Juhyung Lee

Sunnyvale, California, USA

profile.jpg

I am a Principal Researcher in AI/ML at Nokia (Sunnyvale, CA), specializing in architecting and deploying advanced ML models. Currently, I design On-Device AI solutions for Wireless Communication Systems.

Prior to Nokia, I served as a Postdoctoral Scholar at the University of Southern California (USC) under Prof. Andreas F. Molisch, and as a Research Professor at Korea University, where I earned my PhD advised by Prof. Young-Chai Ko.

My academic and industry research has been recognized with multiple Best Paper awards at premier IEEE conferences, a Grand Prize for research excellence, and a 1st-Rank finish in competitive ML challenges.

My research interests include:

  • On-Device AI
  • Multimodal Generative AI & Agent Modeling

news

Dec 06, 2025 Our paper, PEARL: Peer-Enhanced Adaptive Radio via On-Device LLM, has been presented at the NeurIPS’25.
Oct 13, 2025 Our work has been selected as an oral presentation at the NeurIPS’25 (AI4NextG).
Aug 25, 2025 Invited to serve on the Program Committee for the AI4NextG Workshop at NeurIPS’25.
Jul 18, 2025 Invited to join the ML4Wireless panel discussion at the ICML’25 as an industry speaker.
Jul 17, 2025 Our paper, On-Device LLM for Context-Aware Wi-Fi Roaming, has been presented at the ICML’25.
Jul 16, 2025 Our paper, AutoBS: Autonomous Base Station Deployment with Reinforcement Learning and Digital Network Twins, has been presented at the ICML’25.

selected publications

  1. overview_pearl.jpg
    PEARL: Peer-Enhanced Adaptive Radio via On-Device LLM
    Ju-Hyung Lee*, Yanqing Lu* , and Klaus Doppler
    NeurIPS, 2025
  2. overview_cho.jpg
    On-Device LLM for Context-Aware Wi-Fi Roaming
    Ju-Hyung Lee*, Yanqing Lu , and Klaus Doppler
    ICML, 2025
  3. overview_autobs.jpg
    AutoBS: Autonomous Base Station Deployment with Reinforcement Learning and Digital Network Twins
    Ju-Hyung Lee*, and Andreas F. Molisch
    ICML, 2025
  4. ondeviceai_overview.jpg
    Integrating Pre-Trained Language Model with Physical Layer Communications
    Ju-Hyung Lee*, Dong-Ho Lee , Joohan Lee , and 1 more author
    IEEE Trans. Wireless Commun., 2024
  5. pmnet_overview.jpg
    A Scalable and Generalizable Pathloss Map Prediction
    Ju-Hyung Lee*, and Andreas F. Molisch
    IEEE Trans. Wireless Commun., 2024
  6. dho_proposed.jpg
    Handover Protocol Learning for LEO Satellite Networks: Access Delay and Collision Minimization
    Ju-Hyung Lee*Andreas F. Molisch, and  et al.
    IEEE Trans. Wireless Commun., 2024