Seojin Kim (김서진)

I am a M.S student at KAIST Kim Jaechul Graduate School of AI.

I am now a member of BigDyl (at Daejeon, South Korea), advised by Noseong Park. 

My main research interest is Efficient Language Modeling, including KV cache and State Space Models, currently emerging foundational models. 

I am also interested in time series analysis and recommendation system.

I obtained my Bachelor's degree in Applied Statistics and Computer Science at Yonsei University, Seoul, Republic of Korea.

• 2024.09~Present, M.S. in Artificial Intelligence, KAIST Kim Jaechul Graduate School of AI (Daejeon, Rep. of Korea)

• 2020.03~2024.08, B.S. in Applied Statistics & Computer Science, Yonsei Universitiy (Seoul, Rep. of Korea)

• 2022.09~2022.12, Exchange Student, Dept. of Science & Engineering, University of Glasgow (Glasgow, United Kingdom)

• Yehjin Shin, Jeongwhan Choi, Seojin Kim and Noseong Park, "TV-Rec: Time-Variant Convolutional Filter for Sequential Recommendation", The 39th Conference on Neural Information Processing Systems (NeurIPS), 2025

• Yehjin Shin*, Seojin Kim*, and Noseong Park, "Graph Signal Processing Meets Mamba2: Adaptive Filter Bank via Delta Modulation," Workshop on Efficient Systems for Foundation Models at International Conference on Machine Learning (ES-FoMO III), 2025, *Equal Contribution

• Seojin Kim*, Yehjin Shin*, and Noseong Park, "Mitigating Over-Smoothing in Mamba2 via Spectral Domain Analysis," Workshop on Efficient Systems for Foundation Models at International Conference on Machine Learning (ES-FoMO III), 2025, *Equal Contribution

• Sheo yon Jhin, Seojin Kim, and Noseong Park, "Addressing Prediction Delays in Time Series Forecasting: A Continuous GRU Approach with Derivative Regularization," ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2024

• Seonkyu Lim*, Jaehyeon Park*, Seojin Kim*, Hyowon Wi, Haksoo Lim, Jinsung Jeon, Jeongwhan Choi, and Noseong Park, "Long-term Time Series Forecasting based on Decomposition and Neural Ordinary Differential Equations," IEEE International Conference on Big Data (IEEE BigData), 2023, *Equal Contribution

• 2024.09-Present, Graduate Student Researcher, BigDyl, (Advisor: Prof. Noseong Park)

• 2023.02-2024.08, Undergraduate Student Researcher, BigDyl, (Advisor: Prof. Noseong Park)

• 2023.12-2024.02, Data Scientist, Intern @ Minds n Company, Seoul

  • Participated as a field employee in the AI Banker project at Woori Bank, developing a chatbot solution based on LLM.

• 2021.12-2022.02, Data Analyst, Intern @ EcubeLabs, Seoul

  • Executed data-driven marketing for Haulla at the New Business Planning Division of EcubeLabs, conducting site campaigns and A/B tests to analyze results and derive improvement strategies.

  • Successfully conducted a data analysis project based on call marketing logs, resulting in performance improvement and receiving a permanent position offer as a result of this project.

• 2023.01-2024.08, Data Science Lab @ Yonsei University, Member/Academic Director (Core)

• 2021.01-2021.12, The 56th Yonsei Student Council @ Yonsei University, Member of Media and Contents Department

Awards

• 2024, 1st Semester, Honors (awarded to top 10% of the students)

• 2023 Yonsei Big Data Analysis Contest, 2nd place

• 2021 Traditional Culture Youth Startup Idea Contest, Top 10

• 2020, 1st Semester, Honors (awarded to top 10 % of the students)

Scholarship

• 2020, 1st Semester, Yonsei Jilli Scholarship