Taehyun Byun

Github  |  Google Scholar  |  CV

prof_pic.png

taehyun-byun@korea.ac.kr

Ph.D Student, Dept. of Artificial Intelligence

Korea University

I’m a Ph.D. student in Artificial Intelligence at Korea University, working with Professor Sungjoon Choi in the Robot Intelligence Lab. (RILAB).

My doctoral research, Toward Human-Grounded Humanoid Motion Synthesis, connects generative modeling of human motion with controllable, human-like behaviors for humanoid robots.

I received my M.S. in Statistics from Korea University in Seoul, advised by Professor Myoungshic Jhun, and my B.S. in Information and Mathematics from Korea University in Sejong.

Research Interests

I study generative models, including diffusion and flow matching, to learn human motion distributions and build controllable motion priors. Through motion synthesis and retargeting, I aim to translate these priors into natural movements that fit the embodiment and physical constraints of humanoid robots. I am also interested in human–robot interaction and in developing friendly, natural robot behaviors for everyday settings, including the home.

Generative Modeling Controllable Motion Synthesis Humanoid Motion Retargeting

News

2026 :tada: Our paper “Effective Evaluation of Exoskeleton Using Exo-Agnostic Uncertainty-Aware Classification Model with Wearable Sensors” has been accepted to ACM Augmented Humans (AHs) 2026 as a Poster!
2026 :tada: Our paper “Adaptive Hilbert Diffusion Models for Controllable Smoothness in Continuous Function Generation” has been accepted to Computational Visual Media! Check out the project page.
2026 :tada: Our paper “CLIP-Actor-X” has been accepted to IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)!

Publications

  1. arXiv
    Robot Motion Transition with Human Motion Prior
    Robot Motion Transition with Human Motion Prior
    Taehyun Byun, Taemoon Jeong, Yoonbyung Chai, Woojin Wie, Insu Park, Kyungjae Lee, and Sungjoon Choi
    arXiv preprint, 2026
    Paper Project Page
  2. PR
    Controllable Single Motion Synthesis
    Controllable Single Motion Synthesis
    Taehyun Byun, Sunwoo Kim, Minwook Chang, and Sungjoon Choi
    Pattern Recognition (IF: 7.6, 2024), 2026
    Revised and resubmitted
    Project Page
  3. CVM
    Adaptive Hilbert Diffusion Models for Controllable Smoothness in Continuous Function Generation
    Adaptive Hilbert Diffusion Models for Controllable Smoothness in Continuous Function Generation
    Taehyun Byun*, Seulbi Lee*, Kyungjae Lee, Sangheum Hwang, and Sungjoon Choi
    * Equal contribution (co-first authors)
    Computational Visual Media (IF: 18.3, 2024), 2026
    Accepted
    Project Page
  4. TPAMI
    CLIP-Actor-X: Text-driven 4D Human Avatar Generation via Cross-modal Synthesis-through-Optimization
    CLIP-Actor-X: Text-driven 4D Human Avatar Generation via Cross-modal Synthesis-through-Optimization
    Kim Youwang*, Taehyun Byun*, Kim Ji-Yeon, Sungjoon Choi, and Tae-Hyun Oh
    * Equal contribution (co-first authors)
    IEEE Transactions on Pattern Analysis and Machine Intelligence (IF: 18.6, 2024), 2026
    Paper
  5. AHs
    Effective Evaluation of Exoskeleton Using Exo-Agnostic Uncertainty-Aware Classification Model with Wearable Sensors
    Effective Evaluation of Exoskeleton Using Exo-Agnostic Uncertainty-Aware Classification Model with Wearable Sensors
    Taehyun Byun, Chanwoo Kim, Bokman Lim, Sungho Suh, and Sungjoon Choi
    ACM Augmented Humans (AHs), 2026
    Poster
    Paper
  6. IROS
    Robust and Expressive Humanoid Motion Retargeting via Optimization-Based Rig Unification
    Robust and Expressive Humanoid Motion Retargeting via Optimization-Based Rig Unification
    Taemoon Jeong, Taehyun Byun, Jihoon Kim, Keunjun Choi, Jaesung Oh, Sungpyo Lee, Omar Darwish, Joohyung Kim, and Sungjoon Choi
    In 2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2025
    Paper Project Page
  7. RA-L
    Towards Embedding Dynamic Personas in Interactive Robots: Masquerading Animated Social Kinematic (MASK)
    Towards Embedding Dynamic Personas in Interactive Robots: Masquerading Animated Social Kinematic (MASK)
    Jeongeun Park, Taemoon Jeong, Hyeonseong Kim, Taehyun Byun, Seungyoun Shin, Keunjun Choi, Jaewoon Kwon, Taeyoon Lee, Matthew Pan, and Sungjoon Choi
    IEEE Robotics and Automation Letters (IF: 4.6, 2024), 2024
    Paper
  8. NeurIPS
    Score-based generative modeling through stochastic evolution equations in hilbert spaces
    Score-based generative modeling through stochastic evolution equations in hilbert spaces
    Sungbin Lim, Eun Bi Yoon, Taehyun Byun, Taewon Kang, Seungwoo Kim, Kyungjae Lee, and Sungjoon Choi
    Advances in Neural Information Processing Systems, 2023
    Paper
  9. arXiv
    Learning Joint Representation of Human Motion and Language
    Learning Joint Representation of Human Motion and Language
    Jihoon Kim, Youngjae Yu, Seungyoun Shin, Taehyun Byun, and Sungjoon Choi
    arXiv preprint, 2022
    Paper
  10. PR
    Conditional motion in-betweening
    Conditional motion in-betweening
    Jihoon Kim, Taehyun Byun, Seungyoun Shin, Jungdam Won, and Sungjoon Choi
    Pattern Recognition (IF: 8.518, 2021), 2022
    Paper Project Page
  11. A Study of Correlation Curves Estimation
    Taehyun Byun and Myungshic Jhun
    Journal of The Korean Data Analysis Society, 2017
    Paper