Pith. sign in

REVIEW 4 major objections 4 minor 42 references

Harnessing disorder to decouple extension and shear in kirigami metamaterials

T0 review · 4 major / 4 minor · reviewed 2026-08-01 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Randomized kirigami cuts kill parasitic shear, and a graph neural network designs the patterns.

desk verdict Solid inverse-design pipeline for stochastic kirigami with real experimental validation, but the flagship shear-decoupling claim is only simulated, not measured. read the letter →

arxiv 2607.16583 v1 pith:2OENQOD2 submitted 2026-07-18 cond-mat.mtrl-sci cs.LG

Haomin Yu , Hanxun Jin , Mingxuan Bi , Mohammad Jafari , Feng Helen Long , Michael J Greenberg , Farid Alisafaei , Guy Genin This is my paper · ORCID
classification cond-mat.mtrl-scics.LG
keywords stochastickirigamiextension–shearcouplingdisorderasdesignvariablegraphneuralnetworkinverseanisotropymechanicalmetamaterialssurrogatemodeling
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper tries to establish that engineered disorder—randomizing cut orientations and placements in kirigami sheets—is not a defect but a programmable design variable. Periodic cut patterns force panels to rotate in a coordinated way, so stretching one axis always produces a parasitic shear and leaves anisotropy locked to a discrete set of responses. The authors argue that stochastic cut patterns let local positive and negative shears cancel, yielding near-zero net shear even at 100% strain and a continuous range of anisotropic stiffness. Because disordered patterns lack a simple parameterization, they build a graph neural network that maps cut topology to the full nonlinear stress–strain response in two directions, and couple it to a genetic algorithm that inverse-designs patterns matching prescribed targets. Fabricated silicone samples reproduce the predicted anisotropic curves, closing the loop from target to physical part.

What carries the argument

The central object is the cut-network graph: each cut is represented as a polyline of nodes connected by undirected edges, with node coordinates as features. This representation carries the cut topology and spacing that govern kirigami mechanics into a message-passing GNN, which then outputs the 60-component stress vector (30 strain levels in each of two directions). The same graph representation, parsed from a parametric description of cut-orientation ranges and density, lets the genetic algorithm search physically valid disordered patterns without enumerating individual cuts.

What would settle it

If a stochastic kirigami sheet with a new random seed but the same orientation statistics as the fabricated samples is stretched to 100% strain and the measured boundary shear force exceeds, say, 5% of the normal force, the claimed near-complete decoupling fails. Alternatively, retrain the surrogate on a test set of patterns whose cut spacings lie outside the training range; if held-out prediction error jumps well above the reported 3.5%, the graph representation is not enough.

Watch

Extended reading notes

Core claim

In periodic kirigami, identical unit cells rotate coherently, concentrating shear along a biased band; in stochastic kirigami, regions of positive and negative local shear are interspersed and cancel on average, so a sheet stretched along one axis develops essentially no net shear at the boundary. This gives stochastic architectures a behavior periodic ones cannot reach—extension decoupled from shear—and makes in-plane anisotropy a continuous function of cut-orientation statistics rather than a discrete choice of motif. The paper further shows that a geometry-aware graph neural network, trained on finite-element data, predicts the full nonlinear bidirectional stress–strain response of disord

Load-bearing premise

The central claim stands or falls on whether the graph representation fed to the surrogate contains enough information about inter-cut spacing and orientation correlations to predict mechanics for patterns the genetic algorithm has never seen—a premise the paper asserts but does not directly test, and one it acknowledges is limited by extrapolation beyond the training distribution (Discussion, Section 3.2).

Editorial extensions

If this is right

  • If the central claim holds, designers no longer need periodic lattices to get predictable kirigami behavior; statistics of cut orientation and spacing can tune stiffness anisotropy continuously.
  • Stretchable sheets for actuators or tissue-interfacing devices could be made to extend along one axis without generating a parasitic shear that disrupts alignment or sensing.
  • The GNN-based surrogate makes inverse design of disordered architectures practical: each evaluation takes a fraction of a finite-element solve, so a genetic algorithm can search thousands of candidates.
  • Because the representation is connectivity-based, the same pipeline may extend to other discrete-element architected materials that lack a low-dimensional parameterization.
  • The experimental validation suggests the simulated decoupling survives fabrication: at 100% strain, the physical samples deform like the finite-element predictions.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: the decoupling result implies a design rule—'disorder averages local stress states instead of amplifying them'—that could be tested directly by measuring the width of the local shear distribution as cut-orientation randomness increases; the paper reports the distribution is symmetric about zero for stochastic patterns, so a quantitative collapse toward zero as disorder grows w
  • Editorial inference: because the genetic algorithm optimizes the statistics of the cut field, not individual cuts, the approach effectively inverse-designs the probability distribution of cuts; this suggests the same pipeline could target higher-order statistics such as orientation correlation length or two-point spacing, which the current graph encoding captures only globally.
  • Editorial inference: the near-zero net shear in stochastic kirigami may convert to an effective negative or near-zero off-diagonal term in a coarse-grained constitutive law; if so, one could write down a continuum model for disordered kirigami in which the shear modulus decouples from the stretch response—a testable prediction beyond the paper's explicit claims.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The paper proposes that engineered disorder in kirigami cut patterns is a design degree of freedom. It claims that stochastic kirigami can achieve a continuous and much broader range of mechanical responses than periodic patterns, including near-complete elimination of extension-shear coupling. To navigate the disorder space, the authors train a graph neural network (GNN) surrogate on 3,600 finite-element simulations to predict the full nonlinear biaxial stress-strain response, then couple it to a genetic algorithm (GA) for inverse design. Four inverse-designed Ecoflex samples are fabricated and tested under uniaxial tension, and their measured normal stress-strain curves agree with targets and predictions. The paper includes code/data availability, seven-seed training variability, a mesh-convergence study, and an Ogden material fit.

Significance. If the central claims hold, the paper would establish disorder as a programmable design axis for kirigami and demonstrate a practical GNN-GA pipeline for inverse-designing anisotropic mechanical metamaterials. The strengths include reproducible code and dataset, seven independent GNN training runs, mesh convergence checks, and experimental validation of the normal stress-strain response for four inverse-designed patterns. The main physical claim about shear decoupling, however, is supported only by finite-element simulations; no experimental shear measurement is reported. The GNN representation also relies on a graph with no inter-cut edges, so the mechanistic interpretation of 'cut interactions' is not directly encoded. These are load-bearing gaps for the paper's headline claims.

major comments (4)
  1. [Section 2.1, Fig. 1E–H] The claim of 'near-complete elimination of extension-shear coupling' is based entirely on FEM simulation for representative stochastic and periodic patterns. Fig. 1E–H show simulated shear traction, boundary force ratios, and shear-stress distributions, but no fabricated stochastic specimen was tested for shear force, lateral displacement, or shear strain. The four experimental samples in Fig. 5C–F were loaded in uniaxial tension only, and only normal stress-strain curves are reported. Since the abstract and conclusions present this decoupling as a principal advantage, the authors should either provide direct experimental shear measurements on stochastic and periodic samples or explicitly qualify the claim as a simulation-based prediction.
  2. [Materials and Methods, GNN surrogate; Fig. 3A] The graph representation assigns nodes to cut points and undirected edges only between consecutive points on the same cut; the graph is therefore a set of disconnected polylines. Message passing along these edges cannot directly exchange information between different cuts, so the phrase 'encodes cuts and their interactions directly as nodes and edges' is inaccurate. Inter-cut spacing and orientation correlations are captured only indirectly through global mean pooling of coordinate-based node features. To support the surrogate's generalization to GA-proposed designs, the authors should either add inter-cut edges (e.g., proximity-based edges) or provide a quantitative analysis showing that the global pooling preserves the relevant inter-cut statistics for the explored design space.
  3. [Section 2.1, Fig. 1] The claim that stochastic kirigami accesses a 'continuous and far broader region of mechanical response' than periodic patterns is not quantitatively demonstrated. Fig. 1 compares three representative patterns and their stress-strain curves, but no systematic mapping of the reachable response space (e.g., distributions of anisotropic stiffness ratios and shear-coupling measures over ensembles) is given. Such a response-space comparison, even from simulation, would substantiate the 'broader and continuous' statement; alternatively the wording should be moderated to what the data actually show.
  4. [Section 2.3, Fig. 5A] All inverse-design targets are constructed by blending three curves drawn from the training set. This confines the demonstration to interpolation within the training distribution and does not test the claim that arbitrary targets in the broader claimed response space are achievable. The Discussion correctly notes that extrapolation is unreliable, but the abstract's 'programmable anisotropy' statement could be read more strongly than the interpolation-only evidence supports. A statement of this limitation in the Results or an extrapolation test would clarify the scope.
minor comments (4)
  1. [General] There are several typos: 'Graph neutal network' in Fig. 2 caption, 'When, paired with geometry-aware learning' in the Introduction has an errant comma, and 'Ecoflex-0030' should be 'Ecoflex 00-30' to match the rest of the text.
  2. [Materials and Methods] The GA description says 'bounds on cut height' while the data-generation section says 'slit length'; the terminology should be unified. Also, the experimental section does not state how many specimens were tested per design or whether the x-direction and y-direction tests were performed on the same or different specimens.
  3. [Fig. 1F] The p-value 'p < 1e-5' is reported without specifying the sample size or statistical test. Clarify the number of stochastic realizations and the test used.
  4. [Section 2.4, Fig. 5C–F] The agreement between experiment and FEM is shown in insets as deformed shapes and maximum-principal-stress fields. It would be helpful to also quantify the shape agreement (e.g., normalized displacement error) rather than relying on visual comparison.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central claims are grounded in independent FEM, held-out GNN tests, and physical experiments; the main limitations are interpolation and unverified shear experiments, not circular reasoning.

full rationale

The derivation chain is not circular. The extension-shear decoupling claim (Section 2.1) is asserted from finite-element simulations comparing stochastic and periodic cut patterns (Fig. 1C-H); it is not derived from the GNN or from a fitted parameter. The GNN surrogate (Section 2.2) is trained on 3,600 FEM simulations and evaluated on a held-out 20% test split, with MAPE about 3.5% and agreement across seven independently trained networks, so forward predictions are validated against ground truth rather than assumed. Inverse design (Section 2.3) uses targets 'constructed by blending three curves drawn from the training set' — an explicit interpolation within the achievable space, not a hidden fit; the subsequent experimental agreement (Fig. 5C-F) is an external check on the forward surrogate. No load-bearing step relies on a self-citation: references to the authors' prior work (e.g., [6]) motivate the machine-learning approach but do not force the physical result. The paper itself discloses limitations: 'extrapolation beyond the training distribution remains unreliable,' and the shear-decoupling signature is verified only in simulation, not directly by the uniaxial-tension experiments reported. These are scoping and validation concerns, not circularity. Therefore no circular step is present.

Assumptions & free parameters 1 free parameters · 5 assumptions · 0 invented entities

The central claims rest primarily on FEM simulations calibrated with fitted Ogden parameters, on a statistical isotropy argument for shear cancellation, on an unverified graph representation, and on inverse-design targets drawn from the training distribution.

free parameters (1)
  • Ogden hyperelastic coefficients = µ1=1164.77 kPa, α1=0.03, µ2=5.31 kPa, α2=3.48, µ3=267.86 kPa, α3=0.03
    Fitted to uniaxial tests of uncut Ecoflex 00-30; all FEM training data depend on these values, but they are material characterization rather than fit to the kirigami claims.
assumptions (5)
  • domain assumption FEM with S4R shell elements and 0.25 mm mesh accurately represents the kirigami sheet response up to 100% uniaxial strain.
    Mesh convergence reported in Fig. A1, but no direct experimental verification of the local shear fields; out-of-plane buckling is excluded.
  • domain assumption Uniformly random cut orientations produce local shear stresses symmetric about zero, so net shear vanishes.
    Demonstrated for one stochastic iso-kirigami pattern in Fig. 1G/H; relies on statistical isotropy and may not hold for finite anisotropic realizations.
  • ad hoc to paper The graph representation preserves the information needed to predict stress-strain behavior.
    Methods state edges connect consecutive points of the same cut only; no inter-cut edges, so the model must infer interactions from global pooling, which is not established.
  • ad hoc to paper Blending three training curves produces target responses representative of the achievable design space.
    Inverse design is only validated on in-distribution targets; out-of-distribution generalization is explicitly disclaimed.
  • domain assumption Ogden parameters calibrated from uncut Ecoflex dogbones describe the patterned sheet material.
    Standard material calibration, but transfer to thin patterned geometry with stress concentrations is assumed.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Harnessing disorder to decouple extension and shear in kirigami metamaterials." pith.science (2026). https://pith.science/paper/2OENQOD2

@misc{pith2026260716583,
  author       = {Pith},
  title        = {Pith review of: Harnessing disorder to decouple extension and shear in kirigami metamaterials},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2OENQOD2}},
  note         = {Machine review of arXiv:2607.16583}
}
read the original abstract

Kirigami turns stiff sheets into compliant, shape-morphing structures, but its reliance on periodic cut patterns comes at a cost: correlated panel rotations couple extension to shear, so stretching one axis drives a parasitic shear that cannot be suppressed, and also confine anisotropic stiffness to a narrow, discrete set of responses that cannot be tuned independently. Biological tissues overcome an analogous constraint through controlled disorder, such as graded fiber orientations in skin and hierarchical anisotropy in myocardium, achieving direction-dependent mechanics unavailable to regular architectures. Here, we show that engineered disorder is a design degree of freedom for kirigami, with stochastic kirigami accessing a continuous and far broader region of mechanical response than periodic patterns. This includes programmable anisotropy with near-complete elimination of extension-shear coupling. Because disordered patterns lack a simple parameterization, we navigate this design space with a geometry-aware graph neural network (GNN) that maps cut topology to the full nonlinear, bidirectional stress-strain response, coupled to a genetic algorithm that inverse-designs patterns reproducing target responses along two perpendicular axes. The GNN trains an order of magnitude faster and more accurately than image-based models. Fabricated elastomer samples reproduce the predicted nonlinear, anisotropic responses, closing the loop from design to physical component. By turning disorder into a variable to control directional stiffness, this work develops architected materials that stretch without parasitic shear, from soft actuators to tissue-interfacing devices matched to the anisotropy of living tissue.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

42 extracted references

  1. [1]

    & Van Hecke, M

    Bertoldi, K., Vitelli, V., Christensen, J. & Van Hecke, M. Flexible mechanical metamaterials.Nature Reviews Materials2, 1–11 (2017)

    2017

  2. [2]

    & Espinosa, H

    Jin, H. & Espinosa, H. D. Mechanical metamaterials fabricated from self-assembly: A perspective. Journal of Applied Mechanics91, 040801 (2024)

    2024

  3. [3]

    R., Zheng, X

    Jiao, P., Mueller, J., Raney, J. R., Zheng, X. & Alavi, A. H. Mechanical metamaterials and beyond. Nature communications14, 6004 (2023)

    2023

  4. [4]

    U.et al.Mechanical metamaterials and their engineering applications.Advanced Engineering Materials21, 1800864 (2019)

    Surjadi, J. U.et al.Mechanical metamaterials and their engineering applications.Advanced Engineering Materials21, 1800864 (2019)

    2019

  5. [5]

    Jin, H.et al.In situ mechanical characterization of functional and architected materials.Nature Materials1–19 (2026)

    2026

  6. [6]

    Jin, H.et al.Characterization and inverse design of stochastic mechanical metamaterials using neural operators.Advanced Materials37, 2420063 (2025)

    2025

  7. [7]

    C.et al.A kirigami approach to engineering elasticity in nanocomposites through patterned defects.Nature materials14, 785–789 (2015)

    Shyu, T. C.et al.A kirigami approach to engineering elasticity in nanocomposites through patterned defects.Nature materials14, 785–789 (2015)

    2015

  8. [8]

    K.et al.Graphene kirigami.Nature524, 204–207 (2015)

    Blees, M. K.et al.Graphene kirigami.Nature524, 204–207 (2015)

    2015

Show all 42 references
  1. [9]

    Zhang, Y.et al.A mechanically driven form of kirigami as a route to 3d mesostructures in micro/nanomembranes.Proceedings of the National Academy of Sciences112, 11757–11764 (2015)

    2015

  2. [10]

    Advanced Materials34, 2202478 (2022)

    Meng, K.et al.Kirigami-inspired pressure sensors for wearable dynamic cardiovascular monitoring. Advanced Materials34, 2202478 (2022)

    2022

  3. [11]

    K., Chakravarty, S., Ali, M

    Brooks, A. K., Chakravarty, S., Ali, M. & Yadavalli, V. K. Kirigami-inspired biodesign for applications in healthcare.Advanced Materials34, 2109550 (2022)

    2022

  4. [12]

    & Bertoldi, K

    Rafsanjani, A. & Bertoldi, K. Buckling-induced kirigami.Physical review letters118, 084301 (2017)

    2017

  5. [13]

    Genin, G. M. & Thomopoulos, S. Unification through disarray.Nature materials16, 607–608 (2017)

    2017

  6. [14]

    Golman, M.et al.Toughening mechanisms for the attachment of architectured materials: The mechanics of the tendon enthesis.Science Advances7, eabi5584 (2021)

    2021

  7. [15]

    & Zapperi, S

    Zaiser, M. & Zapperi, S. Disordered mechanical metamaterials.Nature Reviews Physics5, 679–688 (2023)

    2023

  8. [16]

    N., Bruy` ere, K., Destrade, M., Gilchrist, M

    Annaidh, A. N., Bruy` ere, K., Destrade, M., Gilchrist, M. D. & Ott´ enio, M. Characterization of the anisotropic mechanical properties of excised human skin.Journal of the mechanical behavior of biomedical materials5, 139–148 (2012). 10

    2012

  9. [17]

    & Genet, M

    Tueni, N., Allain, J.-M. & Genet, M. On the structural origin of the anisotropy in the myocardium: Multiscale modeling and analysis.Journal of the mechanical behavior of biomedical materials138, 105600 (2023)

    2023

  10. [18]

    A.et al.The role of network architecture in collagen mechanics.Biophysical journal114, 2665–2678 (2018)

    Jansen, K. A.et al.The role of network architecture in collagen mechanics.Biophysical journal114, 2665–2678 (2018)

    2018

  11. [19]

    Chen, S., Choi, G. P. & Mahadevan, L. Deterministic and stochastic control of kirigami topology. Proceedings of the National Academy of Sciences117, 4511–4517 (2020)

    2020

  12. [20]

    & Mahadevan, L

    Chaudhary, G., Niu, L., Han, Q., Lewicka, M. & Mahadevan, L. Geometric mechanics of ordered and disordered kirigami.Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences479(2023)

    2023

  13. [21]

    K., Xiao, H., Durian, D

    Fulco, S., Budzik, M. K., Xiao, H., Durian, D. J. & Turner, K. T. Disorder enhances the fracture toughness of 2d mechanical metamaterials.PNAS nexus4, pgaf023 (2025)

    2025

  14. [22]

    & Singh, C

    Choukir, S., Manohara, N. & Singh, C. V. Disorder unlocks the strength-toughness trade-off in metamaterials.Applied Materials Today42, 102579 (2025)

    2025

  15. [23]

    R.et al.Auxetic metamaterials from disordered networks.Proceedings of the National Academy of Sciences115, E1384–E1390 (2018)

    Reid, D. R.et al.Auxetic metamaterials from disordered networks.Proceedings of the National Academy of Sciences115, E1384–E1390 (2018)

    2018

  16. [24]

    & Kochmann, D

    Bastek, J.-H. & Kochmann, D. M. Inverse design of nonlinear mechanical metamaterials via video denoising diffusion models.Nature Machine Intelligence5, 1466–1475 (2023)

    2023

  17. [25]

    S.et al.Rapid inverse design of metamaterials based on prescribed mechanical behavior through machine learning.Nature Communications14, 5765 (2023)

    Ha, C. S.et al.Rapid inverse design of metamaterials based on prescribed mechanical behavior through machine learning.Nature Communications14, 5765 (2023)

    2023

  18. [26]

    & Espinosa, H

    Jin, H., Zhang, E. & Espinosa, H. D. Recent advances and applications of machine learning in experimental solid mechanics: A review.Applied Mechanics Reviews75, 061001 (2023)

    2023

  19. [27]

    & Watanabe, I

    Zheng, X., Zhang, X., Chen, T.-T. & Watanabe, I. Deep learning in mechanical metamaterials: from prediction and generation to inverse design.Advanced Materials35, 2302530 (2023)

    2023

  20. [28]

    Reiser, P.et al.Graph neural networks for materials science and chemistry.Communications Materials 3, 93 (2022)

    2022

  21. [29]

    & Tanwar, S

    Khemani, B., Patil, S., Kotecha, K. & Tanwar, S. A review of graph neural networks: concepts, archi- tectures, techniques, challenges, datasets, applications, and future directions.Journal of Big Data11, 18 (2024)

    2024

  22. [30]

    F., Liang, Y

    Dai, M., Demirel, M. F., Liang, Y. & Hu, J.-M. Graph neural networks for an accurate and interpretable prediction of the properties of polycrystalline materials.npj Computational Materials7, 103 (2021)

    2021

  23. [31]

    & Kochmann, D

    Zheng, L., Karapiperis, K., Kumar, S. & Kochmann, D. M. Unifying the design space and optimizing linear and nonlinear truss metamaterials by generative modeling.Nature Communications14, 7563 (2023). 11

    2023

  24. [32]

    R., Afrasiabi, M

    Frey, R., Tucker, M. R., Afrasiabi, M. & Bambach, M. Multi-objective design of multi-material truss lattices utilizing graph neural networks.Scientific Reports15, 3187 (2025)

    2025

  25. [33]

    Maurizi, M.et al.Designing metamaterials with programmable nonlinear responses and geometric constraints in graph space.Nature Machine Intelligence7, 1023–1036 (2025)

    2025

  26. [34]

    & van Egmond, D

    Dold, D. & van Egmond, D. A. Differentiable graph-structured models for inverse design of lattice materials.Cell Reports Physical Science4(2023)

    2023

  27. [35]

    & Slesarenko, V

    Felsch, G. & Slesarenko, V. Generative models struggle with kirigami metamaterials.Scientific Reports 14, 19397 (2024)

    2024

  28. [36]

    Yu, H.et al.Expansion limits of meshed split-thickness skin grafts.Acta biomaterialia191, 325–335 (2025)

    2025

  29. [37]

    M.et al.Functional grading of mineral and collagen in the attachment of tendon to bone

    Genin, G. M.et al.Functional grading of mineral and collagen in the attachment of tendon to bone. Biophysical journal97, 976–985 (2009)

    2009

  30. [38]

    CS de Alcantara, A.et al.Molecular-scale interactions at mineralized collagen interfaces prevent network percolation, preserving compliance.ACS nano19, 31350–31362 (2025)

    2025

  31. [39]

    Hu, Y.et al.Stochastic interdigitation as a toughening mechanism at the interface between tendon and bone.Biophysical Journal108, 431–437 (2015)

    2015

  32. [40]

    & Monaghan, M

    Olvera, D., Sohrabi Molina, M., Hendy, G. & Monaghan, M. G. Electroconductive melt electrowritten patches matching the mechanical anisotropy of human myocardium.Advanced Functional Materials 30, 1909880 (2020)

    2020

  33. [41]

    A., Someya, T

    Rogers, J. A., Someya, T. & Huang, Y. Materials and mechanics for stretchable electronics.science 327, 1603–1607 (2010)

    2010

  34. [42]

    A., Zhang, Y

    Xue, Z., Song, H., Rogers, J. A., Zhang, Y. & Huang, Y. Mechanically-guided structural designs in stretchable inorganic electronics.Advanced Materials32, 1902254 (2020). 12 A Max. Principal Stress 1200 (kPa) Periodic kirigamiStochastic Iso-kirigamiStochastic Aniso-kirigami 0 0...

    2020

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.