Associate Professor · AI Research Leader
Trustworthy deep learning, and evaluating AI agents honestly.
I direct the Artificial Super Intelligence (ASI) Lab at the University of West Florida, where I am an Associate Professor. I work on deep learning, large language and vision-language models, and uncertainty quantification. More recently, I study how to evaluate AI agents honestly, and what post-training and compression change inside them. Many reported gains in AI are smaller than the run-to-run noise, so I build methods that tell real improvements from chance.

- University of West FloridaAssociate Professor, Computer Science + Intelligent Systems & Robotics
- Artificial Super Intelligence LabDirector
- IHMCResearch Scientist
Research
How it fits togetherVisual & Multi-Agent AI
Agents that reason over images and video, and know when they might be wrong.
Vision-language agents · Multi-agent systems · Collective uncertainty · Hallucination detection
2 projectsEvaluation & Measurement Science
Telling real gains from chance, and finding where systems fail.
Evaluation science · Error bars & noise floors · Falsification · Uncertainty quantification
2 projectsPost-Training, Compression & Safety
What fine-tuning, distillation and quantization change inside a model.
Post-training · Distillation · Quantization · Agent safety
1 projectLatest
All news- Nov 2026
Safety in Numbers, in AI Magazine
"Safety in Numbers: Tackling the deliberative, performative, and infrastructural roots of multi-agent misalignment", led by Jeff Bradshaw at IHMC with 22 co-authors including me, is out in AI Magazine (open access).
- Nov 2026
SCoOP explained in about a minute
A short explainer of SCoOP, Johnny Yu's training-free method for measuring how uncertain a team of vision-language models is. It was a poster at the ICLR 2026 Workshop on Agentic AI in the Wild: From Hallucinations to Reliable Autonomy.
- Nov 2026
Johnny Yu's ORCA paper is published by ACM
"ORCA: An Agentic Reasoning Framework for Hallucination and Adversarial Robustness in Vision-Language Models", led by Ph.D. student Chung-En Johnny Yu, is now in the ACM proceedings of ICCBDC '26.
Work with me