Google Scholar | LinkedIn | Github

I received my PhD from the Graduate School of AI, KAIST, in 2023 (Best PhD Thesis Award, School of Engineering), under the supervision of Prof. Sung Ju Hwang, and my Bachelor's degrees ****in Computer Science (Honors Program) and Applied Mathematics & Statistics from Stony Brook University.

My research interests span — but are not limited to — Multi-Agent System, Automated Machine Learning, Efficient On-Device AI, and Federated & Continual Learning. Integrating and extending these directions, I am designing a Collectively Evolving Federated Foundation System for Physical AI: a collective intelligence whose heterogeneous clients solve problems together, share runtime experience beyond modality, and self-evolve beyond what pretraining alone achieves.

Honors & Awards

• Best PhD Thesis Award, College of Engineering, KAIST, 2024
• Best Student Paper Award, ICML Workshop on Federated Learning (FL-ICML 2020), 2020
• Excellence Graduation Award, Stony Brook University, 2019
• Summa Cum Laude, Stony Brook University, 2019
• Outstanding Academic Achievement Award, Stony Brook University, 2016–2018
• URECA Research Fellowship, Stony Brook University, 2017

Publications

• equal contribution

• Multi-View Encoders for Performance Prediction in LLM-Based Agentic Workflows, ICLR 2026 P. Trirat, Wonyong Jeong, and S. J. Hwang
• AutoML-Agent: A Multi-Agent LLM Framework for Full-Pipeline AutoML, ICML 2025 P. Trirat, Wonyong Jeong, and S. J. Hwang
• Diffusion-Based Neural Network Weights Generation, ICLR 2025 B. Soro, B. Andreis, H. Lee, Wonyong Jeong, S. Chong, F. Hutter, and S. J. Hwang
• A Training-Free Sub-quadratic Cost Transformer Model Serving Framework with Hierarchically Pruned Attention, ICLR 2025 H. Lee, G. Park, Y. Lee, J. Suh, J. Kim, Wonyong Jeong, B. Kim, H. Lee, M. Jeon, and S. J. Hwang
• Personalized Subgraph Federated Learning, ICML 2023 [paper] **J. Baek, Wonyong Jeong, J. Jin, J. Yoon, S. J. Hwang
• Factorized-FL: Personalized Federated Learning with Parameter Factorization & Similarity Matching, NeurIPS 2022 [paper] Wonyong Jeong and S. J. Hwang