I’m interested in NLP, especially in parametric knowledge of LLMs.
Especially understanding (Evaluation on benchmark, membership attacking), unlocking (In-context learning, SFT, alignment learning, reasoning), and expanding (Continual learning) their parametric knowledge to meet ultimate human needs.
I'm currently starting my research on real-world LLM coding ability evaluation and their RL training.
Email: joonwon.lainshower@gmail.com
Curriculum Vitae : JOONWON_CV
LINKEDIN : MY_LINKEDIN_URL
Education
•
Master of Science in Graduate School of Artificial Intelligence, POSTECH. (2023.02 – 2025.02)
•
Bachelor's degree in Hotel &Tourism Management, Sejong University.(2017- 2023) (GPA: 4.37/4.5, Summa Cum Laude)
(International Student in School of Hotel and Tourism Managment HongKong Polytechnic University, 2018.09-2018.12)
Working Experience
•
LG AI Research @EXAONE LAB, LLM Research Intern (2025.03~)
•
ONOUT, LLM Research Engineer (Freelancer, 2024.07-2024.12)
Publications
[C]: Conference [J]: Journal [W]: Workshop [P]: Preprint
•
[C] Are Vision-Language Models Safe in the Wild? A Meme-Based Benchmark Study, EMNLP 2025 Main
DongGeon Lee*, Joonwon Jang*, Jihae Jeong, Hwanjo Yu.
•
[C] How Diversely Can Language Models Solve Problems? Exploring the Algorithmic Diversity of Model-Generated Code, EMNLP 2025 Findings
Seonghyeon Lee, HeeJae Chon, Joonwon Jang, Dongha Lee, Hwanjo Yu.
•
[P] EXAONE 4.0: Unified Large Language Models Integrating Non-reasoning and Reasoning Modes, Technical Report
LG AI Research.
•
[C] Verbosity-Aware Rationale Reduction: Effective Reduction of Redundant Rationale via Principled Criteria, ACL 2025 Findings
Joonwon Jang, Jaehee Kim, Wonbin Kweon, Hwanjo Yu.
•
[C] Eliciting Instruction-tuned Code Language Models' Capabilities to Utilize Auxiliary Function for Code Generation, EMNLP 2024 Findings
Seonghyeon Lee, Suyeon Kim, Joonwon Jang, Heejae Chon, Dongha Lee, Hwanjo Yu.
•
[W] EPR: An Expert Behavior-enhanced Paper Ranking Framework for the Automotive Industry, EMNLP 2024 Workshop
WooJoo Kim, Joonwon Jang, Jinyi Yu, Yunsu Jeon, Hwanjo Yu.
•
[C] Rectifying Demonstration Shortcut in In-Context Learning, NAACL 2024 Main
Joonwon Jang, Sanghwan Jang, Wonbin Kweon, Minjin Jeon, Hwanjo Yu.
•
[C] Hierarchical Graph Convolutional Network Approach for Detecting Low-Quality Documents, LREC-COLING 2024
Jaeyoung Lee, Joonwon Jang, Misuk Kim.
•
[C] Fixed Input Parameterization for Efficient Prompting, ACL 2023 Findings
Eunbi Choi, Yongrae Jo, Joel Jang, Joonwon Jang, Minjoon Seo.
•
[J] Eco-friendly platooning operation algorithm of the autonomous vehicles, JITS, 2023
Joonwon Jang, Sung Il Kwag, Young Dae Ko.
•
[C] Headline Token-based Discriminative Learning for Subheading Generation in News Article, EACL 2023 Findings
Joonwon Jang, Misuk Kim.
•
[J] Detecting incongruent news headlines with auxiliary textual information, ESWA, 2022
Joonwon Jang, Minju Kim, Yoonsik Cho, Misuk Kim.
Projects
•
EXAONE 4.0 (LG AI Research, 2025.03 – 2025.08)
Large Language Model, Vision Language Model
◦
Enhancing coding abilities: code/vision-language corpora for pre/post-training.
◦
Near SOTA on LiveCodeBench among open-source LLMs (strong at 32B).
results
◦
Spark-YARN cluster for large-scale preprocessing.
•
Continual Learning LLM toward Legal Domain (ONOUT, 2024.09 -2024.12)
Domain-Adaptive Pre-training
◦
Data crawling & preprocessing (Spark); DAPT with distributed training (FSDP/DeepSpeed).
◦
Achieved ×2 KMMLU in-domain; preserved general knowledge after DAPT.
•
AI Ranking Model for Promising Technologies Selection (Hyundai, 2023.08 -2024.08)
NLP Application, Recommender System
◦
Data crawling & Post-trained LMs using citation networks (SPECTER framework).
◦
Achieved a 42.8% improvement over a feature-based baseline.
◦
[W] EPR: An Expert Behavior-enhanced Paper Ranking Framework for the Automotive Industry
•
Fashion Advertisement Generation via Quantized LLM (ONOUT, 2023.06 -2023.09)
Supervised Fine-Tuning, Quantization
◦
Fine-tuned Polyglot 5.8B/12.8B with QLoRA (quantization-aware SFT & instruction tuning)
◦
Designed augmentation prompts with the GPT API; combined human-labeled data with self-instruct.
•
Prompt Injection in Chatbot System (KAIST, AI 2022.07-2022.12)
Long Context Handling & Chatbot
◦
Parameterized long dialogue context/persona (MSC) into student models to reduce inference cost.
◦
[C] Fixed Input Parameterization for Efficient Prompting.
•
YoYak (Long Sequence Summarization Framework For Korean) (, 2021.09 -2021.12)
Long Context Handling & Summarization
◦
TAPT Longformer training with Pegasus Objective Function for KoBART Model
◦
Performance comparison with vanilla KoBART
•
Incongruent News Detection (KOCCA, 2020.09-2022.06)
◦
Generating dataset for detecting incongruent news
◦
[J] Detecting incongruent news headlines with auxiliary textual information, ESWA, 2022
◦
[C] Hierarchical Graph Convolutional Network Approach for Detecting Low-Quality Documents, LREC-COLING 2024
•
Optimization of autonomous vehicle platooning (2019.05-2020.12)
◦
[J] Eco-friendly platooning operation algorithm of the autonomous vehicles, JITS, 2023
Activities
•
•
Wemajor (2018.03-2019.12)
◦
Introducing Major to middle/high school students (volunteering)
Awards & Scholarships
•
2024 POSTECH Best Paper Awards (Excellence Award)
•
2024 Hyundai MOBIS AI (Infotainment) Industry–Academic Cooperation Contest, 3rd prize
•
Academic Excellence Scholarship (Spring 2017, Fall 2017, Spring 2018, Fall 2021, Spring 2022)
•
Coding Challenge in Sejong University – Python (Fall 2021, 4th prize)
•
KIISE Undergraduate Student Paper Award (2021, 3rd prize)
Patents
•
가짜 뉴스 탐지용 데이터셋 생성 장치 및 이의 실행 방법, 출원번호: 10-2022-0025684 (2022)
Paper Review and Additional Study
Recent Personal Paper Review Recent Team Paper Review (2023~) Past Personal NLP Paper Review (2021~2022)Linear Algebra
Probability & Statistics
Data Structure
CS224N (2020.09-2020.12)
Skills
•
Programming languages – python, C, R
•
Data Mining - numpy, pandas, sklearn, MYSQL, SAS, SPARK
•
Machine Learning – tensorflow, keras, pytorch, pytorch-lightning, huggingface
•
Web Crawling – request, BeautifulSoup, FastAPI
•
Others – Git & Github, Flask, gephi, Docker
Related Course Work
•
Advanced Machine Learning (A+)
•
Linear Algebra and Programming (A+)
•
Introduction to Open Source (A+)
•
Data Problem&Solution and Practice (A+)
•
C Programming and Lab (A+)
•
Python Programming (A0)
•
Computer Structure (A+)
•
Database (A+)