Pith. sign in

REVIEW 3 major objections 5 minor 2 cited by

Monte Carlo Simulation of the $SU(2)/\mathbb{Z}_2$ Yang--Mills Theory

T0 review · 3 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read This paper claims that promoting the Z2 't Hooft flux to a dynamical variable in hybrid Monte Carlo simulations of SU(2)/Z2 Yang–Mills theory drastically reduces topological-charge autocorrelation, offering a possible substitute for…

desk verdict Clean methodological idea with a correct detailed-balance proof and an honest exploratory study; the SU(2)/Z2-to-SU(2) equivalence is plausible but not yet demonstrated. read the letter →

arxiv 2501.00286 v4 pith:2TNEMF4Q submitted 2024-12-31 hep-lat hep-th

Motokazu Abe , Okuto Morikawa , Hiroshi Suzuki This is my paper · ORCID
classification hep-lathep-th
keywords latticeYang-MillsSU(2)/Z2hybridMonteCarlotopologicalchargefreezing'tHooftfluxZ2two-formgaugefieldautocorrelation
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 tests a way around topological freezing, the well-known slowdown of hybrid Monte Carlo (HMC) simulations of SU(2) Yang–Mills theory as the lattice gets finer. Instead of simulating SU(2) directly, the authors simulate the quotient theory SU(2)/Z2, promoting the Z2 't Hooft flux (a flat Z2 two-form gauge field) to a dynamical variable that is randomly refreshed during the HMC update. They report that the autocorrelation time of the topological charge and of the gradient-flow energy operator drops dramatically compared with ordinary SU(2) HMC. The hope is that, because local observables in SU(2)/Z2 should equal those of SU(2) at large volume, this recipe can serve as a cheaper alternative simulation of SU(2) physics.

What carries the argument

The central object is the dynamical Z2 B-field: a flat Z2-valued two-form gauge field on the lattice, taking the particular form of 't Hooft fluxes and updated by a uniform random choice in the middle of each HMC trajectory. It carries the argument because the random choice shuffles the topological charge as Q = -1/2 * ε B B /8 + Z, so the simulation visits topological sectors rapidly. The 'halfway-updating' HMC algorithm, with two half-trajectories separated by the B-field update, and its detailed-balance proof turn this shuffling into a correct Markov process.

What would settle it

Take the SU(2)/Z2 ensembles at β=2.6, L=20 and a finer lattice, measure the topological susceptibility, and extrapolate $χ_t^{{1/4}}$/√σ to the continuum and infinite volume; if the result does not approach 0.486(10) (the SU(2) value from Ref. [34]) within errors, the substitution fails.

Watch

Extended reading notes

Core claim

In the theory's own terms: the SU(2)/Z2 Yang–Mills partition function can be simulated by HMC in which the flat Z2 B-field is one of the dynamical variables, and doing so removes the topological-sector barrier that traps conventional SU(2) HMC. The paper's 'halfway-updating' HMC algorithm updates the gauge field for half a trajectory, refreshes B uniformly, then completes the trajectory, and the authors prove detailed balance for this update. The topological charge then takes fractional values Q = 1/2 + Z and its HMC history wanders freely, with integrated autocorrelation times far below those of SU(2); the same holds for the energy operator E(t).

Load-bearing premise

The load-bearing premise is that local observables in SU(2)/Z2 match those of SU(2) at the same physical volume in the large-volume limit, since the paper relies on that equality to turn a quotient-theory simulation into a substitute for SU(2) simulation.

Editorial extensions

If this is right

  • If the large-volume equivalence holds, SU(N)/Z_N HMC can replace SU(N) HMC for local observables, eliminating the topological-freezing cost.
  • The approach extends to SU(3) QCD once fundamental quarks are included via the proposed gauged baryon-number construction with a Stückelberg mass for the extra U(1) gauge boson.
  • The random B-field update also shortens autocorrelations of the gradient-flow energy operator E(t), not just Q, indicating a general cure for slow modes.
  • Topological susceptibility can be continuum-extrapolated from quotient-theory ensembles with smaller errors per trajectory.
  • The method pairs with gradient-flow smearing to give fractional topological charges Q = 1/2 + Z, matching the expected fractionalization in the SU(2)/Z2 theory.

Reading between the lines

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

  • A natural stress test is to check whether the autocorrelation gain persists on finer lattices and larger volumes than those tried here; the paper's own data show a small slope but limited statistics.
  • The proposed quark incorporation introduces a dynamical U(1) gauge field that must decouple, so a direct dynamical-fermion simulation would settle whether the Stückelberg mechanism works at finite lattice spacing.
  • The same 'dynamical discrete flux' trick might apply to other theories with 1-form symmetries, such as SU(3) pure gauge, or to other sampling strategies beyond HMC.
  • If the B-dependent gradient flow turns out not to be renormalizable, the fractional-charge definition used here would need replacement, but the autocorrelation benefit itself does not depend on that definition.
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

3 major / 5 minor

Summary. The paper introduces a hybrid Monte Carlo (HMC) algorithm for SU(2)/Z2 Yang–Mills theory on the lattice in which the Z2 2-form flat gauge field (the 't Hooft flux, or B-field) is treated as a dynamical variable. In this "halfway-updating" HMC, the gauge field is evolved for half a trajectory with the current B-field, the B-field is refreshed by a symmetric random update, and the gauge field is evolved for the second half with the new B-field; Appendix A proves detailed balance. The authors report that this algorithm drastically reduces the autocorrelation time of the topological charge Q and of a flowed energy operator E(t) compared with conventional HMC in SU(2), and they present an exploratory continuum extrapolation of the topological susceptibility that is consistent with the SU(2) result within large errors. The paper also sketches a method for including fundamental fermions by gauging baryon number U(1) and giving the unwanted U(1)_B gauge boson a Stückelberg mass.

Significance. If the underlying assumptions hold, this is a potentially important algorithmic idea: simulating SU(N)/Z_N instead of SU(N) could bypass topological freezing by letting random updates of the B-field shuffle the topological charge. The halfway HMC construction is simple, the detailed-balance proof in Appendix A is standard and appears correct, and the public code (footnote 3) is a strength. The reduction in the autocorrelation of Q is clearly visible in the histories and autocorrelation functions. However, the evidence that local observables in SU(2)/Z2 agree with SU(2) in the large-volume limit is preliminary, and the B-dependence of the probe observables weakens the claim that the reduction extends to ordinary physical quantities independent of the B-shuffling mechanism. The paper is an exploratory study with an openly stated open problem (renormalizability of the B-dependent flow), and its central conclusion is conditional.

major comments (3)
  1. [Sec. 3.2, Figs. 7–8] The energy-operator E(t) is not an independent confirmation that autocorrelations of physical observables are reduced. Its definition multiplies every plaquette in the flow and in the observable by the B-field (see the text after Eq. (3.2) and footnote 5), so a random B update directly changes the flowed configuration and hence E(t). The observed short autocorrelation time of O(20) MD time therefore has a large component that is forced by the B-shuffling built into the update, just as for Q. In addition, the flow time t=(0.7L)^2/8 corresponds to a smearing radius of 0.7L, so E(t) is a volume-averaged rather than local quantity. The paper should complement this with a genuinely local and B-invariant or B-independent observable, for example an adjoint Wilson loop or a short-flow-time action density measured at a fixed physical flow time, compared between SU(2) and SU(2)/Z2 at the same β and L.
  2. [Sec. 3.2, Figs. 9–11] The claim that SU(2)/Z2 simulation can serve as an alternative to SU(2) simulation rests on the large-volume equivalence of local observables, and the present evidence is not yet sufficient. The comparison is limited to two physical volumes with La√σ ≈ 2.1 and 2.6, and the infinite-volume limit is taken by a naive linear extrapolation of central values in 1/(La√σ) with no systematic error and no control of exponential finite-volume corrections. Moreover, the scale setting for the SU(2)/Z2 theory uses the SU(2) β–a√σ relation from Ref. [34], although the Wilson line is not gauge invariant in SU(2)/Z2; a scale mismatch would directly misalign the volumes being compared. The authors' own caveat that 'we need further statistics for larger and finer lattices' should be treated as a central limitation of the paper rather than a closing remark.
  3. [Sec. 3.2, footnote 5] Both Q and E(t) are defined through a gradient flow whose equation of motion depends on the discrete B-field, and footnote 5 correctly states that the renormalizability of this B-dependent flow is an open problem. The paper also does not study the dependence of its results on the flow time; t=(0.7L)^2/8 is chosen for all lattices, so the smearing radius is a fixed fraction of the volume. Because the central numerical evidence consists of the autocorrelation of Q and the continuum extrapolation of the topological susceptibility χ_t, this unresolved issue is load-bearing. At minimum, the authors should provide a flow-time independence check for Q and χ_t and discuss whether the fractional-value structure and the autocorrelation reduction persist for other flow times.
minor comments (5)
  1. [Sec. 3.1, bullet (3)] The uniform random choice of B' is indeed symmetric, but since the detailed-balance proof in Eq. (A5) uses P_F(B→B') = P_F(B'→B), it would be clearer to state explicitly that the uniform prescription satisfies this condition.
  2. [Figs. 3–5] The autocorrelation functions are plotted without statistical error bands; given that the integrated autocorrelation times are the central quantitative results, the authors should state the uncertainty on ρ(τ) or show representative error bands.
  3. [Eq. (3.3)] The definition χ_t = ⟨Q^2⟩/(La)^4 is used for both SU(2) and SU(2)/Z2, but in the latter theory Q takes half-integer plus integer values; the meaning of ⟨Q^2⟩ on a finite torus with dynamical B-field and the continuum limit of this quantity should be clarified.
  4. [Sec. 4] The quark-extension proposal is clearly labeled as a possible method, but the Stückelberg decoupling is only sketched; in particular, the U(1)_B gauge field has nontrivial Chern numbers tied to the B-field (Eq. (4.4)), and the compatibility of the twisted fermion boundary conditions with the dynamical B-field update is not discussed. A sentence stating that the decoupling is expected only in the continuum limit and requires further study would be appropriate.
  5. [Appendix B] The histograms in Figs. B1 and B2 would benefit from a description of the binning and of how the half-integer peaks in the SU(2)/Z2 case are formed; currently only raw histograms are shown.

Circularity Check

1 steps flagged · score 2.0 of 10

Q-autocorrelation reduction is by construction and is honestly flagged; the central claim retains independent E(t) evidence.

  1. self definitional [Section 1 (Eq. (1.2)) and Section 3.2 'Autocorrelation functions']
    "The drastic reduction of the autocorrelation in Q and in Q2 is, although quite impressive, somehow expected because the random choice of the B-field inevitably shuffles the topological charge."

    Eq. (1.2) defines the topological charge Q as a direct function of the B-field, Q = -(1/N) epsilon B B /8 + Z. The halfway-updating HMC redraws the B-field uniformly at random every trajectory (Section 3.1, step (3)), so the Monte Carlo proposal changes Q by construction. Thus the short Q autocorrelation time is a designed property of the update rather than an independent dynamical result. The paper explicitly concedes this in Section 3.2. The circularity is only partial because the paper separately measures the gradient-flow energy E(t) and observes reduced autocorrelation there, which is not the direct shuffling of a topological integer. The topological-susceptibility comparison also uses the external SU(2) result from Ref. [34] rather than fitting it.

full rationale

No load-bearing circularity is present. The paper's central algorithmic claim is supported by two observations: the topological-charge autocorrelation and the gradient-flow energy-operator autocorrelation. The first is acknowledged by the authors to be expected from the random B-field update, since Q is a direct function of B via Eq. (1.2). That is a by-construction effect, but the paper explicitly warns that it does not imply reduction for all observables and therefore adds the E(t) measurement. E(t) is B-dependent, but its reduced autocorrelation is not the simple shuffling of a topological integer and provides independent, if imperfect, evidence. The topological-susceptibility comparison is made against the external Teper result, not fitted to it, and the authors themselves state that further statistics on larger and finer lattices are needed before concluding large-volume equivalence. The self-citations [12] and [33] are used as supporting technical constructions, not as the load-bearing justification for the main simulation claim. Overall, the paper is largely self-contained; the score reflects only the explicitly admitted by-construction nature of the Q autocorrelation reduction, not a fatal circularity.

Assumptions & free parameters 3 free parameters · 5 assumptions · 1 invented entities

The main numerical result rests on standard lattice actions plus the large-volume equivalence assumption and the B-dependent flow definition. No constants were fitted to the target observable in this paper; the external string tension scale from Ref. [34] is imported. The quark section adds a further untested Stuckelberg construction.

free parameters (3)
  • Flow time t = (0.7L)^2/8 = t/a^2 = (0.7L)^2/8 in lattice units
    Chosen by hand as the smearing scale for the topological charge and energy operator; the paper states that the dependence on this choice is not studied.
  • HMC step size and trajectory length = Delta_tau = 0.02, tau = 1.0, 50 MD steps per trajectory
    Algorithm tuning parameters; at the largest lattice (beta=2.6, L=20) the Metropolis acceptance was about 86 percent. They are not fitted to a physics target.
  • Scale setting via SU(2) string tension mapping = a*sqrt(sigma) = 0.2673, 0.186, 0.1326 for beta = 2.4, 2.5, 2.6
    Imported from Teper [34] and applied also to the SU(2)/Z2 theory; the authors note this is subtle because Wilson lines are not gauge invariant in the quotient theory.
assumptions (5)
  • standard math The Wilson plaquette action and standard HMC detailed balance are valid for the lattice gauge theory.
    Used in Section 2, Eq. (2.8), and throughout Section 3.1; the proof in Appendix A relies on the standard leapfrog invertibility and Metropolis acceptance.
  • domain assumption Local observables in SU(N)/Z_N are insensitive to the quotienting in the large-volume limit.
    Invoked in the abstract and Section 1 to justify replacing SU(N) by SU(N)/Z_N; the difference is described as a boundary-condition sum that should vanish in large gapped volumes. The paper has only exploratory evidence for this.
  • domain assumption The gradient flow with a fixed B-field is renormalizable in the sense of Refs. [36, 37].
    Footnote 5 admits this is an open problem; the topological charge and E(t) used for the central autocorrelation measurement are defined through this flow.
  • domain assumption Uniform random updates of B in {0,...,N-1} sample all flat Z2 2-cochain sectors on the periodic lattice.
    Used in Section 3.1 step (3); the technical note argues the gauge-fixed form Eq. (2.10) plus random B updates covers the space of flat Z2 2-cochains, relying on the integral lift construction of Ref. [33].
  • domain assumption The lattice Stuckelberg mechanism makes the U(1)_B gauge boson super-heavy and decoupled in the continuum limit.
    Proposed in Section 4 for quark incorporation; this is a sketch with no numerical test, and the paper does not claim it has been validated.
invented entities (1)
  • Stuckelberg scalar field Omega(x) in U(1) for the quark extension
    purpose: To give a large mass to the unwanted U(1)_B gauge boson introduced when embedding Z_N into SU(N) x U(1)_B, so that the framework can approach QCD.
    This is an internal lattice construction proposed in Section 4; no independent experimental or numerical handle is provided, and the paper leaves its validation to future work.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Monte Carlo Simulation of the $SU(2)/\mathbb{Z}_2$ Yang--Mills Theory." pith.science (2026). https://pith.science/paper/2TNEMF4Q

@misc{pith2026250100286,
  author       = {Pith},
  title        = {Pith review of: Monte Carlo Simulation of the $SU(2)/\mathbbZ_2$ Yang--Mills Theory},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2TNEMF4Q}},
  note         = {Machine review of arXiv:2501.00286}
}
abstract

We carry out a hybrid Monte Carlo (HMC) simulation of the $SU(2)/\mathbb{Z}_2$ Yang--Mills theory in which the $\mathbb{Z}_N$ 2-form flat gauge field (the 't~Hooft flux) is explicitly treated as one of the dynamical variables. We observe that our HMC algorithm in the $SU(2)/\mathbb{Z}_2$ theory drastically reduces autocorrelation lengths of the topological charge and of a physical quantity which couples to slow modes in the conventional HMC simulation of the $SU(2)$ theory. Provided that sufficiently large lattice volumes are available, therefore, the HMC algorithm of the $SU(N)/\mathbb{Z}_N$ theory could be employed as an alternative for the simulation of the $SU(N)$ Yang--Mills theory, because local observables are expected to be insensitive to the difference between $SU(N)$ and~$SU(N)/\mathbb{Z}_N$ in the large volume limit. A possible method to incorporate quarks [fermions in the fundamental representation of~$SU(N)$ with the baryon number~$1/N$] in this framework is also considered.

Figures

Figures reproduced from arXiv: 2501.00286 by the authors.

Figure 1
Figure 1. The HMC history of the topological charge [PITH_FULL_IMAGE:figures/full_fig_p009_1.png] view at source ↗
Figure 2
Figure 2. The HMC history of the topological charge [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗
Figure 3
Figure 3. The normalized autocorrelation functions of the topological charge [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figures from the paper (8 more)
Figure 4
Figure 4. Figure 4: The normalized autocorrelation functions of the topological charge [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]
Figure 5
Figure 5. Figure 5: The normalized autocorrelation functions of the topological charge [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]
Figure 6
Figure 6. Figure 6: The integrated autocorrelation lengths in lattice units of the topological charge [PITH_FULL_IMAGE:figures/full_fig_p013_6.png]
Figure 7
Figure 7. Figure 7: The HMC history of E(t) (3.2) in the SU(2) theory (i.e., without the B-field). β = 2.6 and L = 16. susceptibility χt := 1 (La) 4 [PITH_FULL_IMAGE:figures/full_fig_p013_7.png]
Figure 8
Figure 8. Figure 8: The HMC history of E(t) (3.2) in the SU(2)/Z2 theory (i.e., with the B-field). β = 2.6 and L = 16. in units of the string tension σ. For all lattice parameters, the first 50 configurations (500 MD time) are omitted for thermalization and statistical errors are estimate…
Figure 9
Figure 9. Figure 9: The continuum extrapolations of the topological susceptibility [PITH_FULL_IMAGE:figures/full_fig_p015_9.png]
Figure 10
Figure 10. Figure 10: The continuum extrapolation of the topological susceptibility [PITH_FULL_IMAGE:figures/full_fig_p015_10.png]
Figure 11
Figure 11. Figure 11: Continuum extrapolations in Fig. 9 as a function of the physical volume. A naive [PITH_FULL_IMAGE:figures/full_fig_p016_11.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. Direct Monte Carlo Computation of the 't~Hooft Partition Function

    hep-lat 2025-01 conditional novelty 7.0 of 10

    A direct Monte Carlo count of 't Hooft flux sectors yields the 't Hooft partition function for SU(2) Yang-Mills, confirming the ordinary confining phase and, via the Witten effect, indicating oblique confinement at theta=2pi.

  2. Numerical simulation of fractional topological charge in $SU(N)$ gauge theory coupled with $\mathbb{Z}_N$ 2-form gauge fields

    hep-lat 2025-01 conditional novelty 5.0 of 10

    Numerical lattice simulation confirms that Z_2 2-form gauge-field coupling produces fractional (half-integer) topological charge in SU(2) gauge theory and reduces topological autocorrelation.

Reference graph

Works this paper leans on

42 extracted references · 6 canonical work pages · cited by 2 Pith papers

  1. [34]

    M. J. Teper, [arXiv:hep-th/9812187 [hep-th]]

    work page Pith review arXiv

  2. [1]

    ’t Hooft, Nucl

    G. ’t Hooft, Nucl. Phys. B153, 141-160 (1979) doi:10.1016/0550-3213(79)90595-9

    doi:10.1016/0550-3213(79)90595-9 1979

  3. [2]

    Kitano, T

    R. Kitano, T. Suyama and N. Yamada, JHEP09, 137 (2017) doi:10.1007/JHEP09(2017)137 [arXiv:1709.04225 [hep-th]]

    arXiv 2017

  4. [3]

    Tanizaki and M

    Y. Tanizaki and M. ¨Unsal, PTEP2022, no.4, 04A108 (2022) doi:10.1093/ptep/ptac042 [arXiv:2201.06166 [hep-th]]

    arXiv 2022

  5. [4]

    Nguyen, Y

    M. Nguyen, Y. Tanizaki and M. ¨Unsal, JHEP08, 013 (2023) doi:10.1007/JHEP08(2023)013 [arXiv:2306.02485 [hep-th]]

    arXiv 2023

  6. [5]

    Gaiotto, A

    D. Gaiotto, A. Kapustin, N. Seiberg and B. Willett, JHEP02, 172 (2015) doi:10.1007/JHEP02(2015)172 [arXiv:1412.5148 [hep-th]]

    arXiv 2015

  7. [6]

    Gaiotto, A

    D. Gaiotto, A. Kapustin, Z. Komargodski and N. Seiberg, JHEP05, 091 (2017) doi:10.1007/JHEP05(2017)091 [arXiv:1703.00501 [hep-th]]

    arXiv 2017

  8. [7]

    Yamazaki and K

    M. Yamazaki and K. Yonekura, JHEP07, 088 (2017) doi:10.1007/JHEP07(2017)088 [arXiv:1704.05852 [hep- th]]

    arXiv 2017

Show all 42 references
  1. [8]

    Hayashi, Y

    Y. Hayashi, Y. Tanizaki and H. Watanabe, JHEP10, 146 (2023) doi:10.1007/JHEP10(2023)146 [arXiv:2307.13954 [hep-th]]

    2023 arXiv

  2. [9]

    Hayashi and Y

    Y. Hayashi and Y. Tanizaki, JHEP08, 001 (2024) doi:10.1007/JHEP08(2024)001 [arXiv:2402.04320 [hep-th]]

    2024 arXiv

  3. [10]

    Hayashi and Y

    Y. Hayashi and Y. Tanizaki, Phys. Rev. Lett.133, no.17, 171902 (2024) doi:10.1103/PhysRevLett.133.171902 [arXiv:2405.12402 [hep-th]]

    2024 arXiv

  4. [11]

    Hayashi, T

    Y. Hayashi, T. Misumi and Y. Tanizaki, JHEP05, 194 (2025) doi:10.1007/JHEP05(2025)194 [arXiv:2410.21392 [hep-th]]

    2025 arXiv

  5. [12]

    M. Abe, O. Morikawa, S. Onoda, H. Suzuki and Y. Tanizaki, JHEP08, 118 (2023) doi:10.1007/JHEP08(2023)118 [arXiv:2303.10977 [hep-lat]]

    2023 arXiv

  6. [13]

    ’t Hooft, Commun

    G. ’t Hooft, Commun. Math. Phys.81, 267-275 (1981) doi:10.1007/BF01208900

    1981 doi

  7. [14]

    van Baal, Commun

    P. van Baal, Commun. Math. Phys.85, 529 (1982) doi:10.1007/BF01403503

    1982 doi

  8. [15]

    I. G. Halliday and A. Schwimmer, Phys. Lett. B101, 327 (1981) doi:10.1016/0370-2693(81)90055-1

    1981 doi

  9. [16]

    Creutz and K

    M. Creutz and K. J. M. Moriarty, Nucl. Phys. B210, 50-58 (1982) doi:10.1016/0550-3213(82)90248-6

    1982 doi

  10. [17]

    R. G. Edwards, U. M. Heller and R. Narayanan, Phys. Lett. B438, 96-98 (1998) doi:10.1016/S0370- 2693(98)00951-4 [arXiv:hep-lat/9806011 [hep-lat]]

    1998 arXiv

  11. [18]

    de Forcrand and O

    P. de Forcrand and O. Jahn, Nucl. Phys. B651, 125-142 (2003) doi:10.1016/S0550-3213(02)01123-9 [arXiv:hep- lat/0211004 [hep-lat]]

    2003

  12. [19]

    T. G. Kov´ acs and E. T. Tomboulis, Phys. Rev. Lett.85, 704-707 (2000) doi:10.1103/PhysRevLett.85.704 [arXiv:hep-lat/0002004 [hep-lat]]

    2000 arXiv

  13. [20]

    I. G. Halliday and A. Schwimmer, Phys. Lett. B102, 337-340 (1981) doi:10.1016/0370-2693(81)90630-4

    1981 doi

  14. [21]

    Duane, A

    S. Duane, A. D. Kennedy, B. J. Pendleton and D. Roweth, Phys. Lett. B195, 216-222 (1987) doi:10.1016/0370- 2693(87)91197-X

    1987 doi

  15. [22]

    Del Debbio, H

    L. Del Debbio, H. Panagopoulos and E. Vicari, JHEP08, 044 (2002) doi:10.1088/1126-6708/2002/08/044 [arXiv:hep-th/0204125 [hep-th]]

    2002 arXiv

  16. [23]

    Schaeferet al.[ALPHA], Nucl

    S. Schaeferet al.[ALPHA], Nucl. Phys. B845, 93-119 (2011) doi:10.1016/j.nuclphysb.2010.11.020 [arXiv:1009.5228 [hep-lat]]

    2011 arXiv

  17. [24]

    Bonanno, G

    C. Bonanno, G. Clemente, M. D’Elia, L. Maio and L. Parente, JHEP08, 236 (2024) doi:10.1007/JHEP08(2024)236 [arXiv:2404.14151 [hep-lat]]

    2024 arXiv

  18. [25]

    L¨ uscher and S

    M. L¨ uscher and S. Schaefer, JHEP07, 036 (2011) doi:10.1007/JHEP07(2011)036 [arXiv:1105.4749 [hep-lat]]

    2011 arXiv

  19. [26]

    Alexandrou, A

    C. Alexandrou, A. Athenodorou and K. Jansen, Phys. Rev. D92, no.12, 125014 (2015) doi:10.1103/PhysRevD.92.125014 [arXiv:1509.04259 [hep-lat]]

    2015 arXiv

  20. [27]

    L¨ uscher, JHEP08, 071 (2010) [erratum: JHEP03, 092 (2014)] doi:10.1007/JHEP08(2010)071 [arXiv:1006.4518 [hep-lat]]

    M. L¨ uscher, JHEP08, 071 (2010) [erratum: JHEP03, 092 (2014)] doi:10.1007/JHEP08(2010)071 [arXiv:1006.4518 [hep-lat]]

    2010 arXiv

  21. [28]

    Kapustin and N

    A. Kapustin and N. Seiberg, JHEP04, 001 (2014) doi:10.1007/JHEP04(2014)001 [arXiv:1401.0740 [hep-th]]

    2014 arXiv

  22. [29]

    Mack and V

    G. Mack and V. B. Petkova, Annals Phys.125, 117 (1980) doi:10.1016/0003-4916(80)90121-9

    1980 doi

  23. [30]

    Ukawa, P

    A. Ukawa, P. Windey and A. H. Guth, Phys. Rev. D21, 1013 (1980) doi:10.1103/PhysRevD.21.1013

    1980 doi

  24. [31]

    Seiler, Lect

    E. Seiler, Lect. Notes Phys.159, 1-192 (1982)

    1982

  25. [32]

    JuliaQCD: Portable lattice QCD package in Julia language,

    Y. Nagai and A. Tomiya, “JuliaQCD: Portable lattice QCD package in Julia language,” [arXiv:2409.03030 [hep-lat]]

    arXiv

  26. [33]

    M. Abe, O. Morikawa and H. Suzuki, PTEP2023, no.2, 023B03 (2023) doi:10.1093/ptep/ptad009 [arXiv:2210.12967 [hep-th]]

    2023 arXiv

  27. [35]

    L¨ uscher, Commun

    M. L¨ uscher, Commun. Math. Phys.85, 39 (1982) doi:10.1007/BF02029132 23

    1982 doi

  28. [36]

    L¨ uscher and P

    M. L¨ uscher and P. Weisz, JHEP02, 051 (2011) doi:10.1007/JHEP02(2011)051 [arXiv:1101.0963 [hep-th]]

    2011 arXiv

  29. [37]

    Hieda, H

    K. Hieda, H. Makino and H. Suzuki, Nucl. Phys. B918, 23-51 (2017) doi:10.1016/j.nuclphysb.2017.02.017 [arXiv:1604.06200 [hep-lat]]

    2017 arXiv

  30. [38]

    Hern´ andez, K

    P. Hern´ andez, K. Jansen and M. L¨ uscher, Nucl. Phys. B552, 363-378 (1999) doi:10.1016/S0550-3213(99)00213- 8 [arXiv:hep-lat/9808010 [hep-lat]]

    1999 arXiv

  31. [39]

    L¨ uscher, Comput

    M. L¨ uscher, Comput. Phys. Commun.165, 199-220 (2005) doi:10.1016/j.cpc.2004.10.004 [arXiv:hep- lat/0409106 [hep-lat]]

    2005

  32. [40]

    Curci and G

    G. Curci and G. Veneziano, Nucl. Phys. B292, 555-572 (1987) doi:10.1016/0550-3213(87)90660-2

    1987 doi

  33. [41]

    Kouno, Y

    H. Kouno, Y. Sakai, T. Makiyama, K. Tokunaga, T. Sasaki and M. Yahiro, J. Phys. G39, 085010 (2012) doi:10.1088/0954-3899/39/8/085010

    2012 doi

  34. [42]

    Iritani, E

    T. Iritani, E. Itou and T. Misumi, JHEP11, 159 (2015) doi:10.1007/JHEP11(2015)159 [arXiv:1508.07132 [hep-lat]]. 24

    2015 arXiv

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.