Mingchen Zhuge
Hi, I am Mingchen! I am a PhD candidate with a strong interest in multimodal agents, world models, and recursive self-improvement. I am fortunate to have Prof. Jürgen Schmidhuber as my advisor. I have 20+ top-tier publications and over 4,600 citations (70% from first-author papers). My work includes:
- (1) code generation: MetaGPT
, OpenHands
- (2) swarms: NLSOM
, GPTSwarm
- (3) recursive self-improvement: GPTSwarm
, HGM
- (4) open-ended judge: agent-as-a-judge
- (5) multimodal learning, generation, reasoning: Kaleido-BERT
, Pollux
, VideoAuto-R1
My papers were accepted as oral presentations at ICLR (<1.2%), ICML (<1.5%), received the Best Paper Award at the NeurIPS Ro-FoMo Workshop, and Outstanding Paper Nomination at EMNLP 2025. I was nominated for WAIC Future Star Nomination and recognized as an Outstanding Reviewer of CVPR2023 (232 out of 7000+).
👋 I am seeking jobs, and if you are a team lead who is interested, please contact me at mingchen dot zhuge at kaust dot edu dot sa.
