Discovery and Understanding with AI Lab
Forward ModelingandInverse Design:Quantifying structure, modeling dynamics, and guiding design across scales.AI for ScienceandScience for AI:Understanding the world with AI, and understanding AI itself.
We study how multiscale structure and multiscale dynamics shape function in physical systems and AI models. We develop mathematical and AI methods for prediction and use this understanding to guide our growing work in inverse design.
Xiongye Xiao is the UTK Principal Investigator for COMPASS (Center for Complex Particle Systems), an NSF Science and Technology Center.
Includes the PI's earlier work with collaborators at USC and other institutions.
Analyze and predict evolving fields, coupled dynamics, and systems with memory.
GeoWind2Plan connects local, time-averaged 3D wind prediction to UAV path and speed planning, with modeled mission energy evaluated under high-fidelity CFD wind.
The video could not be loaded.
Open videoScaleSplit-NO conditions local, full-resolution patch operators on a global coarse forecast for memory- and data-efficient 3D turbulence prediction.
ASPIRE predicts multiple saddle-point candidates from one atomic configuration, then refines them with Dimer searches to discover transition events for thermally activated dynamics.
CMWNO learns coupled solution operators through interacting kernels in multiwavelet space.
MWT learns PDE solution operators by representing kernels in multiwavelet bases across scales.
Quantify network organization and heterogeneity across scales, and investigate their relation to function.
NeuroMFA characterizes the multifractal organization of AI weight networks across training checkpoints.
Multifractal spiking statistics connect neuronal dynamics to network structure and function.
NMFA uses node-centered box growth to characterize multiscale network structure and compare its complexity and heterogeneity.
From structure-informed materials selection and neuro-inspired AI models toward physics-constrained inverse design.
MolP2SBench tests whether LLMs use molecular properties to select missing structures, and whether their choices follow changes in the supplied physical evidence.
ITHP combines a neuro-inspired processing hierarchy with information bottlenecks for multimodal learning.
Graph and multifractal analysis compare nanofiber networks for structural-battery selection; physical tests assess mechanical performance and ion transport.
Congratulations to James on his selection for a UTK Scholarly Development Grant - Research Assistantship (SDG-RA)!
GeoWind2Plan has been accepted as a Spotlight at NeurIPS 2026. Congratulations to Shaoxiang, Yucheng, and our collaborators!
Our collaborative review of complex nanoparticle systems, from synthesis and self-assembly to graph theory and applications, is published in Chemical Reviews.
Toward a Structure-to-Function Science of Foundation Models is accepted as a visionary paper at the ACM AI Leadership Summit 2026. Congratulations to Yucheng and Shaoxiang!
AI Tennessee supports our research on AI for materials.
Our group receives support from NSF COMPASS, with Xiongye Xiao leading the research at UTK as Principal Investigator.
Principal Investigator
PhD Student
PhD Student
PhD Student
PhD Student (co-advised by Dr. Fei Liu)
Visiting Scholar; PhD Student at McGill University
Undergraduate Research Assistant
Undergraduate Research Assistant
Undergraduate Researcher
We welcome students and collaborators interested in AI methods for scientific systems and in the study and design of AI models.
U.S. National Science FoundationThrough NSF STC COMPASS (Center for Complex Particle Systems)
AI TennesseeUniversity of Tennessee, KnoxvilleXiongye Xiao is the UTK Principal Investigator for NSF STC COMPASS and a member of UTK's Foundational AI cluster.
We gratefully acknowledge support from AI Tennessee. Our COMPASS research is supported by the U.S. National Science Foundation under Award No. 2243104. Any opinions, findings, and conclusions or recommendations expressed here are those of the authors and do not necessarily reflect the views of the U.S. National Science Foundation.
