Dohyeon Lee
I am a Ph.D. Student at the LDI Lab of Seoul National University.
I am fortunate to be advised by Prof. Seung-won Hwang.
My research interests focus on information retrieval and hybrid approaches.
Research Area
Representation-level Hybrid Retrieval
This approach aims to combine the precise lexical matching capability inherent to sparse models with the rich semantic representation strength of dense models.
Distribution-level Hybrid Retrieval
This approach aims to integrate both corpus-wide inverse document frequency (idf) distributions and document-level term distributions into the retrieval training process.
Term-level Hybrid Retrieval
This approach aims to augment pseudo-queries by incorporating terms previously unseen during training or even within the target corpus, thereby reducing the mismatch between pseudo-queries and real-world queries by expanding effective vocabulary coverage.
Agent-level Hybrid Retrieval
This approach consists of multiple specialized yet closely coordinated agents: a retriever, a rewriter, and a reranker. Each agent specializes in a specific aspect of the retrieval process, with coordination facilitated through iterative interactions and state-sharing mechanisms. Through this iterative cooperation, these agents progressively enhance retrieval effectiveness and relevance.
Publications
tRAG: Term-level Retrieval-Augmented Generation for Zero-shot Retrieval
NAACL 2025
DADA: Distribution-Aware Domain Adaptation of PLMs for Information Retrieval
Findings of ACL 2024
HIL: Hybrid Isotropy Learning for Zero-shot Performance in Dense retrieval
NAACL 2024
Interventional Speech Noise Injection for ASR Generalizable Spoken Language Understanding
EMNLP 2024
C2LIR: Continual Cross-lingual Transfer for Low-Resource Information Retrieval
ECIR 2023 (short)
Script, Language, and Labels: Overcoming Three Discrepancies for Low-Resource Language Specialization
AAAI 2023
SCOPA - Soft Code-Switching and Pairwise Alignment for Zero-Shot Cross-lingual Transfer
CIKM 2021 (short)
Educaiton
Seoul National University - Ph.D. in Computer Science and Engineering
2021 - (In progress)Advisor: Prof. Seung-won Hwang
Doctoral disseration: (TBD)
Yonsei University - M.S. in Computer Science
2019 - 2021Advisor: Prof. Seung-won Hwang
Master's thesis: Orthogonal Disentanglement of Semantic and Symbolic Representation for Query-Document Matching
Yonsei University - B.S. in Computer Science
2015 - 2019GPA: 4.06 / 4.5
Awards & Scholarships
BK21+ Outstanding Research Fellowship
2023Korean Government Scholarship Program
Computer Science Department Scholarship
2019 - 2021at Yonsei University (graduate school)
Computer Science Department Scholarship
2017at Yonsei University (undergraduate)
Experience
NAVER Corp. - Research Intern
2020.3 - 2020.6Research title: Sparse-Dense Hybrid Retrieval
Mentor: Ph.D. Sunghyun Park