Sehyun Hwang

prof_sehyun_v4.jpeg

I am a postdoctoral researcher at the Computer Vision Lab at DGIST working with Prof. Sunghoon Im. I received my Ph.D. from POSTECH, where I was advised by Prof. Suha Kwak in the Computer Vision Lab at POSTECH.

During my Ph.D., I worked on active learning, domain adaptation, and debiasing in machine learning and computer vision, with a focus on data-efficient and robust learning. My current research interests include data-centric learning and its applications to improving vision-language models (VLMs), large language models (LLMs), and embodied AI control.

I am open to research collaborations and discussions.
Feel free to reach out if you would like to discuss research-related matters.

News

Jul 22, 2025 📚 A paper about designing novel query for active learning is accepted to TMLR.
Feb 7, 2025 🎓 Successfully defended my Ph.D. dissertation and received my doctoral degree from POSTECH.
Jul 1, 2024 💼 I joined Biomedical Imaging Group at Microsoft Research Cambridge as a research intern.
May 1, 2024 💼 I joined Scalable Trustworthy AI Lab at University of Tübingen as a visiting researcher.

Education

Sep, 2018 - Feb, 2025 Pohang University of Science and Technology (POSTECH), Pohang, South Korea
Integrated M.S./Ph.D. student in Computer Science and Engineering
Advisor: Prof. Suha Kwak
Mar, 2014 - Sep, 2018 Daegu Gyeongbuk Institute of Science and Technology (DGIST), Daegu, South Korea
B.S. in School of Undergraduate Studies

Experience

Aug, 2025 - present Computer Vision Lab at DGIST, Daegu, South Korea
Postdoctoral Researcher
  • Advisor: Prof. Sunghoon Im
  • Participate in research projects about designing diffusion policy for embodied AI control.
Mar, 2025 - May, 2025 Computer Vision Lab at POSTECH, Pohang, South Korea
Postdoctoral Researcher
  • Advisor: Prof. Suha Kwak
  • Participate in research projects about scalable acive learning and efficient image tokenization.
July, 2024 - Sep, 2024 Biomedical Imaging Group at Microsoft Research Cambridge, Cambridge, UK
Research Intern
  • Mentor: Maximilian Ilse
  • Participate in research projects about tube segmentation on chest X-ray.
Mar, 2024 - May, 2024 Scalable Trustworthy AI Lab at University of Tübingen, Tübingen, Germany
Visiting Researcher
  • Mentor: Prof. Seongjoon Oh
  • Participate in research projects about active learning for OOD.
Sep, 2018 - Feb, 2025 Computer Vision Lab at POSTECH, Pohang, South Korea
Ph.D. Student
  • Advisor: Prof. Suha Kwak
  • Participate in research projects about acive learning.

Publications

* indicates equal contribution.

  1. Enhancing Cost Efficiency in Active Learning with Candidate Set Query
    Yeho Gwon*,  Sehyun Hwang*, Hoyoung Kim, Jungseul Ok,  and Suha Kwak
    Transactions on Machine Learning Research (TMLR), 2025
  2. Active Label Correction for Semantic Segmentation with Foundation Models
    Hoyoung Kim,  Sehyun Hwang, Suha Kwak,  and Jungseul Ok (*equal contribution)
    International Conference on Machine Learning (ICML), 2024
  3. Extreme Point Supervised Instance Segmentation
    Hyeonjun Lee,  Sehyun Hwang,  and Suha Kwak
    IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024
  4. Active Learning for Semantic Segmentation with Multi-class Label Query
    Sehyun Hwang, Sohyun Lee, Hoyoung Kim, Minhyeon Oh, Jungseul Ok,  and Suha Kwak
    Conference on Neural Information Processing Systems (NeurIPS), 2023
  5. Adaptive Superpixel for Active Learning in Semantic Segmentation
    Hoyoung Kim, Minhyeon Oh,  Sehyun Hwang, Suha Kwak,  and Jungseul Ok
    International Conference on Computer Vision (ICCV), 2023
  6. Combating label distribution shift for active domain adaptation
    Sehyun Hwang, Sohyun Lee, Sungyeon Kim, Jungseul Ok,  and Suha Kwak
    European Conference on Computer Vision (ECCV), 2022
    Qualcomm Innovation Fellowship Winner, Gold Prize at IPIU paper Award
  7. Learning debiased classifier with biased committee
    Nayeong Kim,  Sehyun Hwang, Sungsoo Ahn, Jaesik Park,  and Suha Kwak
    Conference on Neural Information Processing Systems (NeurIPS), 2022
  8. Learning to Detect Semantic Boundaries with Image-Level Class Labels
    Namyup Kim*,  Sehyun Hwang*,  and Suha Kwak (*equal contribution)
    International Journal of Computer Vision (IJCV), 2022
    Honorable Mention @ Samsung HumanTech Paper Award

Honors and Awards

BK21 Best Paper Award, POSTECH CSE
  • (2025) Best Paper Award - Active Label Correction for Semantic Segmentation with Foundation Models
  • (2024) Best Paper Award - Adaptive Superpixel for Active Learning in Semantic Segmentation
  • (2023) Excellence Award - Learning debiased classifier with biased committee
  • (2023) Excellence Award - Combating label distribution shift for active domain adaptation
POSTECHIAN Fellowship Award (2023)
  • Winner ($5,000)
Qualcomm Innovation Fellowship South Korea (2022)
  • Winner ($3,000) - Combating label distribution shift for active domain adaptation
IPIU Best Paper Award, IPIU, (2022)
  • Gold Prize - Combating label distribution shift for active domain adaptation
The 26th HumanTech Paper Award, Samsung Electronics Co., Ltd. (2020)
  • The Honorable Mention ($3,000) - Learning to Detect Semantic Boundaries with Image‑Level Class Labels

Professional Services

Reviewer
  • IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
  • Conference on Computer Vision and Pattern Recognition (CVPR): 2023-2025
  • International Conference on Learning Representations (ICLR): 2025
  • European Conference on Computer Vision (ECCV): 2024
  • International Conference on Machine Learning (ICML): 2024
  • Asian Conference on Computer Vision (ACCV 2022,) 2024
  • Conference on Neural Information Processing Systems (NeurIPS): 2022-2023
  • International Conference on Computer Vision (ICCV): 2023
  • Winter Conference on Applications of Computer Vision (WACV): 2022