Pith. sign in

REVIEW 4 major objections 5 minor 2 cited by

Spiral renormalization group flow and universal entanglement spectrum of the non-Hermitian 5-state Potts model

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read Ordinary tensor-network algorithms can simulate the non-Hermitian 5-state Potts model at its complex fixed point, yielding the spiral RG flow and a boundary CCFT spectrum.

desk verdict A careful numerical study that delivers the first lattice spiral-flow observation and a refined critical coupling, with honest limitations that a referee should push on but that do not sink the paper. read the letter →

arxiv 2507.14732 v3 pith:VA5HA64Q submitted 2025-07-19 cond-mat.str-el quant-ph

Vic Vander Linden , Boris De Vos , Kevin Vervoort , Frank Verstraete , Atsushi Ueda This is my paper · ORCID
classification cond-mat.str-elquant-ph
keywords non-Hermitianquantumspinchains5-statePottsmodelcomplexconformalfieldtheoryfixedpointsmatrixproductstatesDMRGrenormalizationgroupflowentanglementspectrum
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

The quantum 5-state Potts model's weakly first-order transition is believed to be controlled by a pair of complex conformal field theory (CCFT) fixed points that cannot be reached by tuning real couplings alone. To realize them on the lattice, one adds a non-Hermitian term with a complex coefficient $\lambda$, which destroys the variational principle that normally guarantees tensor-network accuracy. This paper argues that in the 'walking' regime, where RG flow slows and the non-Hermitian deformation remains small, ordinary DMRG and quasiparticle-ansatz methods still converge to the correct ground and excited states. With systems up to $L=28$, the authors refine the critical point to $\lambda_c = 0.0788 + 0.0603i$, observe the theoretically predicted spiral flow of the running coupling $g_{\epsilon'}$, and recover an entanglement spectrum that matches the free-free boundary CCFT. If correct, this turns complex fixed points from an analytic curiosity into a numerically accessible window on weakly first-order transitions.

What carries the argument

The workhorse is the running coupling $g_{\epsilon'}$ of the leading perturbing operator in the CCFT description. It is recovered from finite-size energy data by a cost function built from conformal perturbation theory, $\delta E_{\phi} = 2\pi g_{\epsilon'} C_{\phi\phi\epsilon'}$, with the descendant shift fixed by the scaling dimension $\Delta_{\epsilon'}$. Plotting $g_{\epsilon'}$ as the chain grows shows the spiral RG flow. For entanglement, the central object is the non-Hermitian density matrix $\rho_{RL} = |\psi_L\rangle\langle\psi_R|/\langle\psi_R|\psi_L\rangle$ and the associated reduced density matrix $\rho^A_{RL}$ built from the singular value decompositions of the right and left ground states; the paper proposes that its entanglement Hamiltonian satisfies $\rho_{RL} = e^{-2\pi K_C}$. The MPS ansatz's singular values serve as the truncation criterion, justified here by the smallness of the non-Hermitian deformation.

What would settle it

A direct test would be to compute, at $\lambda_c$ and $L=24$, the full eigenvalue spectrum of the reduced density matrix $\rho_{RL}^A$ (using ED for $L\lesssim 12$ and a much larger bond dimension on longer chains) and compare it with the singular-value ordering used in the MPS truncation: if the largest discarded singular value corresponds to an eigenvalue whose real or imaginary part is comparable to the smallest level retained in the fits, the MPS ground state is missing essential entanglement. Alternatively, if the fitted minimum of the cost function $J(v,g_{\epsilon'})$ stops moving toward $\lambda_c = 0.0788 + 0.0603i$ as $L$ grows beyond 24 and instead reverses its $1/L^2$ drift, the refined critical point is an artifact of finite-size extrapolation.

Watch

Extended reading notes

Core claim

On the paper's terms, the discovery is that the small non-Hermiticity of the deformed 5-state Potts Hamiltonian makes the MPS singular-value truncation nearly optimal even though the variational principle is invalid, so the unadapted DMRG and QPA reach algorithmic tolerance ($10^{-6}$) for the ground state at $\lambda_c = 0.0788 + 0.0603i$. With the refined coupling, subleading finite-size scaling gives central charge $c = 1.1375 - 0.0211i$ and conformal dimensions consistent with the theoretical CCFT data; the cost-function fit yields the logarithmic-spiral RG flow of $g_{\epsilon'}$ around the two conjugate fixed points; and the entanglement Hamiltonian extracted from $\rho_{RL}^A$ reproduces the free-free boundary CCFT operator content (towers with degeneracies 1, 4, and 11). The failure modes are acknowledged: imaginary parts converge more slowly, the density matrix is not positive semi-definite, and the MPS ordering property is only approximate.

Load-bearing premise

The load-bearing premise is that the MPS truncation based on the singular values of the right ground-state eigenvector preserves the ordering of the true non-Hermitian entanglement spectrum well enough to neglect the discarded states, even though the variational principle no longer guarantees this and the paper flags in Supplementary Section E that the disconnect between singular values and eigenvalues leaves the neglected entanglement unknown.

Editorial extensions

If this is right

  • Standard DMRG and QPA can be used for other slightly non-Hermitian CCFT lattice models without biorthonormal adaptations, with errors controlled by algorithmic tolerance.
  • The refined $\lambda_c = 0.0788 + 0.0603i$ improves the predicted conformal data beyond exact diagonalization at the previous estimate, because subleading $1/L^2$ finite-size effects are included.
  • The observed spiral flow of $g_{\epsilon'}$ in a 1+1D lattice system confirms the theoretical picture of complex fixed points as the origin of walking and quasi-critical behavior.
  • The entanglement spectrum matching the free-free boundary CCFT (degeneracies 1, 4, 11) supports the proposed relation $\rho_{RL} = e^{-2\pi K_C}$ for complex fixed points.
  • Weakly first-order transitions with hidden complex fixed points become directly addressable by tensor networks, not only through analytic continuation of theory space.

Reading between the lines

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

  • A practical criterion suggested by the paper: unadapted MPS methods should work when the imaginary part of the complex fixed-point data is small compared with the real part; more strongly non-Hermitian CCFTs (such as O(N>2) sigma models) should require biorthonormal or fidelity-based truncation.
  • The same entanglement-spectrum diagnostic could be applied to other candidate weakly first-order transitions, such as the Néel–valence-bond-solid transition, where characteristic boundary degeneracies could reveal a complex fixed point even without an explicit non-Hermitian lattice deformation.
  • The anomalous ~$1/L^{1.05}$ scaling of Im $\Delta_\epsilon$ is a candidate systematic artifact: checking whether it survives in a biorthonormal DMRG calculation at fixed $\lambda_c$ would separate a true OPE effect from the MPS truncation bias in imaginary entanglement.
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.

Referee Report

4 major / 5 minor

Summary. The manuscript studies the non-Hermitian deformation of the 5-state Potts model proposed by Tang et al. and claims that standard MPS methods (DMRG and QPA), despite the breakdown of the variational principle, faithfully capture the complex fixed point. The authors determine a refined critical coupling λ_c = 0.0788 + 0.0603i by finite-size scaling of conformal dimensions, extract bulk conformal data, observe a spiral RG flow of the coupling g_ε′ in agreement with the CCFT prediction, and reconstruct a boundary entanglement spectrum matching the free-free conformal boundary condition. The paper includes an extensive supplementary discussion of error sources and a data-collapse cross-check of the fixed point.

Significance. If the central claims are correct, this is a valuable demonstration that tensor network algorithms can simulate non-Hermitian deformations of weakly first-order transitions at complex fixed points, going beyond exact diagonalization and providing numerical evidence for the CCFT scenario of the 5-state Potts model. The paper is unusually candid in documenting the limitations of the non-Hermitian MPS approximation, and the data-collapse check in Supplement B is a partial independent validation. However, the headline results inherit two load-bearing issues: the uncontrolled truncation of the non-Hermitian density matrix and the use of theory-derived conformal data in the fixed-point determination while simultaneously comparing extracted conformal data with those same inputs.

major comments (4)
  1. [Tensor networks / Supp. Sec. E] The central computational assumption is that truncating the right eigenvector |ψ_R⟩ by its singular values yields a controlled approximation to the non-Hermitian ground state. The manuscript itself states in Sec. E that "due to the observed disconnect between singular and eigenvalues, it remains unknown how much entanglement is actually being neglected by the MPS approximation." Because the variational principle is invalid, DMRG provides no error bound, and the ED comparison at L=12 and bond-dimension saturation at L=24 do not validate the L=64 entanglement data or the QPA excited states used in the RG flow. This affects all three headline claims. The authors should quantify the truncation error (for example by comparing against a biorthonormal DMRG variant or by presenting convergence in the entanglement spectrum with respect to discarded weight in ρ_RL), or explicitly restrict claims to system sizes where an independent validation exists.
  2. [Conformal data / Eq. (4) / Supp. Sec. B] The fixed-point search uses theoretical scaling dimensions and OPE coefficients as input: Eq. (4) defines δΔ_n relative to Δ_th^n, and Eq. (5) minimizes J(v,g_ε′) built from the theoretical C_ϕϕε′. The extracted conformal data in Table I are then compared with those same theoretical values. This is a consistency check, not an independent verification. The data collapse in Supp. Sec. B is a genuinely independent confirmation for the σ operator, which strengthens the paper, but it is not used for the main quoted precision and does not cover the other operators. Please reframe the conformal-data section accordingly and, if possible, provide a quantitative measure of the degree of independence between the input theory and the extracted numbers.
  3. [Fig. 3 / Table I] The paper reports that Im(Δ_ε) scales as 1/L^1.05 rather than the expected 1/L^2 and states that "it remains unclear whether this observation has a deeper meaning or if it is due to the same simulation artifacts." Nevertheless, Table I lists Im(Δ_ε) = −0.2245(1)i fitted with 1/L^2. This unresolved discrepancy means the quoted precision of the imaginary part of ε is not supported by the presented scaling analysis. Either fit with the observed power law and state the resulting uncertainty, or identify the origin of the anomalous exponent.
  4. [Complex entanglement spectrum / Eq. (7)] The identification ρ_RL = e^{−2πK_C} is assumed rather than derived, and the paper acknowledges that ρ_RL is not positive semi-definite. The agreement with the free-free boundary CCFT is argued from degeneracies and the first two fitted levels of the zero sector; however, the imaginary parts converge more slowly and no error estimates are given for the extracted levels. Please provide a quantitative comparison (for example, residuals for all identified towers) and discuss how the non-positive semi-definiteness of ρ_RL affects the logarithm in Eq. (7).
minor comments (5)
  1. [Introduction] The symbols C and C are introduced only through the phrase "fixed points C and C"; since the conjugate fixed point is denoted in several places by an overline, this notation should be defined explicitly at first use.
  2. [Conformal data] In the sentence "The best subleading scaling for the real and imaginary part of the six most relevant operators is visible in Fig. 3," the six operators are not enumerated in the main text; please list them explicitly.
  3. [Tensor networks] The statement "we use χ=400 and χ=600 for the ground state and excitation calculations" should specify which value applies to which algorithm and whether the same bond dimensions are used for the L=64 entanglement spectrum.
  4. [Conformal data] The phrase "the maximal reach of ED methods L∼12 barely captures the right physics due to the logarithmic finite-size effects" would benefit from a reference or a quantitative definition of the logarithmic corrections.
  5. [Conformal data / Supplementary] Equation (S.5) is cited in the main text before the supplementary material is introduced; please ensure the cross-referencing is clear to the reader.

Circularity Check

2 steps flagged · score 4.0 of 10

Fixed-point search and running-coupling coordinate both import the CCFT's own conformal data, so the reported lambda_c refinement, Table I conformal dimensions, and spiral-flow agreement are partly built in; independent ED checks and energy-gap collapse keep the central tensor-network claims non-circular.

  1. fitted input called prediction [Main text, 'Conformal data' section, Eq. (4) and Table I; SI Sec. B]
    "Now we evaluate the deviation between the leading order contribution to Δn in Eq. (S.5) and the theoretical value Δth n [9]: δΔn = L/2πv (En − E0) − Δth n . By analyzing the scaling of these values for different system sizes, we recover the fixed point as the value for λ where we observe the correct 1/L2 scaling."

    lambda_c is selected as the point where delta_Delta_n = (L/2*pi*v)(E_n - E_0) - Delta_th^n scales as 1/L^2, i.e., where the measured gaps approach the theoretical dimensions Delta_th^n of [9] at the expected rate. The conformal data in Table I are then fitted at this same lambda_c and presented as 'more accurate estimates for the conformal dimensions' that agree with [9]; the real-part agreement is therefore largely a restatement of the selection criterion rather than an independent prediction. The authors concede the theory-dependence in SI Sec. B: 'The downside to either the cost function Eq.

  2. fitted input called prediction [Main text, 'RG flow' section, Eq. (5)]
    "Using the cost function proposed in [11], J(v, gε′) = Σ n∈{ϕ,∂ϕ} |δΔn −δE n|, δEϕ = 2πgε′Cϕϕε′, δE ∂ϕ = δE ϕ (1 + Δε′(Δε′ −2)/4Δϕ), we systematically recover the parameter gε′ for different system sizes and λ values in theory space. ... We recognize the theoretical cyclic behavior of the RG flow."

    g_epsilon' is extracted by fitting measured energy deviations to the theory's own first-order perturbation formula with OPE coefficient C_phi_phi_epsilon' and dimension Delta_epsilon' taken from [9]. The coupling is thus a measurement in the CCFT's own coordinates, so the subsequent statement 'We recognize the theoretical cyclic behavior of the RG flow' is partly a consistency test of the assumed operator data rather than an independent derivation of the flow. The spiral trajectory itself is empirical (the cost function does not force a spiral), so the circularity is mild: the coordinate system is theory-defined, but the flow topology, the winding direction, and the near-zero flow at the blue-dot fixed point are data-driven content.

full rationale

The paper's central methodological claim — that unadapted DMRG/QPA faithfully simulate the weakly non-Hermitian deformed 5-state Potts Hamiltonian — is independently supported: ground-state and excitation energies agree with exact diagonalization at L=12 (Fig. 2), bond-dimension saturation is checked at L=24, and the energy-gap collapse at the conjugate pair of fixed points (SI Fig. S2) cross-checks lambda_c. Those results are not circular. The circularity is partial and confined to two conformal-data channels. (1) The refined fixed point lambda_c = 0.0788 + 0.0603i is located by demanding delta_Delta_n = (L/2*pi*v)(E_n - E_0) - Delta_th^n (Eq. 4) scale as 1/L^2 with the theoretical Delta_th^n of [9] as target; Table I then reports the conformal dimensions extracted at this lambda_c as 'more accurate estimates' agreeing with [9]. The real-part agreement is substantially a restatement of the selection criterion; the authors themselves flag this in SI Sec. B ('these methods ... depend on theoretical predictions of the CCFT data'). Residual independent content exists: one lambda_c must clean up six operators simultaneously, the imaginary parts do not fully comply (Im(Delta_epsilon) requires a 1/L^1.05 fit), and the conjugate collapse plus the proximity to the ED-based lambda_prev_c of [11] anchor the value externally. (2) The running coupling g_epsilon' of cost function (5) is defined using the theory's own OPE coefficient C_phi_phi_epsilon' and dimension Delta_epsilon', so the observed spiral flow is a measurement in theory coordinates; the spiral topology itself is empirical and not forced. The entanglement-spectrum claim is benchmarked against the external boundary-CCFT spectrum of [31] and is robust to the fixed-point choice (SI Sec. D), and no load-bearing self-citation chain is present: the theory inputs [9, 11, 22, 31] are all external to the present authors, with self-citations to [12] merely deferring to the paper's own SI error analysis. Finally, SI Sec. E's concession that 'it remains unknown how much entanglement is actually being neglected by the MPS approximation' is a flagged evidence gap in the truncation premise, not a circular step; it bears on correctness rather than on circularity and does not increase the score. Overall: partial circularity on the conformal-data and coupling-flow claims, central tensor-network results retaining independent content — score 4.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The paper introduces no new particles or fields. The key assumptions are the modeling choice of the non-Hermitian deformation, the validity of Hermitian MPS algorithms for a mildly non-Hermitian problem, and the proposed relation between the non-Hermitian density matrix and the entanglement Hamiltonian. The free parameters are the refined critical coupling, the speed of light, the fitted running coupling, and the fitted mapping to the walking parameter.

free parameters (4)
  • lambda_c = 0.0788 + 0.0603i
    Critical coupling of the non-Hermitian deformation, refined by finite-size scaling with subleading corrections. This is the central parameter of the paper.
  • speed of light v = 2.8812 + 0.7050i
    Non-universal velocity fitted from ground-state energies via Eq. (S.5) and then refined to improve the scaling; used to convert energy gaps to scaling dimensions.
  • running coupling g_epsilon' at each (lambda, L) = varies
    Obtained by minimizing the cost function J(v, g_epsilon') in Eq. (5), which uses theoretical OPE coefficients C_{phi phi epsilon'}; plotted to reveal the spiral flow.
  • mapping parameters x and y = x = 0.076 - 0.015i, y = 0.245 - 0.057i
    Supplement Section C: fitted so that real lambda values map to real lambda_w values, used to construct the walking parameter flow.
assumptions (5)
  • domain assumption The extension term H_1 (Eq. S.1) is the unique nearest-neighbor, S_Q-symmetric, Kramers-Wannier-duality-invariant deformation that realizes the CCFT perturbation.
    Supplement Section A argues uniqueness among nearest-neighbor terms; other terms such as H_2 exist when range is extended, so the identification is a modeling assumption.
  • domain assumption The weakly first-order Q=5 Potts transition is governed by complex fixed points from the analytic continuation of the Coulomb gas.
    Taken from Gorbenko et al. [9]; the entire analysis presupposes this CCFT scenario.
  • ad hoc to paper The non-Hermitian deformation is small enough that standard DMRG truncation based on right-eigenvector singular values is accurate.
    Tensor networks section and Supplement Section E; explicitly acknowledged as an assumption, with the caveat that the amount of neglected entanglement is unknown.
  • ad hoc to paper The non-Hermitian density matrix relates to the entanglement Hamiltonian as rho_RL = e^{-2 pi K_C} (Eq. 7).
    Proposed by the authors in the Complex entanglement spectrum section; no proof is given for the complex case.
  • domain assumption The MPS ground state is the eigenstate with the smallest real part and has exponentially decaying Schmidt coefficients.
    Defined and assumed in the Tensor networks section; a priori not guaranteed for non-Hermitian critical systems.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Spiral renormalization group flow and universal entanglement spectrum of the non-Hermitian 5-state Potts model." pith.science (2026). https://pith.science/paper/VA5HA64Q

@misc{pith2026250714732,
  author       = {Pith},
  title        = {Pith review of: Spiral renormalization group flow and universal entanglement spectrum of the non-Hermitian 5-state Potts model},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VA5HA64Q}},
  note         = {Machine review of arXiv:2507.14732}
}
abstract

The quantum $5$-state Potts model is known to possess a perturbative description using complex conformal field theory (CCFT), the analytic continuation of ``theory space" to a complex plane. To study the corresponding complex fixed point on the lattice, the model must be deformed by an additional non-Hermitian term due to its complex coefficient $\lambda$. Although the variational principle breaks down in this case, we demonstrate that tensor network algorithms are still capable of simulating these non-Hermitian theories. We access system sizes up to $L = 28$, which enable the observation of the theoretically predicted spiral flow of the running couplings. Moreover, we reconstruct the full boundary CCFT spectrum through the entanglement Hamiltonian encoded in the ground state. Our work demonstrates how tensor networks are the correct approach to capturing the approximate conformal invariance of weakly first-order phase transitions.

Figures

Figures reproduced from arXiv: 2507.14732 by the authors.

Figure 2
Figure 2. FIG. 2. The difference between the ground state energy [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. The 1 [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figure 4
Figure 4. FIG. 4. The fifteen largest eigenvalues [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figures from the paper (1 more)
Figure 5
Figure 5. Figure 5: FIG. 5. The entanglement spectrum of the ground state in the (a) zero sector and (b) a charged sector of [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Complex Conformal Manifolds

    hep-th 2026-06 unverdicted novelty 7.0 of 10

    Complexified exactly-marginal couplings produce solvable complex CFTs, with the Ising defect verified numerically in non-Hermitian chains.

  2. Extracting Boundary Conformal Data from Periodic Non-Hermitian Critical Chains

    cond-mat.stat-mech 2026-06 conditional novelty 7.0 of 10

    Periodic-chain projected overlaps with paired left duals extract universal boundary CFT coefficients, including a negative Yang-Lee ratio and complex Potts boundary data.

Reference graph

Works this paper leans on

46 extracted references · 20 canonical work pages · cited by 2 Pith papers

  1. [1]

    D. B. Kaplan, J.-W. Lee, D. T. Son, and M. A. Stephanov, Physical Review D80, 10.1103/phys- revd.80.125005 (2009)

    doi:10.1103/phys- 2009

  2. [2]

    Corboz and F

    P. Corboz and F. Mila, Phys. Rev. B87, 115144 (2013)

    work page 2013

  3. [3]

    N. Xi, H. Chen, Z. Y. Xie, and R. Yu, Phys. Rev. B107, L220408 (2023)

    work page 2023

  4. [4]

    Koga and N

    A. Koga and N. Kawakami, Phys. Rev. Lett.84, 4461 (2000)

    work page 2000

  5. [5]

    J. Y. Lee, Y.-Z. You, S. Sachdev, and A. Vishwanath, Phys. Rev. X9, 041037 (2019)

    2019

  6. [6]

    Naichuk, J

    E. Naichuk, J. van den Brink, and F. S. Nogueira, Physi- cal Review B110, 10.1103/physrevb.110.224505 (2024)

    work page doi:10.1103/physrevb.110.224505 2024

  7. [7]

    Ma and Y.-C

    H. Ma and Y.-C. He, Physical Review B99, 10.1103/physrevb.99.195130 (2019)

    work page doi:10.1103/physrevb.99.195130 2019

  8. [8]

    Kumar, S

    S. Kumar, S. Pujari, and J. D’Emidio, Pseudo- criticality in antiferromagnetic spin chains (2025), arXiv:2506.16424 [cond-mat.str-el]

    arXiv 2025

Show all 46 references
  1. [10]

    Gorbenko, S

    V. Gorbenko, S. Rychkov, and B. Zan, Journal of High Energy Physics2018, 10.1007/jhep10(2018)108 (2018)

    2018 doi

  2. [12]

    See supplementary material

  3. [13]

    J. I. Cirac, D. P´ erez-Garc ´ ıa, N. Schuch, and F. Ver- straete, Reviews of Modern Physics93, 10.1103/revmod- phys.93.045003 (2021)

    2021 doi

  4. [14]

    Schollw¨ ock, Reviews of modern physics77, 259 (2005)

    U. Schollw¨ ock, Reviews of modern physics77, 259 (2005)

    2005

  5. [15]

    Verstraete, V

    F. Verstraete, V. Murg, and J. I. Cirac, Advances in physics57, 143 (2008)

    2008

  6. [16]

    Ashida, Z

    Y. Ashida, Z. Gong, and M. Ueda, Advances in Physics 69, 249–435 (2020)

    2020

  7. [17]

    Verstraete and J

    F. Verstraete and J. I. Cirac, Physical Review B73, 10.1103/physrevb.73.094423 (2006)

    2006 doi

  8. [19]

    Van Damme, R

    M. Van Damme, R. Vanhove, J. Haegeman, F. Ver- straete, and L. Vanderstraeten, Phys. Rev. B104, 115142 (2021)

    2021

  9. [20]

    Haegeman, S

    J. Haegeman, S. Michalakis, B. Nachtergaele, T. J. Os- borne, N. Schuch, and F. Verstraete, Phys. Rev. Lett. 111, 080401 (2013)

    2013

  10. [21]

    Pirvu, G

    B. Pirvu, G. Vidal, F. Verstraete, and L. Tagliacozzo, Phys. Rev. B86, 075117 (2012)

    2012

  11. [23]

    Devos and J

    L. Devos and J. Haegeman, Tensorkit.jl: A julia pack- age for large-scale tensor computations, with a hint of category theory (2025), arXiv:2508.10076 [cs.MS]

    2025 arXiv

  12. [24]

    Van Damme, L

    M. Van Damme, L. Devos, and J. Haegeman, Mpskit (2025)

    2025

  13. [25]

    J. L. Cardy, Nuclear Physics B270, 186 (1986)

    1986

  14. [26]

    Affleck, Phys

    I. Affleck, Phys. Rev. Lett.56, 746 (1986)

    1986

  15. [27]

    H. W. J. Bl¨ ote, J. L. Cardy, and M. P. Nightingale, Phys. Rev. Lett.56, 742 (1986)

    1986

  16. [30]

    Cardy and E

    J. Cardy and E. Tonni, Journal of Statistical Mechanics: Theory and Experiment2016, 123103 (2016)

    2016

  17. [32]

    J. J. Bisognano and E. H. Wichmann, J. Math. Phys.16, 985 (1975)

    1975

  18. [33]

    Ohmori and Y

    K. Ohmori and Y. Tachikawa, Journal of Statistical Me- chanics: Theory and Experiment2015, P04010 (2015)

    2015

  19. [34]

    A. M. L¨ auchli, Operator content of real-space entan- glement spectra at conformal critical points (2013), arXiv:1303.0741 [cond-mat.stat-mech]

    2013 arXiv

  20. [35]

    W. Tang, F. Verstraete, and J. Haegeman, Physical Re- view B111, 10.1103/physrevb.111.035107 (2025)

    2025 doi

  21. [36]

    Saleur and M

    H. Saleur and M. Bauer, Nuclear Physics B320, 591 (1989). 7 Supplementary Material CONTENTS A. Extension Term 7 B. Fixed Point Precision 7 C. Walking RG Flow 10 D. Open Boundary Conditions 11 E. Sources of error 11 References 14 A. EXTENSION TERM In this section, we investigat...

    1989

  22. [37]

    Gorbenko, S

    V. Gorbenko, S. Rychkov, and B. Zan, SciPost Physics5, 10.21468/scipostphys.5.5.050 (2018)

    2018 doi

  23. [38]

    Y. Tang, H. Ma, Q. Tang, Y.-C. He, and W. Zhu, Phys. Rev. Lett.133, 076504 (2024)

    2024

  24. [39]

    Cobanera, G

    E. Cobanera, G. Ortiz, and Z. Nussinov, Phys. Rev. Lett.104, 020402 (2010)

    2010

  25. [40]

    F. C. Alcaraz, Physical Review B94, 10.1103/physrevb.94.115116 (2016)

    2016 doi

  26. [41]

    Huang, L

    R.-Z. Huang, L. Zhang, A. M. L¨ auchli, J. Haegeman, F. Verstraete, and L. Vanderstraeten, Phys. Rev. Lett.132, 086503 (2024)

    2024

  27. [42]

    J. T. Schneider, A. Ueda, Y. Liu, A. L¨ auchli, M. Oshikawa, and L. Tagliacozzo, SciPost Physics18, 10.21468/scipost- phys.18.4.142 (2025)

    2025 doi

  28. [43]

    Reinicke, Journal of Physics A: Mathematical and General20, 5325 (1987)

    P. Reinicke, Journal of Physics A: Mathematical and General20, 5325 (1987)

    1987

  29. [44]

    Lao and S

    B.-X. Lao and S. Rychkov, SciPost Physics15, 10.21468/scipostphys.15.6.243 (2023)

    2023 doi

  30. [45]

    D. B. Kaplan, J.-W. Lee, D. T. Son, and M. A. Stephanov, Physical Review D80, 10.1103/physrevd.80.125005 (2009)

    2009 doi

  31. [46]

    Haldar, O

    A. Haldar, O. Tavakol, H. Ma, and T. Scaffidi, Physical Review Letters131, 10.1103/physrevlett.131.131601 (2023)

    2023 doi

  32. [47]

    Shimizu and K

    H. Shimizu and K. Kawabata, Complex entanglement entropy for complex conformal field theory (2025), arXiv:2502.02001 [cond-mat.stat-mech]

    2025 arXiv

  33. [48]

    Y. Tang, Q. Liu, Q. Tang, and W. Zhu, SciPost Phys.19, 164 (2025), arXiv:2512.07625 [cond-mat.stat-mech]

    2025

  34. [49]

    Yang and T

    C. Yang and T. Scaffidi, Asymptotic freedom, lost: Complex conformal field theory in the two-dimensionalo(n >2) nonlinear sigma model and its realization in the spin-1 heisenberg chain (2026), arXiv:2601.02459 [cond-mat.stat-mech]

    2026 arXiv

  35. [50]

    Zhong, W

    P. Zhong, W. Pan, H. Lin, X. Wang, and S. Hu, Phys. Rev. Lett.135, 106502 (2025)

    2025

  36. [51]

    Yamamoto, M

    K. Yamamoto, M. Nakagawa, M. Tezuka, M. Ueda, and N. Kawakami, Phys. Rev. B105, 205125 (2022)

    2022

  37. [52]

    S. R. White, Phys. Rev. Lett.69, 2863 (1992)

    1992

  38. [53]

    Haegeman, S

    J. Haegeman, S. Michalakis, B. Nachtergaele, T. J. Osborne, N. Schuch, and F. Verstraete, Phys. Rev. Lett.111, 080401 (2013)

    2013

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.