About Me

ezCaretech Research & Development Center
AI Research Engineer
Technical Research Personnel (전문연구요원), 2023.2 ~

* Email: hyeongmin0121@gmail.com

Research Interests: Large Language Model (LLM), Anomaly Detection, Generative Models, Natural Language Processing

[CV] [Github] [Google Scholar] [LinedIn]

---

Experience

ezCaretech Research & Development Center 2023.02 - Current
- AI Research Engineer (전문연구요원)

---

Education

Master’s in Engineering 2020.03 - 2022.02
- Korea University, Anam
(Advisor: Prof. Sangkyun Lee [Lab])

Bachelor’s in Computer Engineering 2015.03 - 2019.02
- Hanyang University, ERICA

---

Publications

2024

[6] Extracting lung cancer staging descriptors from pathology reports: a generative language model approach
Hyeongmin Cho, Sooyoung Yoo, Borham Kim, Sowon Jang, Leonard Sunwoo, Sanghwan Kim, Donghyoung Lee, Seok Kim, Sejin Nam and Jin-Haeng Chung
Journal of Biomedical Informatics (SCIE, IF 4.0)
(Topic: LLM-based Information Extraction)

[5] AutomatedPathologic TN Classification Prediction and Rationale Generation from LungCancer Surgical Pathology Reports using a Large Language Model Fine-Tunedwith Chain-of-Thought
Sanghwan Kim, Sowon Jang, Borham Kim, Leonard Sunwoo, Seok Kim, Jin-Haeng Chung, Sejin Nam, Hyeongmin Cho, Donghyoung Lee, Keehyuck Lee and Sooyoung Yoo
JMIR Medical Informatics (SCIE, IF 3.1)
(Topic: LLM Reasoning)

[4] Impact of Prompt Formats on Medical Question-Answering Performance of Large Language Model
Hyeongmin Cho, Donghyoung Lee, Inhu Kim and Sejin Nam
Conference on Korean Society of Medical Informatics (KOSMI)
(Topic: LLM Prompt Engineering)

2021

[3] Data Quality Measures and Efficient Evaluation Algorithms for Large-Scale High-Dimensional Data
Hyeongmin Cho, Sangkyun Lee
Applied Sciences (SCIE, IF 2.7), 11(2)
(Topic: ML Data Complexity)

[2] A study on Neural Network-based Anomaly Detection and Data Augmentation Method to Improve the Performance of Network IDS
Master’s Thesis
(Topic: Anomaly Detection)

2020

[1] Machine Learning Data Poisoning Quantification using Linear Discriminant Analysis
Hyeongmin Cho, Sangkyun Lee
Conference on Information Security and Cryptography
(Topic: Neural Network Vulnerability Attack)

---

Patents

[3] 생성형 모델을 학습시키는 장치 및 방법
(Application No. 10-2024-0186432)

[2] Network attack detection system and network attack detection method
(Application No. 10-2021-0057656, Patent No. 10-2525-5930000)

[1] Network intrusion detection system and network intrusion detection method
(Application No. 10-2021-0057630, Patent No. 10-2526-9350000)

---

Projects

Predicting Particulate Matter Concentration of Fine Dust for Air Quality Forecasting 2019 - 2022
with National Institute of Environmental Research (NIER)

Research on AI-based Network Instrusion Detection System (NIDS) & Explainable AI (XAI) 2021 - 2022
with LIG Nex1

Research on AI-based Network Instrusion Detection System (NIDS) 2020 - 2021
with LIG Nex1

Research on Quality Measures for Machine Learning Data 2019 - 2020
with Telecommunications Technology Association (TTA)

Predicting Stroke Severity from MRI Images 2019
with Korea University Ansan Hospital

---

Teaching

Samsung DS Expert 2021
- Teaching Assistant

Whimoon High School Special Lecture on AI 2021
- Instructor

Yangji High School C Programming 2018
- After School Instructor

---

Seminar

[14] Variational Autoencoder (Comprehensive) 2021
 - Bayesian Inference, Variational Inference, Variational AutoEncoder

[13] Anomaly Detection (Comprehensive) 2021
 - ODIN, AnoGAN, Deep-SVDD, DASVDD

[12] Neural Attention Distillation: Erasing Backdoor Triggers from Deep Neural Networks 2021
 - AI Backdoor Defense

[11] WaNet - Imperceptible Warping-based Backdoor Attack 2021
 - AI Backdoor Attack

[10] Sampling Methods 2021
 - Monte Carlo, Rejection Sampling, MCMC

[9] Weight Uncertainty in Neural Networks 2020
 - Bayesian Neural Network

[8] Hands-on Bayesian Neural Networks - a Tutorial for Deep Learning Users 2020
 - Bayesian Neural Network

[7] SinGAN: Learning a Generative Model from a Single Natural Image 2020
 - Generative Models

[6] Flow-based Generative models (NICE, GLOW) 2020
 - Generative Models

[5] Wasserstein GAN 2019
 - Generative Models

[4] Accessorize to a Crime: Real and Stealthy Attacks on State-of-the-Art Face Recognition 2019
 - AI Attack

[3] Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization 2019
 - Explainable AI (XAI)

[2] Part-based motion descriptor image for human action recognition 2019
 - Vision

[1] Image zooming using directional cubic convolution interpolation 2019
 - Vision

---