pith. machine review for the scientific record. sign in

arxiv: 2605.12116 · v1 · submitted 2026-05-12 · ⚛️ physics.plasm-ph

Recognition: 2 theorem links

· Lean Theorem

MPEX AI Digital Twins Milestone Report

Atul Kumar, Ben Dudson, Cory Hauck, Gary Staebler, Gregory Watson, John Duggan, Mark Cianciosa, Minglei Yang, Pablo Seleson, Rhea Barnett, Richard Archibald, Rinkle Juneja, Sam Reeve, Vasily Geyko, Viktor Reshniak, Wouter Tierens

Pith reviewed 2026-05-13 03:28 UTC · model grok-4.3

classification ⚛️ physics.plasm-ph
keywords AI digital twinsMPEXhelicon heatingelectron beam damageGalaxy interfaceplasma operationsfusion energydigital twin prototypes
0
0 comments X

The pith

MPEX project reports two AI digital twin prototypes on track for June 2026 demonstration.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This six-month progress report states that the MPEX AI Digital Twins effort has advanced two phase I milestones to the point of readiness for a June demonstration. The first milestone is a Helicon AI Hot-Spot Controller that manages the helicon heating component of the planned full digital twin. The second is a reduced-scale E-beam Damage Assessment Digital Twin that prototypes material damage prediction. The report also describes configuration of the Galaxy software interface that now links selected physics simulation codes to DOE high-performance computers and will link to the MPEX data acquisition system for both human and AI-driven analysis and validation.

Core claim

The authors state that the Helicon AI Hot-Spot Controller and the E-beam Damage Assessment Digital Twin prototypes are progressing on schedule, supported by the Galaxy interface that connects simulations to DOE HPC resources and the experiment data system, thereby positioning the project to demonstrate AI advantages for MPEX operations and discovery by June 2026.

What carries the argument

The two phase I prototypes—the Helicon AI Hot-Spot Controller and E-beam Damage Assessment Digital Twin—together with the Galaxy software interface that automates simulation execution, validation, and data analysis.

If this is right

  • The Galaxy interface will allow scientists or AI agents to run and validate MPEX simulations on DOE HPC resources.
  • Once connected to the American Science Cloud, Galaxy will serve as the MPEX data and simulation gateway.
  • Successful hot-spot control will enable predictive adjustment of helicon heating during plasma experiments.
  • The damage-assessment twin will provide early estimates of electron-beam effects on target materials.
  • These prototypes form the foundation for the full MPEX AI Hot Spot and Damage Assessment Digital Twins.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the milestones succeed, similar AI twin methods could be tested on other plasma devices to reduce trial-and-error operation.
  • The current report leaves open whether the AI components deliver faster or more accurate results than existing control methods.
  • Integration with cloud resources may allow real-time AI agents to adjust experiment parameters during shots.
  • The approach could later extend to additional MPEX diagnostics such as probe arrays or spectroscopy.

Load-bearing premise

That the unreported technical details in the milestone sections actually establish operational readiness and measurable AI advantage without any validation data, performance metrics, or error analysis provided.

What would settle it

Absence of a completed demonstration by June 2026 or release of quantitative metrics showing the AI components outperform non-AI controls on the same MPEX tasks.

Figures

Figures reproduced from arXiv: 2605.12116 by Atul Kumar, Ben Dudson, Cory Hauck, Gary Staebler, Gregory Watson, John Duggan, Mark Cianciosa, Minglei Yang, Pablo Seleson, Rhea Barnett, Richard Archibald, Rinkle Juneja, Sam Reeve, Vasily Geyko, Viktor Reshniak, Wouter Tierens.

Figure 1
Figure 1. Figure 1: Drawing of the Material Plasma Exposure eXperiment (MPEX) showing the Helicon [PITH_FULL_IMAGE:figures/full_fig_p008_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Proto-MPEX experimental observations of helicon-window heat flux. (a) Magnetic topology [PITH_FULL_IMAGE:figures/full_fig_p011_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Proto-MPEX target heat-flux evolution from the same experimental shot as Fig. [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: COMSOL RF heating simulations for Proto-MPEX. The 2D model shows di [PITH_FULL_IMAGE:figures/full_fig_p012_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: HERMES-3 time-averaged plasma profiles at a representative axial location. Left: plasma [PITH_FULL_IMAGE:figures/full_fig_p014_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Flux-tube heat-flux mapping from Proto-MPEX measurements. Left: measured IR heat-flux [PITH_FULL_IMAGE:figures/full_fig_p015_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Comparison of IR measurements at the Helicon window with AI-generated images for the [PITH_FULL_IMAGE:figures/full_fig_p016_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Depiction of a VAE used to predict experimental data that were not measured. The top [PITH_FULL_IMAGE:figures/full_fig_p017_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Left: VAE-predicted heat flux at the Helicon window for coil currents [PITH_FULL_IMAGE:figures/full_fig_p017_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Left: VAE-predicted heat flux at the Helicon window for coil currents [PITH_FULL_IMAGE:figures/full_fig_p018_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Workflow connections for MPEX AI Damage Assessment, including experimental inputs, [PITH_FULL_IMAGE:figures/full_fig_p020_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Overview of the collected surface-scan dataset. (a) Number of experiments and images [PITH_FULL_IMAGE:figures/full_fig_p021_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: Image-processing pipeline for crack detection. Top row: original image and successive [PITH_FULL_IMAGE:figures/full_fig_p022_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: Top row: Experimental characterization for a single heat flux and base temperature [PITH_FULL_IMAGE:figures/full_fig_p023_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: Top row: Representative SEM images at di [PITH_FULL_IMAGE:figures/full_fig_p024_15.png] view at source ↗
Figure 16
Figure 16. Figure 16: Embedding pipeline for DINOv2 patch features. From left to right: PCA variance decay, [PITH_FULL_IMAGE:figures/full_fig_p025_16.png] view at source ↗
Figure 17
Figure 17. Figure 17: Patch embeddings in diffusion coordinate space colored by base temperature, heat flux, and crack density, which exhibit clear spatial correlation in the embedding. However, PCA is inherently a linear method and may not capture the intrinsic structure of complex image representations. In particular, features derived from natural images and scientific imaging data often lie on a low-dimensional nonlinear ma… view at source ↗
Figure 18
Figure 18. Figure 18: Parity plots of predicted versus observed crack density for kernel ridge regression using [PITH_FULL_IMAGE:figures/full_fig_p028_18.png] view at source ↗
Figure 19
Figure 19. Figure 19: Top row: Experimental test condition matrix across microstructure, alloy, base temperature, [PITH_FULL_IMAGE:figures/full_fig_p029_19.png] view at source ↗
Figure 20
Figure 20. Figure 20: Experimental stress-strain data with model parameterizations for longitudinal and trans [PITH_FULL_IMAGE:figures/full_fig_p031_20.png] view at source ↗
Figure 21
Figure 21. Figure 21: Anisotropic wave propagation for the transversely isotropic forged tungsten microstructures [PITH_FULL_IMAGE:figures/full_fig_p032_21.png] view at source ↗
Figure 22
Figure 22. Figure 22: Stress-strain curves for ultra-high purity tungsten in the transverse and longitudinal [PITH_FULL_IMAGE:figures/full_fig_p033_22.png] view at source ↗
Figure 23
Figure 23. Figure 23: E-beam physics predictions with (a) imposed temperature profiles for CabanaPD matching [PITH_FULL_IMAGE:figures/full_fig_p034_23.png] view at source ↗
Figure 24
Figure 24. Figure 24: An example of a tool output that can be visualized directly in Galaxy. Here we visualize a [PITH_FULL_IMAGE:figures/full_fig_p036_24.png] view at source ↗
Figure 25
Figure 25. Figure 25: An example of the Galaxy workflow builder. [PITH_FULL_IMAGE:figures/full_fig_p036_25.png] view at source ↗
read the original abstract

This is the six month progress report to Fusion Energy Science (FES) and the American Science Cloud (AmSC) on the MPEX AI Digtial Twins project that was started in October 2025. There are two milestones to demonstrate the Artificial Intelligence (AI) advantage for MPEX operations and scientific discovery, that will be completed by June 2026. The first is a Helicon AI Hot-Spot Controller (Sec. 3.1), which is the helicon heating component of the more comprehensive planned MPEX AI Hot Spot Digital Twin (Sec. 3). The second is an E-beam Damage Assessment Digital Twin (Sec. 4.1), which is a reduced electron beam damage modality prototype for the MPEX AI Damage Assessment Digital Twin (Sec. 4). These two phase I milestones are on track for the June demonstration. In addition to these two milestones, progress on configuring the Galaxy software interface for automation, validation and data analysis is reported (Sec. 5). This interface now connects a subset of the main physics simulation codes to DOE HPC resources and will connect to the MPEX data acquisition system so that analysis of data, validation and execution of simulations can be performed by the scientist or by AI-Agents. When AmSC is ready to accept connections and data, Galaxy will be the MPEX interface to AmSC

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 2 minor

Summary. This manuscript is a six-month progress report on the MPEX AI Digital Twins project started in October 2025. It asserts that two Phase I milestones—the Helicon AI Hot-Spot Controller (Sec. 3.1) and the E-beam Damage Assessment Digital Twin (Sec. 4.1)—are on track for demonstration by June 2026, and reports progress on configuring the Galaxy software interface (Sec. 5) to connect physics simulation codes to DOE HPC resources and the MPEX data acquisition system for automated analysis and AI-agent execution.

Significance. If the milestones are successfully demonstrated with the claimed AI advantage, the work would mark incremental progress toward AI-enabled digital twins for controlling helicon heating and assessing electron-beam damage in plasma-material interaction experiments. The Galaxy interface development could enable reproducible workflows linking simulations and data, supporting both operations and discovery in fusion research.

major comments (1)
  1. [Sec. 3.1, Sec. 4.1] Sec. 3.1 and Sec. 4.1: The central claim that the two milestones are 'on track' for the June demonstration is not supported by any performance metrics, validation results against MPEX data, error analyses, or quantitative comparisons to non-AI baselines. The sections contain only declarative statements of progress and high-level plans, leaving the assertion unevaluable from the reported evidence.
minor comments (2)
  1. [Abstract] Abstract: 'Digtial' is a typographical error and should read 'Digital'.
  2. The manuscript lacks any tables, figures, or numerical results that would normally accompany milestone claims in a technical report; adding at least preliminary benchmarks would improve clarity.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their thoughtful review of our six-month progress report on the MPEX AI Digital Twins project. We provide a point-by-point response to the major comment below.

read point-by-point responses

  1. Referee: [Sec. 3.1, Sec. 4.1] Sec. 3.1 and Sec. 4.1: The central claim that the two milestones are 'on track' for the June demonstration is not supported by any performance metrics, validation results against MPEX data, error analyses, or quantitative comparisons to non-AI baselines. The sections contain only declarative statements of progress and high-level plans, leaving the assertion unevaluable from the reported evidence.

    Authors: We agree that the current version of the manuscript does not include specific performance metrics or validation data, as these are part of the ongoing development toward the June 2026 demonstration. The report focuses on the progress achieved in the first six months, including the configuration of the Galaxy interface and initial setups for the AI controllers. The 'on track' assessment is based on the successful completion of these preparatory steps and alignment with the project timeline. To strengthen the manuscript, we will make a partial revision by adding explicit statements in Sections 3.1 and 4.1 clarifying that this is an interim qualitative evaluation and outlining the next steps for quantitative validation. We believe this addresses the evaluability concern while accurately reflecting the project's status. revision: partial

Circularity Check

0 steps flagged

No circularity: factual status report with no derivations or predictions

full rationale

This is a six-month progress report asserting that two AI milestones are on track for June demonstration. The text contains no equations, no first-principles derivations, no fitted parameters, no predictions of physical quantities, and no load-bearing self-citations. The central statements are declarative progress summaries (e.g., “These two phase I milestones are on track”) rather than results obtained by reducing one quantity to another via the paper’s own logic. Consequently no step satisfies any of the enumerated circularity patterns, and the circularity score is zero.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The report introduces no mathematical models, physical derivations, or new entities; it contains no free parameters, axioms, or invented entities that the central claim depends upon.

pith-pipeline@v0.9.0 · 5608 in / 1157 out tokens · 141180 ms · 2026-05-13T03:28:02.540704+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

Reference graph

Works this paper leans on

37 extracted references · 37 canonical work pages · 1 internal anchor

  1. [1]

    Laplacian eigenmaps for dimensionality reduction and data representation

    Mikhail Belkin and Partha Niyogi. “Laplacian eigenmaps for dimensionality reduction and data representation”. In:Neural Computation15.6 (2003), pp. 1373–1396

    work page 2003

  2. [2]

    DINOv2 rocks geological image analysis: Classification, segmentation, and interpretability

    Florent Brondolo and Samuel Beaussant. “DINOv2 rocks geological image analysis: Classification, segmentation, and interpretability”. In:Journal of Rock Mechanics and Geotechnical Engineering 17.11 (2025), pp. 6853–6867.doi:10.1016/j.jrmge.2025.01.057

    work page doi:10.1016/j.jrmge.2025.01.057 2025

  3. [3]

    Emerging Properties in Self-Supervised Vision Transformers

    Mathilde Caron et al. “Emerging Properties in Self-Supervised Vision Transformers”. In:Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV). 2021, pp. 9650–9660

    work page 2021

  4. [4]

    Applied and computational harmonic analysis21(1), 5–30 (2006) https://doi.org/10.1016/j.acha.2006.04.006

    Ronald R. Coifman and Stéphane Lafon. “Diffusion maps”. In:Applied and Computational Harmonic Analysis21.1 (2006), pp. 5–30.doi:10.1016/j.acha.2006.04.006

    work page doi:10.1016/j.acha.2006.04.006 2006

  5. [5]

    Geometric diffusions as a tool for harmonic analysis and structure definition of data: Diffusion maps

    Ronald R. Coifman et al. “Geometric diffusions as a tool for harmonic analysis and structure definition of data: Diffusion maps”. In:Proceedings of the National Academy of Sciences102.21 (2005), pp. 7426– 7431.doi:10.1073/pnas.0500334102

    work page doi:10.1073/pnas.0500334102 2005

  6. [6]

    Thermal shock response via weakly coupled peridynamic thermo-mechanics

    Pietro D’Antuono and Marco Morandini. “Thermal shock response via weakly coupled peridynamic thermo-mechanics”. In:International Journal of Solids and Structures129 (2017), pp. 74–89.issn: 0020-7683.doi:https://doi.org/10.1016/j.ijsolstr.2017.09.010

    work page doi:10.1016/j.ijsolstr.2017.09.010 2017

  7. [7]

    AnomalyDINO: Boosting Patch-Based Few-Shot Anomaly Detection with DINOv2

    Simon Damm et al. “AnomalyDINO: Boosting Patch-Based Few-Shot Anomaly Detection with DINOv2”. In:Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV). 2025, pp. 1319–1329

    work page 2025

  8. [8]

    Simulation of thermal shock cracking in ceramics using bond-based peridynamics and FEM

    Ilias N. Giannakeas, Theodosios K. Papathanasiou, and Hamid Bahai. “Simulation of thermal shock cracking in ceramics using bond-based peridynamics and FEM”. In:Journal of the European Ceramic Society38.8 (2018), pp. 3037–3048.issn: 0955-2219.doi: https : / / doi . org / 10 . 1016 / j . jeurceramsoc.2017.12.039

    work page 2018

  9. [9]

    Jolliffe.Principal Component Analysis

    Ian T. Jolliffe.Principal Component Analysis. Springer, 2002

    work page 2002

  10. [10]

    Performance of different tungsten grades under transient thermal loads

    J. Linke et al. “Performance of different tungsten grades under transient thermal loads”. en. In:Nuclear Fusion51.7 (June 2011), p. 073017.issn: 0029-5515.doi: https://doi.org/10.1088/0029- 5515/51/7/073017. (Visited on 01/09/2026)

    work page doi:10.1088/0029- 2011

  11. [11]

    A user’s guide to the encyclopedia of dna elements (encode).PLoS Biology, 9(4):e1001046, 2011

    Théo Moutakanni et al. “Cell-DINO: Self-supervised image-based embeddings for cell fluorescent microscopy”. In:PLOS Computational Biology21.12 (2025), e1013828.doi: 10.1371/journal. pcbi.1013828

    work page doi:10.1371/journal 2025

  12. [12]

    Continuous Radon Monitoring during Seven Years of V olcanic Unrest at Campi Flegrei Caldera (Italy)

    Gustav Müller-Franzes et al. “Medical slice transformer for improved diagnosis and explainability on 3D medical images with DINOv2”. In:Scientific Reports15 (2025), p. 23979.doi: 10.1038/s41598- 025-09041-8

    work page doi:10.1038/s41598- 2025

  13. [13]

    Murphy.Machine Learning: A Probabilistic Perspective

    Kevin P. Murphy.Machine Learning: A Probabilistic Perspective. MIT Press, 2012

    work page 2012

  14. [14]

    On Estimating Regression

    E. A. Nadaraya. “On Estimating Regression”. In:Theory of Probability&Its Applications9.1 (1964), pp. 141–142

    work page 1964

  15. [15]

    DINOv2: Learning Robust Visual Features without Supervision

    Maxime Oquab et al. “DINOv2: Learning Robust Visual Features without Supervision”. In:Transac- tions on Machine Learning Research(2024). Also available as arXiv:2304.07193

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  16. [16]

    A Threshold Selection Method from Gray-Level Histograms

    Nobuyuki Otsu. “A Threshold Selection Method from Gray-Level Histograms”. In:IEEE Transactions on Systems, Man, and Cybernetics9.1 (1979), pp. 62–66.doi:10.1109/TSMC.1979.4310076

    work page doi:10.1109/tsmc.1979.4310076 1979

  17. [17]

    Taylor, P

    Sam Reeve et al.CabanaPD. Version 0.4.0. Aug. 2025.url: https://doi.org/10.5281/zenodo. 16997636

    work page doi:10.5281/zenodo 2025

  18. [18]

    Nonlinear dimensionality reduction by locally linear embed- ding

    Sam T. Roweis and Lawrence K. Saul. “Nonlinear dimensionality reduction by locally linear embed- ding”. In:Science290.5500 (2000), pp. 2323–2326. 36

    work page 2000

  19. [19]

    Three-dimensional multi-scale line filter for segmentation and visualization of curvilinear structures in medical images

    Yoshinobu Sato et al. “Three-dimensional multi-scale line filter for segmentation and visualization of curvilinear structures in medical images”. In:Medical Image Analysis2.2 (1998), pp. 143–168.issn: 1361-8415.doi:https://doi.org/10.1016/S1361-8415(98)80009-1

    work page doi:10.1016/s1361-8415(98)80009-1 1998

  20. [20]

    Ridge Regression Learning Algorithm in Dual Variables

    Craig Saunders, Alexander Gammerman, and Vladimir V ovk. “Ridge Regression Learning Algorithm in Dual Variables”. In:Proceedings of the 15th International Conference on Machine Learning (ICML) (1998), pp. 515–521

    work page 1998

  21. [21]

    Smola.Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond

    Bernhard Schölkopf and Alexander J. Smola.Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond. MIT Press, 2002

    work page 2002

  22. [22]

    Survey over image thresholding techniques and quantitative performance evaluation

    Mehmet Sezgin and Bülent Sankur. “Survey over image thresholding techniques and quantitative performance evaluation”. In:Journal of Electronic Imaging13.1 (2004), pp. 146–165.doi: https: //doi.org/10.1117/1.1631315

    work page doi:10.1117/1.1631315 2004

  23. [23]

    Peridynamic States and Constitutive Modeling

    S. A. Silling et al. “Peridynamic States and Constitutive Modeling”. In:Journal of Elasticity88.2 (Aug. 2007), pp. 151–184.issn: 1573-2681.doi:10.1007/s10659-007-9125-1

    work page doi:10.1007/s10659-007-9125-1 2007

  24. [24]

    Reformulation of elasticity theory for discontinuities and long-range forces

    S.A. Silling. “Reformulation of elasticity theory for discontinuities and long-range forces”. In:Journal of the Mechanics and Physics of Solids48.1 (2000), pp. 175–209.issn: 0022-5096.doi: https : //doi.org/10.1016/S0022-5096(99)00029-0

    work page doi:10.1016/s0022-5096(99)00029-0 2000

  25. [25]

    A meshfree method based on the peridynamic model of solid mechanics

    S.A. Silling and E. Askari. “A meshfree method based on the peridynamic model of solid mechanics”. In:Computers&Structures83.17 (2005), pp. 1526–1535.issn: 0045-7949.doi: https://doi.org/ 10.1016/j.compstruc.2004.11.026

    work page doi:10.1016/j.compstruc.2004.11.026 2005

  26. [26]

    UMAP: Uniform Manifold Approximation and Projection

    Stuart Slattery et al. “Cabana: A Performance Portable Library for Particle-Based Simulations”. In: Journal of Open Source Software7.72 (2022), p. 4115.doi: https://doi.org/10.21105/joss. 04115

    work page doi:10.21105/joss 2022

  27. [27]

    Stephens et al.A Brief Introduction to Generative Models

    Christopher R. Stephens et al.A Brief Introduction to Generative Models. 2021. arXiv: 2103.00265 [stat.ML].url:https://arxiv.org/abs/2103.00265

    work page arXiv 2021

  28. [28]

    Science290(5500), 2319–2323 (2000) https://doi.org/10.1126/ science.290.5500.2319

    Joshua B. Tenenbaum, Vin de Silva, and John C. Langford. “A global geometric framework for nonlinear dimensionality reduction”. In:Science290.5500 (2000), pp. 2319–2323.doi: 10.1126/ science.290.5500.2319

    work page arXiv 2000

  29. [29]

    Anisotropic two-dimensional, plane strain, and plane stress models in classical linear elasticity and bond-based peridynamics

    Jeremy Trageser and Pablo Seleson. “Anisotropic two-dimensional, plane strain, and plane stress models in classical linear elasticity and bond-based peridynamics”. In:arXiv preprint arXiv:1905.12761 (2019)

    work page arXiv 1905

  30. [30]

    Kokkos 3: Programming Model Extensions for the Exascale Era

    Christian R. Trott et al. “Kokkos 3: Programming Model Extensions for the Exascale Era”. In: IEEE Transactions on Parallel and Distributed Systems33.4 (2022), pp. 805–817.doi: https : //doi.org/10.1109/TPDS.2021.3097283

    work page doi:10.1109/tpds.2021.3097283 2022

  31. [31]

    A 3-D conjugated bond-pair-based peridynamic formulation for initiation and propagation of cracks in brittle solids

    Yunteng Wang et al. “A 3-D conjugated bond-pair-based peridynamic formulation for initiation and propagation of cracks in brittle solids”. In:International Journal of Solids and Structures134 (Mar. 2018), pp. 89–115.issn: 0020-7683.doi: https://doi.org/10.1016/j.ijsolstr.2017.10.022. (Visited on 03/25/2026)

    work page doi:10.1016/j.ijsolstr.2017.10.022 2018

  32. [32]

    Smooth Regression Analysis

    G. S. Watson. “Smooth Regression Analysis”. In:Sankhy ¯a: The Indian Journal of Statistics, Series A 26.4 (1964), pp. 359–372

    work page 1964

  33. [33]

    Learning Microstructure–Property Relation- ships in Materials with Robust Features from Vision Transformers

    Sheila E. Whitman, Guangyu Hu, and Marat I. Latypov. “Learning Microstructure–Property Relation- ships in Materials with Robust Features from Vision Transformers”. In:Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). 2024, pp. 8125–8130

    work page 2024

  34. [34]

    Machine learning of microstructure–property relationships in materials leveraging microstructure representation from foundational vision transformers

    Sheila E. Whitman and Marat I. Latypov. “Machine learning of microstructure–property relationships in materials leveraging microstructure representation from foundational vision transformers”. In:Acta Materialia296 (2025), p. 121217.doi:10.1016/j.actamat.2025.121217

    work page doi:10.1016/j.actamat.2025.121217 2025

  35. [35]

    Thermal shock tests to qualify different tungsten grades as plasma facing material

    M Wirtz et al. “Thermal shock tests to qualify different tungsten grades as plasma facing material”. en. In:Physica ScriptaT167 (Feb. 2016), p. 014015.issn: 0031-8949, 1402-4896.doi: https : //doi.org/10.1088/0031-8949/T167/1/014015. (Visited on 01/12/2026). 37

    work page doi:10.1088/0031-8949/t167/1/014015 2016

  36. [36]

    Oliver Marius Wirtz.Thermal shock behaviour of different tungsten grades under varying conditions. V ol. 161. Forschungszentrum Jülich, 2013

    work page 2013

  37. [37]

    Automatic measurement of sister chromatid exchange frequency

    G W Zack, W E Rogers, and S A Latt. “Automatic measurement of sister chromatid exchange frequency.” In:Journal of Histochemistry&Cytochemistry25.7 (1977), pp. 741–753.doi: https: //doi.org/10.1177/25.7.70454. eprint:https://doi.org/10.1177/25.7.70454. 38

    work page doi:10.1177/25.7.70454 1977