Three connected research directions
Our work connects model foundations, multimodal representation learning, embodied interaction, and real-world deployment into one agenda for rigorous, reliable, and human-aware multimodal intelligence.
Foundation Models & Reasoning
We build models that connect language, vision, video, audio, and structured signals for grounded understanding and trustworthy generation.
The focus is multimodal representation, event understanding, robust adaptation, uncertainty estimation, and evidence-backed reasoning.
Embodied Agents & World Models
We develop agents that combine perception, memory, prediction, and planning for navigation, manipulation, and dynamic interaction.
The focus is human-aware navigation, world modeling, robotic manipulation, planning, and reasoning over possible futures.
Mobility AI & Secure Deployment
We design deployable systems that integrate multimodal sensing, real-time analytics, security, edge intelligence, and accountable oversight for mobility intelligence and public safety.
The focus is traffic analytics, secure sensing, deployable perception, edge intelligence, and accountable decision support.
Research Opportunities
MILab welcomes students and collaborators interested in multimodal foundation models, embodied AI, mobility intelligence, public safety, and reliable deployment. The Join Us page describes Ph.D., postdoctoral, RA, and current UW student research-credit pathways.
Applied Work & Media
Before joining the University of Washington, I contributed to applied collaborations at Carnegie Mellon University, including DARPA KAIROS/AIDA, IARPA DIVA, NIST PSIAP, USDOT Mobility21, CMU MFI, and CMU-AIST human-inclusive navigation. Selected media coverage and public-interest applications show how this background connects to my current UW research agenda in reliable multimodal AI for real-world environments.