PBISC: Reference-free Single-cell-resolution Deconvolution of PBMC Bulk RNA-seq
Under Review.
Research
I build AI systems that connect genomic prediction with biological mechanisms and clinically meaningful insight. The goal is to produce robust models whose explanations can guide experiments and generate testable hypotheses.
Regulatory genomics
Foundation models
Interpretability
Robustness
Evaluation
Research record
Publications & ongoing work
* denotes equal contribution. “Under Review” items are ongoing manuscripts.
BioAI & AI for Science
RNA-LLM: Transcriptomes as One Language-model Token
Under Review.
Leveraging Biokinetic Knowledge Priors for Data-Scarce Bioprocess Modeling
ICML 2026 AI for Science Workshop.
Towards the Practical Utility of Federated Learning in the Medical Domain [code]
CHIL 2023.
EHRSQL: A Practical Text-to-SQL Benchmark for Electronic Health Records [code]
NeurIPS 2022 Datasets and Benchmarks.
Language Models
Assessing Patent Value with LLMs: DRAM Dataset [pdf]
PACIS 2025.
Predictive Pipelined Decoding: A Compute-Latency Trade-off for Exact LLM Decoding
TMLR 2024.
Improving Lexically Constrained Neural Machine Translation with Source-Conditioned Masked Span Prediction [arXiv] [code]
ACL-IJCNLP 2021 (Short Papers).
Robust Machine Learning
Task Agnostic and Post-hoc Unseen Distribution Detection [code]
WACV 2023.
For citation details and the latest updates, visit Google Scholar.
Experience & education
Background
-
Present
PhD Student · Cold Spring Harbor Laboratory
Advised by Prof. Peter Koo; researching interpretable AI for genomics.
-
2025.12 – 2026.05
AI Researcher · Gruve AI
Developed AI agents in collaboration with a genetic testing company at a U.S.-based AI company.
-
Previously
NLP Research Engineer · KRAFTON AI
Worked on LLM post-training, prompting, efficient inference, and applications for games.
-
Earlier
AI Researcher · NHN Cloud
Designed tutorials and evaluation workflows for Korean language models.
-
2020 – 2022
M.S. in AI · KAIST
Advised by Prof. Edward Choi; studied federated learning, healthcare AI, distribution shifts, and NLP.
-
Undergraduate
B.S. in Computer Science · Yonsei University
Worked on language and information-processing research, including an industry–academia project with Hyundai NGV.