Publications

Google Scholar carries the most current list and citation counts.

Journal articles

Scalable super-resolution of flow conveyance systems through adaptive domain decomposition

Wenzhuo Xu, Akibi Archer, Mike McCarrell, Scott Hesser, Noelia Grande Gutiérrez, Christopher McComb

Journal of Computing and Information Science in Engineering 26 (2026)

Splits a large conveyance geometry into adaptively sized subdomains so a learned super-resolution model runs on full industrial ducts rather than toy meshes.

Energy-based feature extraction with adaptive local domain decomposition for prediction of transient and turbulence flow with operator regression models

Wenzhuo Xu, Madhav Karthikeyakannan, Christopher McComb, Noelia Grande Gutiérrez

Computers & Fluids 307, 106958 (2026)

Learns which energy content actually matters in a transient or turbulent flow, so an operator-regression surrogate stays accurate where fixed-resolution surrogates drift.

Enforcing the principle of locality for physical simulations with neural operators

Jiangce Chen, Wenzhuo Xu, Zeda Xu, Noelia Grande Gutiérrez, Sneha Prabha Narra, Christopher McComb

Journal of Computational Physics 538, 114131 (2025)

Shows that restricting a neural operator to a bounded physical neighbourhood is what lets it generalize to meshes far larger than anything it was trained on.

Taylor series error correction network for super-resolution of discretized partial differential equation solutions

Wenzhuo Xu, Christopher McComb, Noelia Grande Gutiérrez

Journal of Computational Physics 521, 113569 (2025)

Learns the discretization error between a coarse and a fine solver, so a coarse mesh recovers fine-mesh accuracy at a fraction of the cost.

Capturing local temperature evolution during additive manufacturing through Fourier neural operators

Jiangce Chen, Wenzhuo Xu, Martha Baldwin, Björn Nijhuis, Ton van den Boogaard, Noelia Grande Gutiérrez, Sneha Prabha Narra, Christopher McComb

Journal of Manufacturing Science and Engineering 146(9) (2024)

Predicts how temperature evolves locally during metal additive manufacturing — the thermal history that determines whether the finished part is sound.

Conference papers

Fast super-resolution analysis of low-pressure duct air flow through adaptive domain decomposition

Wenzhuo Xu, Akibi Archer, Mike McCarrell, Scott Hesser, Noelia Grande Gutiérrez, Christopher McComb

ASME IDETC-CIE, Vol. 2A: 45th Computers and Information in Engineering Conference (2025)

The adaptive decomposition applied to Eaton's production duct geometries, cutting turnaround on a design-loop simulation from hours to minutes.

Capturing local temperature evolution during additive manufacturing through Fourier neural operators

Jiangce Chen, Wenzhuo Xu, Martha Baldwin, Björn Nijhuis, Ton van den Boogaard, Noelia Grande Gutiérrez, Sneha Prabha Narra, Christopher McComb

ASME IDETC-CIE (2023)

The conference version of the additive-manufacturing thermal-history work.

MegaFlow2D: A parametric dataset for machine learning super-resolution in computational fluid dynamics simulations

Wenzhuo Xu, Noelia Grande Gutiérrez, Christopher McComb

Proceedings of Cyber-Physical Systems and Internet of Things Week 2023 (CPS-IoT Week '23)

A parametric CFD dataset built so super-resolution methods can be compared on the same footing instead of each on its own bespoke flow.

Preprints and technical reports

Thinking with Anchors: Grounded and Efficient Document Reasoning to appear

S. Zhu, Y. Zhu, Wenzhuo Xu, J. Kuen, …, Jiuxiang Gu

arXiv preprint (2026)

Grounds document reasoning in explicit anchors in the source, so an answer can be traced back to the span it came from rather than asserted.

PhyReAct: Agentic Physical Reasoning for Video Evaluation in preparation

Adobe Research — core contributor

Technical report

Reframes video evaluation from one opaque quality score into an auditable verdict: the generation prompt is compiled into typed physical obligations and a measurement plan, frozen perception operators return only measurements, and deterministic rules compose them into supported / contradicted / unknown with full provenance.

Talks

Conference presentations

  • Graph-based domain decomposition for scalable cardiovascular flow super-resolution
    APS Division of Fluid Dynamics Meeting, Houston, TX — November 2025
  • Adaptive local domain decomposition for learning large-scale multi-physics numerical simulations
    APS Division of Fluid Dynamics Meeting, Salt Lake City, UT — November 2024
  • Taylor series error correction network for super-resolution of discretized fluid solutions
    APS Division of Fluid Dynamics Meeting, Washington, DC — November 2023
  • MegaFlow2D: a parametric dataset for machine learning super-resolution in CFD simulations
    CPS-IoT Week, San Antonio, TX — May 2023

Invited talks

  • Machine learning for large-scale multi-physics engineering simulations
    CMU Mechanical Engineering Ph.D. Research Symposium — March 2025
  • Machine learning in large-scale engineering simulations
    Autodesk Research, San Francisco, CA — November 2024