I’m a research scientist at Columbia Business School, working with Prof. Hongseok Namkoong. My current work focuses on adaptive and reliable AI: foundation models and LLM agents for sequential decision-making, with clinical applications in AI for Healthcare.
View Publications ↓
Summary TEP uses local prompt optimization to mitigate exploding and vanishing textual gradients in deep, long-horizon compound AI systems.
Summary FedTextGrad studies federated LLM prompt optimization with textual gradients, identifying aggregation challenges and improving prompt summarization.
Summary Local Superior Soups improves client training and pre-trained model adaptation in federated learning through regularized model interpolation.
Summary FedSoup uses selective model interpolation to balance local and global performance, reduce overfitting, and seek flat minima for improved generalization.
Summary VITA generates diverse on-manifold samples through multi-source vicinal transfer to improve robustness to image corruptions.
Summary The first comprehensive study of five ReID corruption benchmarks compares transformer and CNN robustness and examines cross-dataset generalization.
ICLR 2027 — International Conference on Learning Representations
2022–present
2023–present
Professional experience and positions.
Decision, Risk, and Operations division at Columbia Business School, working with Prof. Hongseok Namkoong on continual, meta, and reinforcement learning.
Nanyang Technological University, working with Prof. Xiaoxiao Li and Prof. Han Yu on federated learning and multi-agent.
Vector Institute, working with Prof. Xiaoxiao Li on federated learning.
University of British Columbia, working with Prof. Xiaoxiao Li and Prof. Zehua Wang on federated learning.
I’m Minghui (pronounced: Ming-hway) Chen, a research scientist in the Decision, Risk, and Operations division at Columbia Business School, working with Prof. Hongseok Namkoong.
I work on adaptive and reliable AI, with a focus on LLM agents, continual and reinforcement learning, learning under distribution shift, and Deep Learning. I currently work on foundation models for sequential decision-making and their applications in real clinical settings, with a particular interest in AI for Healthcare.
I received my M.Sc. in Computer Science from Southern University of Science and Technology, China, and B.Sc. in Software Engineering from Sun Yat-sen University, China.
