About
Bio, career, education, honors, and service.
I am a Ph.D. candidate in the Minnesota NLP group led by Prof. Dongyeop Kang at University of Minnesota Twin Cities.
My research interests center on developing a “meta-scaffolding paradigm” that integrates discourse structures, dataset metadata, and metacognitive feedback into the training loop of large language models to stabilize learning and produce interpretable, coherent long-form text. I design disciplined training schemes that slash data and compute demands, curtail manual prompt engineering, and steer model reasoning toward human-like cognitive patterns.
Before joining the Ph.D. program, I worked as a researcher at Naver Labs Europe and Papago team at Naver Korea, where I researched on various topics in neural machine translation (NMT), such as analysis of language-pair-specific multilingual representation, document-level NMT with discourse information, cross-attention-based website translation, and quality estimation for evaluating NMT models.
During the Ph.D. I have interned at Grammarly, Salesforce AI Research, Amazon AGI, and Amazon AWS.
I received a B.Eng. degree in Computer Science from Imperial College London in 2011. From 2012 to 2013, I served in the Republic of Korea Army Special Forces as an army interpreter and a geospatial image analyst. In 2016, I received an M.S. degree in Computer Science from Korea Advanced Institute of Science and Technology (KAIST).
During the Spring 2024 semester, I organized the meetings and seminars for Textgroup, an interdisciplinary reading group focused on language. Additionally, since October 2020, I have been leading Seeking-SOTA (Korean), a deep learning study group that convenes weekly. This group brings together researchers, academics, and professionals in South Korea, all dedicated to staying at the cutting edge of the field of deep learning. Let me know if you are interested in joining either of these groups!
My Ph.D. program is generously supported by 3M Science and Technology Fellowship and Doctoral Dissertation Fellowship.
Experience
- Developing a metric recommendation system for Amazon EC2 ASG using time-series and heterogeneous data streams.
- Proposed and developed structural alignment of LLMs for long-form generation.
- Implemented a large-scale challenging benchmarking system for scientific-paper retrieval.
- Built a co-writing system with large language models for scientific writing.
- Developed a human-in-the-loop iterative text-revision system.
- Trained span-based text-revision models with data augmentation.
- Studied language-specific representations in multilingual NMT.
- Conducted cross-lingual discourse analysis for document-level NMT.
- Developed machine translation models for Thai–Korean and German–Korean pairs.
- Implemented tools for NMT: sentence-segmenter, corpus pre-processor, aligner, and crawler.
- Created website-translation service using NMT + heuristic attention.
- Ranked 1st (doc-level) and 4th (sent-level) at WMT20 quality estimation shared task.
- Built an adaptive vocabulary learning app using learner interest profiles.
- Developed a healthcare app that recommends Pokémon Go events by forecasting users' physical activity patterns using recurrent neural network.
- Researched extending named entity recognition for Korean with hierarchical categories.
- [1st Year] Intent classification for travel-service personalization.
- [2nd Year] Modeled long-term activeness using RNNs for wellness prediction.
- Surveyed plan and intention recognition.
- Researched temporal information extraction for English and Korean.
- Served as an army interpreter, aviation image analyst, and squad leader.
Education
- Group Project: Intention Recognition in Ambient Homes
- Individual Project: Computing Fictitious Plays in Graph-Based Games