Co-corresponding author.

PI-FD: Property Inference Attacks on Federated Learning With Distillation via Synthetic Data Generation

• Yungi Cho, Miseon Yu, Jaejin Kim, Younghan Lee, and Yunheung Paek.
• IEEE Access 14, 101398–101414. DOI 10.1109/ACCESS.2026.3700259.
• Role: Co-corresponding author.

Poster: A Case Study on Automated Vulnerability Repair Using Pre-Trained Language Models

• Woorim Han, Miseon Yu, Younghan Lee, Hyungon Moon, and Yunheung Paek.
• IEEE Silicon Valley Cybersecurity Conference, 1–3. DOI 10.1109/SVCC65277.2025.11133621.
• Role: Co-corresponding author.

VFLIP: A Backdoor Defense for Vertical Federated Learning via Identification and Purification

• Yungi Cho, Woorim Han, Miseon Yu, Younghan Lee, Ho Bae, and Yunheung Paek.
• ESORICS, LNCS 14985, 291–312. DOI 10.1007/978-3-031-70903-6_15.
• Role: Co-author.

FLGuard: Byzantine-Robust Federated Learning via Ensemble of Contrastive Models

• Younghan Lee, Yungi Cho, Woorim Han, Ho Bae, and Yunheung Paek.
• ESORICS, LNCS 14347, 65–84. DOI 10.1007/978-3-031-51482-1_4.
• Role: First author.

Precise Extraction of Deep Learning Models via Side-Channel Attacks on Edge/Endpoint Devices

• Younghan Lee, Sohee Jun, Yungi Cho, Woorim Han, Hyungon Moon, and Yunheung Paek.
• ESORICS, LNCS 13556, 364–383. DOI 10.1007/978-3-031-17143-7_18.
• Role: First author.

Learn2Evade: Learning-Based Generative Model for Evading PDF Malware Classifiers

• Ho Bae, Younghan Lee, Yohan Kim, Uiwon Hwang, Sungroh Yoon, and Yunheung Paek.
• IEEE Transactions on Artificial Intelligence 2(4), 299–313. DOI 10.1109/TAI.2021.3103139.
• Role: Co-first author.

AI-Based Automated Vulnerability Detection and Secure Code Generation

• Participating Researcher, project RS-2024-00439819

Quantitative Sleep Microstructure Optimisation for Cognitive and Emotional Enhancement

• Participating Researcher, project RS-2024-00359247

Confidential-Computing-Based Privacy-Preserving Federated Learning Platform and Security Technologies

• Participating Researcher, project RS-2024-00351898

Leadership and Conference Service

• Editorial Board Member, KIPS Review, 2025–present
• Executive Secretary, Artificial Intelligence System Research Group, 2024–present
• Member, Executive Director Board, 2024–2025
• Vice Chair, Academic Committee, ASK 2025
• Programme Committee Member, WISA 2026
• Programme Organising Committee Member, 2026 Joint Workshop of the AI Systems and Data Security & Privacy Research Groups
• Session Chair, ASK 2025, ACK 2025, and ASK 2026

External and Alumni Service

• Secretariat Member, Imperial College London Alumni Association Korea
• External Evaluator, Seocho Resident Living Lab Idea Competition

Peer Review

• IEEE Transactions on Dependable and Secure Computing, 2025
• ACM Transactions on Software Engineering and Methodology, 2025
• Journal of Information Processing Systems, 2025–2026
• KIPS Transactions on Computer and Communication Systems, 2025–2026

Korean Patent Applications 10-2025-0189598 and 10-2026-0110736

• Backdoor-robust multimodal-model training and inference
• Co-inventors: Ho Bae, Youngseo Kim, and Younghan Lee

External Advisory Engagement

• Consultation on the impact of AI development on the software industry

Korean Patent Application 10-2026-0016350

• Multi-LLM clinical decision support for automated interpretation of sleep questionnaires and sleep-disorder correlation search
• Co-inventors: Jooyoung Kim and Younghan Lee

Computer Program Copyright Registration C-2025-056782

• Multimodal Learning Model and Method for Backdoor Defence Using Randomised Response
• Joint institutional registration by the university-industry cooperation foundations of Ewha Womans University and Sungshin Women's University
• Younghan Lee is named in the institutional inventor disclosure

Computer Program Copyright Registration C-2024-056087

• Positional and REM-Dominant OSA Classification Model with Differential Privacy
• Joint institutional registration by the university-industry cooperation foundations of Ewha Womans University and Sungshin Women's University
• Younghan Lee is named as the representative inventor in Sungshin Women's University's institutional disclosure

LLM Security: RAG Poisoning Attacks and Defences

AI Security: Leveraging Large Language Models to Address Security Challenges

• Three-hour English-medium international lecture

LLM-Based Automated Vulnerability Detection and Patch Generation

AI Agents and Automation: Current Research Trends and Innovative Applications

Multi-Agent Systems

Security of Deep Learning Models in IoT/Edge Devices

Invited Seminar

Theory of Information Security