About Me
👋 Hi! I’m Kiyoung Om, a research intern at NAVER LABS. I received my M.S. in Data Science from the Graduate School of Data Science at the Korea Advanced Institute of Science and Technology (KAIST), where I worked in the System Intelligence Lab under the supervision of Professor Jinkyoo Park. Here’s my CV.
My current research focuses on the closed-loop covariate shift problem in traffic simulation for autonomous driving. I am interested in building learning and simulation methods that remain reliable when autonomous vehicle policies interact with evolving traffic environments.
More broadly, my research bridges generative models (diffusion models and flow-based models) with sequential decision-making (reinforcement learning and black-box optimization). I develop algorithms that learn complex probability distributions and use them to make robust decisions under uncertainty.
I am also interested in the alignment of generative models, including diffusion models and LLMs, with downstream objectives. My work explores search and fine-tuning strategies that guide generative models toward reliable, goal-oriented behavior while preserving diversity.
🔥 News
- 🚗 June 2026: Working as a research intern at NAVER LABS on closed-loop covariate shift in traffic simulation for autonomous driving
- 📄 January 2026: Two papers accepted to ICLR 2026
- 🎓 February 2026: Completed my M.S. in Data Science at KAIST
- 📄 Sep 2025: One paper accepted at the NeurIPS 2025 workshop on SPIGM (Oral)
- 📄 May 2025: One paper accepted at the International Conference on Machine Learning (ICML) 2025
- 🎓 March 2024: Joined the System Intelligence Lab as a Master’s student at KAIST
🎓 Education
KAIST M.S. in Data Science (2024 – 2026) - Advisor: Prof. Jinkyoo Park
- Lab: System Intelligence Lab
Korea University B.S. in Industrial Management Engineering (2018 – 2024)