Hi! I'm Dayoon Ko, a Ph.D. student in CSE at Seoul National university, advised by Gunhee Kim. My primary research interest lies in model evaluation for several challenges. My research has focused on retrieval-augmented generation (RAG) and multimodal learning, enhancing the scalability and adaptability of large language models (LLMs). Moving forward, I aim to apply RAG techniques to video-based LLMs, exploring how these models can integrate and process both text and video data for more effective understanding and generation.

Recent News!

Our paper "DynamicER: Resolving Emerging Mentions to Dynamic Entities for RAG" was accepted in EMNLP 2024 Main!

Publications

DynamicER: Resolving Emerging Mentions to Dynamic Entities for RAG

Jinyoung Kim, Dayoon Ko, Gunhee Kim

EMNLP 2024

This work addresses challenges in resolving temporally evolving mentions to entities. Resolving mentions is key to improve retrieval, enhancing RAG accuracy in dynamic environments.

GrowOVER: How Can LLMs Adapt to Growing Real-World Knowledge?

Dayoon Ko, Jinyoung Kim, Hahyeon Choi, Gunhee Kim

ACL 2024

We propose QA & dialogue benchmarks that are continuously and automatically updated, to assess whether LLM can handle evolving knowledge. By making LLM evaluate its confidence, we enable RAG systems to adapt to new knowledge without retraining.

Can Language Models Laugh at YouTube Short-form Videos?

Dayoon Ko, Sangho Lee, Gunhee Kim

EMNLP 2023

A video humor explanation benchmark via a multimodal-filtering pipeline to evaluate LLMs’ understanding of complex multimodal tasks like humor. We generate several frame captions and then filter them based on video segments to enhance LLMs with vision capabilities.

Education

M.S/Ph.D in Computer Science Engineering (2022.9 - )

Seoul National University

Advisor: Gunhee Kim

B.S. in Computer Science Engineering (2018.3 - 2022.8)

Yonsei University

GPA: 4.12/4.30 (Rank: 1/28)