An $H^{-1}$ least-squares UnCut FEM on domains defined by a level set function

Buyang Li; Han Yang; Jiashun Hu

An $H^{-1}$ least-squares UnCut FEM on domains defined by a level set function

T0 review · 0 major / 5 minor · reviewed 2026-08-04 · deepseek-v4-flash

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

Pith's one-line read An H⁻¹ least-squares reformulation makes unfitted finite element methods stable with fixed unit weights, proving optimal-order convergence for Poisson and Stokes on level-set domains.

desk verdict The paper's main stability claim survives the reader's objection: in Lemma 3.8 the sigma from Lemma 3.6 is an auxiliary parameter, not the scheme weight, so the unit-weight scheme stands. read the letter →

arxiv 2608.02015 v1 pith:H42GZ4XX submitted 2026-08-03 math.NA cs.NA

Jiashun Hu , Buyang Li , Han Yang This is my paper

classification math.NAcs.NA MSC 65N1265N1565N85

keywords unfittedfiniteelementmethodlevelsetfunctionphi-FEMleast-squaresH⁻¹residualStokesequationsPoissonequationstabilization-free

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 proposes a new unfitted finite element method for solving the Poisson and Stokes equations on domains described by a level set function. Instead of relying on large, user-chosen stabilization parameters, the method minimizes a discrete H⁻¹ residual of the governing equations, which makes the scheme stable with all weights fixed at one. The authors prove optimal-order convergence in the H¹ norm for the Poisson solution and, for Stokes, in H¹ for velocity and L² for pressure, on meshes that cut the boundary arbitrarily. Numerical experiments confirm the predicted rates and show that the method is far less sensitive to the stabilization weight than the earlier phi-FEM approach. The method avoids numerical integration over curved cut cells because it works on the full background mesh.

What carries the argument

The key object is the discrete H⁻¹ residual, realized through the Riesz map (I-Δ_h)⁻¹: the auxiliary variable w_h is the H¹ function whose inner product with any test function reproduces the strong residual of the equation extended to the active mesh. The variational formulation then has a saddle-point structure in (w_h, u_h) (or (w_h, u_h, p_h)) whose diagonal term is ||w_h||_{H¹}², and the proof of stability uses a generalized inf-sup condition with a theoretical test function (w_h + α u_h, u_h) for arbitrarily small α, so no large parameter enters the actual scheme. The jump and residual penalties j_h and h²||...||² are only needed to control the narrow boundary strip.

What would settle it

Run scheme (2.19) for the two-dimensional Stokes equations on a background mesh with a straight boundary that cuts a row of elements at a small volume fraction, then assemble the full matrix with unit weights and compute its smallest nonzero singular value on a sequence of meshes h, h/2, h/4. If this singular value decays like a negative power of h, or if the H¹ velocity error for a smooth solution fails to converge at the predicted rate, then the unit-weight coercivity claim is false.

Watch

Extended reading notes

Core claim

The central claim is that introducing an auxiliary variable w_h defined as the H¹ Riesz representative of the discrete residual, w_h = (I-Δ_h)⁻¹(Δ̃_h u_h + P_h f) for Poisson and similarly for Stokes, converts the phi-FEM stabilization problem into a least-squares problem whose intrinsic H⁻¹ residual structure provides stability. The resulting schemes (2.11) and (2.19) are parameter-free—their stabilization terms carry fixed unit weights—and the paper proves that for k≥2 Taylor-Hood elements, the velocity and pressure satisfy ||u_h - ũ||_{H¹(Ω_h)} + ||p_h - Π̃p||_{L²(Ω_h)} ≤ C(h^k + ||φ-φ_h||_{H¹})(||f||_{H^{k-1}} + ||f||_{H¹}). The discovery is that the H⁻¹ residual norm itself controls the

Load-bearing premise

The stability proof uses a coercivity estimate (Lemma 3.6) that the cited phi-FEM theory guarantees only for a sufficiently large stabilization weight σ, while the proposed scheme fixes the corresponding weight at 1; the paper never proves the estimate holds at σ=1, so the parameter-free stability claim rests on an unverified transfer.

Editorial extensions

If this is right

For Poisson and Stokes on level-set domains, the user can set every stabilization weight to 1 and still get optimal-order H¹ (velocity) and L² (pressure) accuracy; no penalty-parameter scan is needed.
When the level set is approximated by its degree-k Lagrange interpolant, the φ approximation error term becomes O(h^k), so the methods converge at the full order of the finite element space.
Because all volume integrals are on full background elements and only C⁰ finite elements are used, the same code can handle complex or moving interfaces by simply changing the level set function.
The Stokes result covers the standard Taylor-Hood P_k/P_{k-1} pair for k≥2, so the method is directly usable with existing finite element software.

Reading between the lines

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

A likely extension is to other saddle-point problems (e.g., Navier-Stokes or Darcy flow): the recipe of adding an H⁻¹ residual variable and unit-weight jump terms may reproduce the same stabilization-free behavior.
The proof's reliance on a large-σ coercivity lemma suggests the first check a reader should perform is whether the unit-weight coercivity actually holds for a single straight-cut element; if it fails on that configuration, the parameter-free claim would need a different proof.
The observed-but-unproved optimal L² rate for both Poisson and Stokes is a natural next theorem: a duality argument could upgrade the H¹ estimates to L², similar to standard Aubin-Nitsche techniques.
For time-dependent problems with moving interfaces, the parameter-free nature of the scheme would remove the need to re-tune penalties at each time step, a concrete practical benefit worth testing.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Bottom line up front: the central claim holds. The reader's weakest-assumption worry about Lemma 3.6 requiring large sigma does not survive close reading. In the proof of Lemma 3.8, sigma is an auxiliary parameter used to control the alpha-scaled combination alpha[(grad u,grad u) - (∂_n u,u)] + h^2||Δu||^2 + j_h. Writing A=(grad u,grad u), B=(∂_n u,u), D=h^2||Δu||^2, J=j_h, the expression is alpha(A-B)+D+J = alpha(A-B+sigma D+sigma J)+(1-alpha sigma)(D+J). Choosing alpha < 1/sigma and substituting (3.10) gives the lower bound in (3.16); no version of (3.10) at sigma=1 is ever needed. So the parameter-free stability claim is not endangered by that gap.

The genuinely new thing is the H^{-1} least-squares UnCut FEM itself: a discrete H^{-1} residual minimized via the Riesz map (I-Δ_h)^{-1}, with fixed unit weights, avoiding integration over cut subregions. That is a real mechanism change from phi-FEM and it directly targets the practical pain point of unknown stabilization parameters. The analysis is extensive and mostly careful: coercivity, generalized inf-sup, interpolation with explicit level-set error, and optimal H1 velocity plus L2 pressure estimates. The numerical experiments show optimal rates and a nice sigma-sensitivity comparison where the new method stays accurate over a wide range while phi-FEM degrades. No code is shipped, but the FEniCS setup is straightforward and the results look reproducible.

Softer spots: the observed L2 convergence rates are not proved, and the paper acknowledges that. The modified pressure inf-sup pair in Lemma 3.12 relies on a discrete extension result from the CutFEM literature and a constant-shift argument; I would ask a referee to verify the extension preserves the needed bounds in this unfitted setting, but nothing in the text suggests it breaks. Also, the "parameter-free" claim should be parsed carefully: there are still jump-penalty terms, but their weights are fixed at unit values, which is exactly the point and is a reasonable use of the phrase.

This is a serious, honestly written numerical analysis paper. The method is new, the proofs are detailed, and the main objection raised by the first read does not land. I would send it to peer review and would cite it if I work on unfitted or phi-FEM methods.

Referee Report

0 major / 5 minor

Summary. The paper proposes an UnCut finite element method for Poisson and Stokes equations on domains defined by a level set function. Building on the φ-FEM framework, the new scheme replaces the usual large stabilization parameters with an H^{-1} least-squares term: an auxiliary variable w_h represents the residual through the discrete Riesz map (I-Δ_h)^{-1}. For Poisson the method is (2.11), and for Stokes it is (2.19), with fixed unit weights and a jump penalty. The main results are Theorem 2.1 and Theorem 2.2, claiming unique solvability and optimal-order H^1 velocity error and L^2 pressure error (up to a constant), under Assumptions 2.1–2.3, with error bounds of the form C(h^k + \|φ-φ_h\|_{H^1(Ω_h)}). The proofs use generalized inf-sup conditions with an auxiliary parameter α that does not appear in the scheme. Numerical experiments in 2D/3D Poisson and 2D Stokes confirm the predicted rates and compare favorably with φ-FEM when the stabilization parameter is varied.

Significance. If the results are correct, the paper makes a valuable contribution to unfitted finite element methods: it removes the a priori unknown stabilization threshold that complicates standard φ-FEM, while preserving the implementation advantages of avoiding integration over cut elements. The analysis is extensive and largely self-contained, with cited φ-FEM lemmas used under the stated assumptions. There are no fitted parameters and the numerical experiments support the theoretical convergence rates. I specifically examined the reader's concern that the stability proof requires a large stabilization parameter in Lemma 3.6 while the scheme fixes the corresponding weight to 1. This concern does not land: in Lemma 3.8, σ in (3.10) is an auxiliary parameter, not the scheme weight; the choice α<1/σ turns (3.10) into the needed lower bound (3.16), and no version of (3.10) at σ=1 is required. The central claims are therefore not weakened by that issue.

minor comments (5)

[§4.3] The pressure convergence experiment does not state that the exact pressure is mean-adjusted over Ω_h before computing the L2 error. Since p_h ∈ Q_h has zero mean over Ω_h and the theorem is stated for Π \tilde p, the reported optimal L2 convergence for p would otherwise be impossible. Please state the normalization explicitly, as is done in §4.4.
[Lemma 3.6 / Remark 3.1] The auxiliary role of σ in Lemma 3.6 deserves a sentence of clarification. The inequality (3.10) is used only with an auxiliary parameter σ, and the scheme weight remains 1. It would also be helpful to state explicitly that the threshold for σ in Lemma 3.6 can be chosen independent of h, so that the parameter α in Lemma 3.8 can be fixed independently of h and the inf-sup constant is uniform. This is standard in the cited φ-FEM lemmas, but making it explicit would prevent a misreading.
[Eq. (3.46)] There is a typographical error in the display: '1/4|∇·u h∥2' is malformed; it should read '\frac{1}{4}\|\nabla\cdot u_h\|_{L^2(Ω_h)}^2'. Please correct.
[§4.4 and Conclusion] The conclusion states that the method is robust with respect to 'arbitrarily small or irregularly cut' elements, but the numerical comparison in Figure 4 varies only the stabilization parameter σ. A separate small-cut-cell experiment, or more cautious wording, would be appropriate.
[Lemma 3.12] In the proof of the new inf-sup pair, the estimate (3.36b) benefits from the cancellation ∫_{Ω_h} p_h = 0. It would be clearer to write c_p = -|Ω|^{-1}∫_Ω p_h = -|Ω|^{-1}(∫_{Ω_h\Ω}p_h - ∫_{Ω\Ω_h}p_h), so the reader can see how the measure of the symmetric difference enters.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the derivation is self-contained and external benchmarks are used; the σ=1 concern is resolved by the auxiliary-σ argument.

full rationale

The paper's central derivation is not circular. The method is a genuinely new least-squares formulation; the auxiliary variable w_h is the standard Riesz representative of the residual, and no parameter is fitted to data and then renamed as a prediction. The stability proof of Lemma 3.8 uses the phi-FEM coercivity estimate Lemma 3.6, but only with an auxiliary parameter σ: testing with (w_h+αu_h,u_h) gives α(A−B)+D+J = α(A−B+σD+σJ)+(1−ασ)(D+J), and choosing α<1/σ yields (3.16). Thus no version of (3.10) at σ=1 is required, and the scheme's unit weights are not being silently replaced. The remaining ingredients—interpolation (Lemma 3.5), Poisson/Stokes consistency bounds, and the new inf-sup pair (Lemma 3.12)—are proved from explicit hypotheses and cited external lemmas in the phi-FEM literature ([11]–[14]), which are not authored by the present authors. There are no self-citations that carry the argument. The one limitation explicitly noted, that optimal L2 convergence is observed but not proved, is an honest remark and does not constitute circularity. Overall, the derivation chain is independent of its conclusions.

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

The method introduces no free parameters in the scheme, and no invented physical entities. The auxiliary variable w_h is a mathematical device, not a new entity. The central claims rest on standard geometric assumptions on the level set approximation and on standard regularity of the data, plus classical inequalities borrowed from the phi-FEM literature.

assumptions (5)

domain assumption Assumption 2.1: phi is smooth, |nabla phi|>0 on Gamma, phi != 0 off Gamma
Used to ensure the level-set representation is well-posed and to apply Lemmas 3.1-3.2.
domain assumption Assumption 2.2: each cut element connects to an interior element through at most N cut elements, N independent of h
Used in Lemmas 3.6 and 3.10 to keep the narrow band quasi-uniform; standard in phi-FEM analysis.
domain assumption Assumption 2.3: |nabla phi_h| bounded away from 0 near Gamma_h and dist(Gamma, Gamma_h)=O(h)
Required for Lemma 3.11 and the interpolation analysis.
domain assumption Regularity of data: f in H^{k-1}(Omega_h union Omega) intersection H^1, u_tilde in H^{k+1}(Omega_h union Omega) intersection H^3, p_tilde in H^k(Omega_h union Omega) and existence of Stein extensions
Stated in Theorems 2.1-2.2; standard for a priori error estimates; used to bound the consistency error via Lemmas 3.1 and 3.5.
standard math Standard inequalities: Poincare on S_h (Lemma 3.4), trace inequalities (Lemma 3.3, (3.3a)), Hardy inequality (Lemma 3.2), Korn inequality (Lemma 3.9), discrete inf-sup for Stokes (Lemmas 3.11-3.12)
These are cited (with proofs or references) and used throughout the stability and error analysis.

how reviews work

0 comments

Cite this review

Pith. "Pith review of An $H^{-1}$ least-squares UnCut FEM on domains defined by a level set function." pith.science (2026). https://pith.science/paper/H42GZ4XX

@misc{pith2026260802015,
  author       = {Pith},
  title        = {Pith review of: An $H^-1$ least-squares UnCut FEM on domains defined by a level set function},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/H42GZ4XX}},
  note         = {Machine review of arXiv:2608.02015}
}

abstract

We propose a novel UnCut finite element method (FEM) for the Poisson and Stokes equations on domains with curved boundaries represented by a level set function. Like the $\phi$-FEM, the method avoids numerical integration over cut subregions of boundary elements, but introduces a novel least-squares formulation that minimizes an $H^{-1}$ residual of the governing equations. This formulation ensures stability without requiring large stabilization parameters, thereby eliminating the need for user-tuned penalty parameters and improving the robustness of the computation. Optimal-order convergence of the UnCut FEM solutions is rigorously established in the $H^1$ norm for both the Poisson and Stokes equations, and numerical experiments are presented to support the theoretical analysis.

Figures

Figures reproduced from arXiv: 2608.02015 by the authors.

**Figure 1.** Illustration of the extended domain Ωh, cut cells in Ω, boundary strip Bh := Ωh\{ϕh < 0}, full cut element region ΩΓ h and the edge set F Γ h involving jump penalties. The recently proposed ϕ-FEM [11–14] is an UnCut FEM which offers an appealing alternative that overcomes these integration challenges. In the ϕ-FEM, the domain of the differential equation is extended to Ωh, which consists of all full elements T ∈ Th … view at source ↗

**Figure 2.** Poisson with P 1 and P 2 elements 4.2 Poisson Equation in Three Dimensions In this example, we investigate the convergence behavior of the scheme (2.11) for the threedimensional Poisson problem (1.1). To this end, we consider a spherical computational domain Ω defined by Ω = {(x, y, z) ∈ R 3 : (x − 0.5)2 + (y − 0.5)2 + (z − 0.5)2 < 1 8 }. and utilize a manufactured exact solution uexact = [PITH_FULL_IMAGE:figures/… view at source ↗

**Figure 3.** Convergence results for the 3D Poisson and Stokes problems [PITH_FULL_IMAGE:figures/full_fig_p022_3.png] view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: Sensitivity of the relative velocity H1 -error (left) and pressure L 2 -error (right) to the stabilization parameter σ, for the original ϕ-FEM and the proposed method with h = √ 2/80 and h = √ 2/240. 5 Conclusion We have introduced and analyzed a novel unfitted finite …

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

29 extracted references

[1]

Badia, F

S. Badia, F. Verdugo, and A. F. Martín,The aggregated unfitted finite element method for elliptic problems, Comput. Methods Appl. Mech. Engrg., 336 (2018), pp. 533–553

2018
[2]

Burman, S

E. Burman, S. Claus, P. Hansbo, M. G. Larson, and A. Massing,CutFEM: Discretizing geometry and partial differential equations, Internat. J. Numer. Methods Engrg., 104 (2015), pp. 472–501

2015
[3]

Burman and M

E. Burman and M. A. Fernández,An unfitted nitsche method for incompressible fluid–structure interaction using overlapping meshes, Comput. Methods Appl. Mech. Engrg., 279 (2014), pp. 497–514. 24

2014
[4]

Burman, P

E. Burman, P. Hansbo, and M. Larson,Cut finite element method for divergence- free approximation of incompressible flow: a Lagrange multiplier approach, SIAM J. Numer. Anal., 62 (2024), pp. 893–918

2024
[5]

Burman, P

E. Burman, P. Hansbo, M. Larson, and S. Zahedi,Cut finite element methods, Acta Numer., 34 (2025), pp. 1–121

2025
[6]

Burman, P

E. Burman, P. Hansbo, and M. G. Larson,CutFEM based on extended finite element spaces, Numer. Math., 152 (2022), pp. 331–369

2022
[7]

Comp., 93 (2024), pp

,Low regularity estimates for CutFEM approximations of an elliptic problem with mixed boundary conditions, Math. Comp., 93 (2024), pp. 35–54

2024
[8]

Z. Chen, K. Li, and X. Xiang,An adaptive high-order unfitted finite element method for elliptic interface problems, Numer. Math., 149 (2021), pp. 507–548

2021

Show all 29 references

[9]

Chen and Y

Z. Chen and Y. Liu,An arbitrarily high order unfitted finite element method for elliptic interface problems with automatic mesh generation, J. Comput. Phys., 491 (2023), p. 112384

2023
[10]

Piecewise-smooth interfaces, Appl

,An arbitrarily high order unfitted finite element method for elliptic interface problems with automatic mesh generation, Part II. Piecewise-smooth interfaces, Appl. Numer. Math., 206 (2024), pp. 247–268

2024
[11]

Duprez, V

M. Duprez, V. Lleras, and A. Lozinski,A new ϕ-FEM approach for problems with natural boundary conditions, Numer. Methods Partial Differential Equations, 39 (2023), pp. 281–303

2023
[12]

, ϕ-FEM: an optimally convergent and easily implementable immersed boundary method for particulate flows and Stokes equations, ESAIM Math. Model. Numer. Anal., 57 (2023), pp. 1111–1142

2023
[13]

Duprez, V

M. Duprez, V. Lleras, A. Lozinski, and K. Vuillemot, ϕ-FEM for the heat equation: optimal convergence on unfitted meshes in space, Comptes Rendus. Mathé- matique, 361 (2023), pp. 1699–1710

2023
[14]

Duprez and A

M. Duprez and A. Lozinski, ϕ-FEM: a finite element method on domains defined by level-sets, SIAM J. Numer. Anal., 58 (2020), pp. 1008–1028

2020
[15]

M. A. Fernández and M. Landajuela,Splitting schemes and unfitted-mesh methods for the coupling of an incompressible fluid with a thin-walled structure, IMA J. Numer. Anal., 40 (2019), pp. 1407–1453

2019
[16]

Gürkan, S

C. Gürkan, S. Sticko, and A. Massing,Stabilized cut discontinuous Galerkin methods for advection-reaction problems, SIAM J. Sci. Comput., 42 (2020), pp. A2620– A2654

2020
[17]

Guzmán and M

J. Guzmán and M. Olshanskii,Inf-sup stability of geometrically unfitted Stokes finite elements, Math. Comp., 87 (2018), pp. 2091–2112

2018
[18]

Hansbo and P

A. Hansbo and P. Hansbo,An unfitted finite element method, based on Nitsche’s method, for elliptic interface problems, Comput. Methods Appl. Mech. Engrg., 191 (2002), pp. 5537–5552. 25

2002
[19]

Johansson and M

A. Johansson and M. G. Larson,A high order discontinuous Galerkin Nitsche method for elliptic problems with fictitious boundary, Numer. Math., 123 (2013), pp. 607–628

2013
[20]

Lehrenfeld and M

C. Lehrenfeld and M. Olshanskii,An Eulerian finite element method for PDEs in time-dependent domains, ESAIM Math. Model. Numer. Anal., 53 (2019), pp. 585–614

2019
[21]

H. Liu, M. Neilan, and M. Olshanskii,A cutFEM divergence–free discretization for the Stokes problem, ESAIM Math. Model. Numer. Anal., 57 (2023), pp. 143–165

2023
[22]

C. Ma, Q. Zhang, and W. Zheng,A fourth-order unfitted characteristic finite element method for solving the advection-diffusion equation on time-varying domains, SIAM J. Numer. Anal., 60 (2022), pp. 2203–2224

2022
[23]

Ma and W

C. Ma and W. Zheng,A fourth-order unfitted characteristic finite element method for free-boundary problems, J. Comput. Phys., 469 (2022), p. 111552

2022
[24]

Massing, M

A. Massing, M. G. Larson, A. Logg, and M. E. Rognes,A stabilized Nitsche fictitious domain method for the Stokes problem, J. Sci. Comput., 61 (2014), pp. 604– 628

2014
[25]

N. Moës, E. Béchet, and M. Tourbier,Imposing Dirichlet boundary conditions in the extended finite element method, Internat. J. Numer. Methods Engrg., 67 (2006), pp. 1641–1669

2006
[26]

N. Moës, J. Dolbow, and T. Belytschko,A finite element method for crack growth without remeshing, Internat. J. Numer. Methods Engrg., 46 (1999), pp. 131–150

1999
[27]

Mohammadi,Extended Finite Element Method: For Fracture Analysis of Structures, John Wiley & Sons, Chichester, 2008

S. Mohammadi,Extended Finite Element Method: For Fracture Analysis of Structures, John Wiley & Sons, Chichester, 2008

2008
[28]

Neilan and M

M. Neilan and M. Olshanskii,An Eulerian finite element method for the linearized Navier–Stokes problem in an evolving domain, IMA J. Numer. Anal., 44 (2024), pp. 3234–3258

2024
[29]

Olshanskii and H

M. Olshanskii and H. von W ahl,A conservative Eulerian finite element method for transport and diffusion in moving domains, Comput. Methods Appl. Math., 25 (2025), pp. 961–979. 26

2025

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.

[1] [1]

Badia, F

S. Badia, F. Verdugo, and A. F. Martín,The aggregated unfitted finite element method for elliptic problems, Comput. Methods Appl. Mech. Engrg., 336 (2018), pp. 533–553

2018

[2] [2]

Burman, S

E. Burman, S. Claus, P. Hansbo, M. G. Larson, and A. Massing,CutFEM: Discretizing geometry and partial differential equations, Internat. J. Numer. Methods Engrg., 104 (2015), pp. 472–501

2015

[3] [3]

Burman and M

E. Burman and M. A. Fernández,An unfitted nitsche method for incompressible fluid–structure interaction using overlapping meshes, Comput. Methods Appl. Mech. Engrg., 279 (2014), pp. 497–514. 24

2014

[4] [4]

Burman, P

E. Burman, P. Hansbo, and M. Larson,Cut finite element method for divergence- free approximation of incompressible flow: a Lagrange multiplier approach, SIAM J. Numer. Anal., 62 (2024), pp. 893–918

2024

[5] [5]

Burman, P

E. Burman, P. Hansbo, M. Larson, and S. Zahedi,Cut finite element methods, Acta Numer., 34 (2025), pp. 1–121

2025

[6] [6]

Burman, P

E. Burman, P. Hansbo, and M. G. Larson,CutFEM based on extended finite element spaces, Numer. Math., 152 (2022), pp. 331–369

2022

[7] [7]

Comp., 93 (2024), pp

,Low regularity estimates for CutFEM approximations of an elliptic problem with mixed boundary conditions, Math. Comp., 93 (2024), pp. 35–54

2024

[8] [8]

Z. Chen, K. Li, and X. Xiang,An adaptive high-order unfitted finite element method for elliptic interface problems, Numer. Math., 149 (2021), pp. 507–548

2021