Baek Seong-Eun
I am a M.S. student at the Advanced Machine Intelligence Lab (AMI Lab), advised by Prof. Tae-Hyun Oh. I am currently enrolled in the Graduate School of Artificial Intelligence at POSTECH.
My research interests include multimodal foundation models, video understanding, and black-box optimization (Bayesian Optimization).
News
| Apr 30, 2026 | 🎉 Our paper “A Language-Guided Bayesian Optimization for Efficient LoRA Hyperparameter Search” has been accepted to ICML 2026. (First-author) |
|---|---|
| Nov 08, 2025 | 🎉 Our paper “Training-free Multimodal Embedding for Structure-Aware Retrieval of Scalable Vector Graphics and Images” has been accepted to WACV 2026. |
| Sep 16, 2025 | 🎉 Our paper “Do MLLMs Understand Arrows? Visual Motion Prompting and Motion-Aware Reasoning Make Motion Easier to Understand” has been accepted to ICCV 2025 Workshop on Short-Form Video Understanding (SVU). |
| Aug 31, 2024 | 🎓 Started M.S. program at the Graduate School of Artificial Intelligence, POSTECH. |
Publications
- ICMLA Language-Guided Bayesian Optimization for Efficient LoRA Hyperparameter SearchIn International Conference on Machine Learning (ICML), 2026
- WACVTraining-free Multimodal Embedding for Structure-Aware Retrieval of Scalable Vector Graphics and ImagesIn IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2026
- Accepted as a Round 1 paper (6.4% acceptance rate)
- Presented at the Workshop on Graphic Design Understanding and Generation (GDUG), in conjunction with ICCV 2025
- ICCV-WDo MLLMs Understand Arrows? Visual Motion Prompting and Motion-Aware Reasoning Make Motion Easier to UnderstandIn ICCV Workshop on Short-Form Video Understanding (SVU), 2025Under Review in ICML
Education
- M.S. in Artificial Intelligence, Graduate School of Artificial Intelligence, POSTECH (2024 – present)
- B.S. in Biomedical Engineering & Artificial Intelligence, Korea University (2020 – 2024)