Sunghyun Lee
Incoming KAIST AI Researcher. AI Creativity Explorer.
Kim Jaechul Graduate School of AI, Artificial Intelligence
KAIST(Korean Institute of Advanced Science and Technology)
Daejeon, Republic of Korea
Email: romanticbox@kaist.ac.kr
๐ Hi, Iโm Sunghyun Lee
Iโm a Deep Learning researcher with a deep interest in Multimodal ToM, AI for media, and Computational Creativity.
Currently, my keyword of research is conversation.
Iโm keep updating my website!
Beyond that, I explore the creativity of AI โ how machines can generate novel ideas, expressions, or artifacts that go beyond human expectations, and how we might evaluate such output meaningfully.
I majored Computer Science and French Language and Literature in Yonsei University, graduated as summa cum laude (top 3%).
๐ง Research Interests
- Multimodal understanding and generation
- HumanโAI interaction and creativity
- Pragmatics-based human interaction understanding and generation
- Storytelling and media generation by AI
- Computational artistry and creativity
โจ Motto
โThe future belongs to those who can hear it coming.โ โ David Bowie
Feel free to check out my publications and recent blog posts, or get in touch through the icons below!
news
| Nov 08, 2025 | Iโm so happy to share a news that our paper has been accepted to AAAI-26 as oral presentation!! Do you think MLLMs understand sound as human understands? For some models, the answer is YES, for others, the answer is NO. GPT and Gemini seems to not infer the sound as humans does. However, Qwen2.5 resembled the result of experiment conducted by humans. Our research digged down to this question. This is the first paper I participated as co-author, researching with Jinhong Jeong and supervised by professor Youngjae Yu. I would like to appreciate all authors; Jinhong Jeong, Jaeyoung Lee, Seonah Han, and Youngjae Yu. Jinhong Jeong provided fabulous and deep insights with linguistics, leading the whole research. Jaeyoung Lee provided marvelous ideas with the research, helping to turn the idea of the research to consider mechanistic interpretability with our research. Seonah Han devoted a lot of effort with ideation during our meeting, and spent lots of effort in constructing and preprocessing the dataset. And, Youngjae Yu supervised our research a lot, always helped our research to think out of the box and to think about key questions that researchers would ask. I, Sunghyun Lee, took effort in constructing the dataset, set the experiments to be precise and persuasive (introducing semantic dimension for the experiment), and analyzed the attention layers. Here I share you the links: [github] [arxiv] See you in Singapore! |
|---|---|
| Sep 18, 2025 | I am pleased to announce my admission to KAIST (Korean Advanced Institute of Science and Technology) for a Masterโs degree program. I will enroll in March 2026 at the Kim Jaecheol AI Graduate School (๊น์ฌ์ฒ AI ๋ํ์) under the supervision of Professor Ro Yong Man. I am delighted to join IVY Lab & IVL Lab as a new member! Furthermore, I would like to express my sincere and profound gratitude to Professor Yu Young Jae (Seoul National University) for supervising and supporting all research activities during my undergraduate research internship period. All my participation and accomplishments during my undergraduate studies would not have been possible without the guidance of the professor and the members of the MIR lab. See you in Daejeon! |
| Jun 26, 2025 | Welcome to my personal website! Iโm excited to share my research and experiences in AI and machine learning. |
