About Me
I am a second year Computer Science Ph.D. student at CLIP Lab under the supervision of Professor Marine Carpuat and Professor Tianyi Zhou at University of Maryland, College Park.
Prior to that, I received B.S. in Statistics and Computer Science from Korea University in 2023.
I was also an undergraduate researcher at the Language & Knowledge Lab at KAIST, advised by Professor Minjoon Seo and at the Data and Visual Analytics Lab at KAIST where I worked with Professor Jaegul Choo.
My primary research interest lies in natural language processing.
I am more focusing on enhancing large language model's ability to jointly reason over multilingual text and generalize to different languages and contexts.
Through my research, I hope to develop methods that can effectively align large language models with multilingual users, aiming to bridge the gap between them.
Here is Curriculum Vitae for more information about me.
Work Experience
- In Adobe Research, Document Intelligence team.
- Large Language Models and Agents.
- In NAVER Papago, Multilingual NMT team under Hyunjoong Kim.
- Disentanglement of cross-lingual sentence embedding research.
- In LG AI Research Language team advised by Sunkyoung Kim.
- Cross-lingual prompting of multilingual language models research.
Publication
Conferences
(* indicates equal contribution)
- Rewriting Inputs for Translation with Large Language Models
Dayeon Ki, Marine Carpuat
Under Review
- Mitigating Semantic Leakage in Cross-lingual Embeddings via Orthogonality Constraint
Dayeon Ki, Cheonbok Park, Hyunjoong Kim
RepL4NLP Workshop, ACL 2024
- Cross-lingual QA: A Key to Unlocking In-context Cross-lingual Performance
Sunkyoung Kim*, Dayeon Ki*, Yireun Kim, Jinsik Lee
ICL Workshop, ICML 2024
- Guiding Large Language Models to Post-Edit Machine Translation with Error Annotations
Dayeon Ki, Marine Carpuat
Findings, NAACL 2024
- Towards Accurate Translation via Semantically Appropriate Application of Lexical Constraints
Yujin Baek*, Koanho Lee*, Dayeon Ki, Hyoung-Gyu Lee, Cheonbok Park, and Jaegul Choo
Findings, ACL 2023