Mingchen Zhuge
Mingchen Zhuge's research focuses on scalable multimodal agent systems, including code generation, agent swarms, agentic societies and economies, recursive self-improvement, open-ended evaluation, multimodal reasoning, and neural computers.
Biography
Mingchen Zhuge is a Ph.D. candidate in the Computer Science program at King Abdullah University of Science and Technology (KAUST), working on AI agents, world models, and recursive self-improvement under the supervision of Professor Jürgen Schmidhuber. He has authored more than 20 top-tier publications with over 6,300 citations, of which more than 70% are from first-author papers. His representative projects include MetaGPT, NLSOM, GPTSwarm, Agent-as-a-Judge, Kaleido-BERT, and NeuralComputer.
His work has been recognized through six oral presentations at leading AI conferences, including the International Conference on Learning Representations (ICLR 2024, 2025, 2026; acceptance rate <1.2%) and the International Conference on Machine Learning (ICML 2024; acceptance rate <1.5%). He also received the Best Paper Award at the NeurIPS Ro-FoMo Workshop and an Outstanding Paper Nomination at the Conference on Empirical Methods in Natural Language Processing (EMNLP 2025). Beyond research, he was nominated for the WAIC Future Star Award, recognized as an Outstanding Reviewer at the Conference on Computer Vision and Pattern Recognition (CVPR 2023; 232 out of 7,000+ reviewers), served as Lead Organizer of ICLR RSI 2026, and served as an Area Chair for COLM 2026 and ACM CAIS 2026.
Awards and Distinctions
Education
- Master of Science (M.S.)
- Computer Science, China University of Geosciences (CUG), China, 2021
- Bachelor of Science (B.S.)
- Computer Science, China University of Petroleum (CUP), China, 2019
Related Media
