Members

Member

Byeonghu NA

PostDocs

Applied Artificial Intelligence Laboratory Industrial and Systems Engineering, KAIST 291 Daehak-ro, Yuseong-gu, Daejeon 305-338, Republic of Korea

KAIST E4, C325

byeonghu.na@kaist.ac.kr

https://sites.google.com/view/byeonghu-na

Research Interest

  • Deep Generative Model: Diffusion Model, Variational Autoencoder, Large Language Models
  • Insufficient/incomplete Dataset: Noisy Label Learning, Positive-Unlabeled Learning, Distribution Shift

Education

    Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea (Mar. 2025 - )

  • Postdoctoral Researcher in KAIST Institute for Robotics
  • Advisor: Professor Il-Chul Moon

    Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea (Mar. 2021 - Feb. 2025)

  • Ph.D. in Industrial and Systems Engineering, AAILab
  • Advisor: Professor Il-Chul Moon
  • Thesis: Improving Conditional Information Estimation for Diffusion-based Generative Models

    Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea (Feb. 2019 - Feb. 2021)

  • M.S. in Industrial and Systems Engineering, AAILab
  • Advisor: Professor Il-Chul Moon
  • Thesis: Deep Generative Positive-Unlabeled Learning under Selection Bias

    Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea (Mar. 2014 - Feb. 2019)​

  • B.S. in Mathematical Sciences
  • B.S. in Industrial and Systems Engineering

Publication

    International Conference

  • Byeonghu Na, Mina Kang, Jiseok Kwak, Minsang Park, Jiwoo Shin, SeJoon Jun, Gayoung Lee, Jin-Hwa Kim, and Il-Chul Moon. 2025. Diffusion Adaptive Text Embedding for Text-to-Image Diffusion Models. In The Thirty-ninth Annual Conference on Neural Information Processing Systems (NeurIPS 2025), December 2–7, 2025, San Diego, USA.
  • Byeonghu Na, Minsang Park, Gyuwon Sim, Donghyeok Shin, HeeSun Bae, Mina Kang, Se Jung Kwon, Wanmo Kang, and Il-Chul Moon. 2025. Diffusion Adaptive Text Embedding for Text-to-Image Diffusion Models. In The Thirty-ninth Annual Conference on Neural Information Processing Systems (NeurIPS 2025), December 2–7, 2025, San Diego, USA.
  • Yeongmin Kim, HeeSun Bae, Byeonghu Na, and Il-Chul Moon. 2025. Preference Optimization by Estimating the Ratio of the Data Distribution. In The Thirty-ninth Annual Conference on Neural Information Processing Systems (NeurIPS 2025), December 2–7, 2025, San Diego, USA.
  • Yeongmin Kim, Kwanghyeon Lee, Minsang Park, Byeonghu Na, and Il-Chul Moon. 2025. Diffusion Bridge AutoEncoders for Unsupervised Representation Learning. In The Thirteenth International Conference on Learning Representations (ICLR 2025), April 24-28, 2025, Singapore, Singapore.
  • Byeonghu Na*, Suhyeon Jo*, Yeongmin Kim, and Il-Chul Moon. 2024. Reward-based Input Construction for Cross-document Relation Extraction. In The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024), August 11–16, 2024, Bangkok, Thailand. (Oral presentation)
  • Kwanghyeon Lee, Mina Kang, Hyungho Na, Heesun Bae, Byeonghu Na, Doyun Kwon, Seungjae Shin, Yeongmin Kim, Taewoo Kim, Seungmin Yun, and Il-Chul Moon. 2024. DPO-Finetuned Large Multi-Modal Planner with Retrieval-Augmented Generation @ EgoPlan Challenge ICML 2024. In The Multi-modal Foundation Model meets Embodied AI Workshop @ ICML 2024, July 26, 2024, Vienna, Austria.
  • Byeonghu Na, Yeongmin Kim, Minsang Park, Donghyeok Shin, Wanmo Kang, and Il-Chul Moon. 2024. Diffusion Rejection Sampling. In The Forty-first International Conference on Machine Learning (ICML 2024), July 21–27, 2024, Vienna, Austria.
  • Byeonghu Na, Yeongmin Kim, HeeSun Bae, Jung Hyun Lee, Se Jung Kwon, Wanmo Kang, Il-Chul Moon, Label-Noise Robust Diffusion Models, International Conference on Learning Representations (ICLR 2024), Vienna, Austria, May 7-11, 2024
  • Yeongmin Kim, Byeonghu Na, JoonHo Jang, Minsang Park, Dongjun Kim, Wanmo Kang, Il-Chul Moon, Training Unbiased Diffusion Models From Biased Dataset, International Conference on Learning Representations (ICLR 2024), Vienna, Austria, May 7-11, 2024 
  • HeeSun Bae, Seungjae Shin, Byeonghu Na, Il-chul Moon, Dirichlet-based Per-Sample Weighting by Transition Matrix for Noisy Label Learning, International Conference on Learning Representations (ICLR 2024), Vienna, Austria, May 7-11, 2024 
  • Seungjae Shin, Heesun Bae, Byeonghu Na, Yoon-Yeong Kim, Il-Chul Moon, Unknown Domain Inconsistency Minimization for Domain Generalization, International Conference on Learning Representations (ICLR 2024), Vienna, Austria, May 7-11, 2024 
  • Byeonghu Na, Hyemi Kim, Kyungwoo Song, Weonyoung Joo, Yoon-Yeong Kim, and Il-Chul Moon. 2020. Deep Generative Positive-Unlabeled Learning under Selection Bias. In The 29th ACM International Conference on Information and Knowledge Management (CIKM ’20), October 19–23, 2020, Virtual Event, Ireland.
  • HeeSun Bae*, Seungjae Shin*, Byeonghu Na, JoonHo Jang, Kyungwoo Song, and Il-Chul Moon. 2022. From Noisy Prediction to True Label: Noisy Prediction Calibration via Generative Model. In The 39th International Conference on Machine Learning (ICML 2022), July 17-23, 2022, Baltimore, Maryland, USA.
  • Seungjae Shin, Byeonghu Na, HeeSun Bae, JoonHo Jang, Hyemi Kim, Kyungwoo Song, Youngjae Cho, and Il-Chul Moon. 2022. Improving Group-based Robustness and Calibration via Ordered Risk and Confidence Regularization. In The Workshop on Spurious Correlations, Invariance, and Stability, International Conference on Machine Learning (SCIS at ICML 2022), July 22, 2022, Baltimore, Maryland, USA. 
  • Byeonghu Na, Yoonsik Kim, and Sungrae Park. 2022. Multi-modal Text Recognition Networks: Interactive Enhancements between Visual and Semantic Features. In European Conference on Computer Vision (ECCV 2022), October 23-27, 2022, Tel-Aviv, Israel. 
  • JoonHo Jang, Byeonghu Na, DongHyeok Shin, Mingi Ji, Kyungwoo Song, and Il-Chull Moon. 2022. Unknown-Aware Domain Adversarial Learning for Open-Set Domain Adaptation. In The Conference on Neural Information Processing Systems (NeurIPS 2022), Nov 28-Dec 9, 2022, New Orleans, USA. 
  • Dongjoun Kim*, Byeonghu Na*, Se Jung Kwon, Dongsoo Lee, Wanmo Kang, and Il-Chul Moon. 2022. Maximum Likelihood Training of Implicit Nonlinear Diffusion Models. In The Conference on Neural Information Processing Systems (NeurIPS 2022), Nov 28-Dec 9, 2022, New Orleans, USA. 
  • Yoon-Yeong Kim*, Youngjae Cho*, JoonHo Jang, Byeonghu Na, Yeongmin Kim, Kyungwoo Song, Wanmo Kang, Il-Chul Moon, SAAL: Sharpness-Aware Active Learning, In The 40th International Conference on Machine Learning (ICML 2023), Jul 25-27, 2023, Hawaii, USA. 
  • Suhyeon Jo, DongHyeok Shin, Byeonghu Na, JoonHo Jang, and Il-Chul Moon. 2023. Hierarchical Multi-Label Classification with Partial Labels and Unknown Hierarchy. In The 32nd ACM International Conference on Information and Knowledge Management (CIKM 2023), October 21–25, 2023, Birmingham, United Kingdom.

    Domestic Conference

  • 신승재, 나병후, 신동혁, 나영연, 문일철 (2017. 12) 인공 신경망 기반 뉴스기사의 토픽 분석 및 문서 분류 동시 수행 모델 개발, 한국정보과학회

Award

  • Outstading Reviewers, CVPR 2025.
  • Best PhD Dissertation Award, KAIST College of Engineering, 2025.
  • Winners, Qualcomm Innovation Fellowship Korea, 2024.
  • Outstanding Champion & Innovation Award, EgoPlan Challenge @ ICML 2024 Workshop, 2024.
  • Best Poster, Samsung AI Forum, 2022, 2024.
  • Top Reviewer, NeurIPS, 2022, 2023.
  • 관정 국내대학원 장학생, 관정이종환장학재단, 2021 - 2022.
  • SIGIR Student Travel Grant, SIGIR, 2020.
  • 김영한 글로벌리더 장학생, KAIST, 2020.
  • Summa Cum Laude, KAIST, 2019.
  • 학부생부문 최우수상, KSC2017 학부생 논문경진대회, 2017.
  • 이공계 국가우수장학금, 한국장학재단, 2014 - 2017.
  • 고교 상상 장학생, KT&G 장학재단, 2012 - 2013.