Pith. sign in

REVIEW 3 major objections 4 minor 40 references

This paper claims that in strongly coupled three-dimensional QED, both Schwinger-Dyson and 3PI non-perturbative vertex prescriptions violate the Ward identity by roughly 30%, and that the popular Ball-Chiu vertex ansatz only tracks the full

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-02 09:59 UTC pith:7RW6VD3M

load-bearing objection Useful numerical benchmark with a real caveat: the headline ~30% Ward-violation number is biased by the imposed λ4=0 constraint, so treat it as indicative rather than precise. the 3 major comments →

arxiv 2606.27213 v2 pith:7RW6VD3M submitted 2026-06-25 hep-th hep-ph

The gauge invariance of non-perturbative vertex prescriptions

M.E. Carrington , A.R. Frey , B.A. Meggison This is my paper
classification hep-th hep-ph
keywords gauge invarianceWard identitySchwinger-Dyson equations3PI effective actionBall-Chiu ansatzthree-dimensional QEDnon-perturbative vertexLandau gauge
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper asks whether widely used non-perturbative vertex approximations in gauge theories can be trusted at strong coupling. Working in Landau-gauge QED in three dimensions, it solves the coupled Schwinger-Dyson equations and the 3PI effective-action equations for the electron-photon vertex, then tests gauge invariance by checking how well the Ward identity is satisfied. At the largest coupling studied (α=5), the Ward identity is violated by about 30% with either method, with 3PI slightly better. The Ball-Chiu ansatz, which satisfies the Ward identity by construction, reproduces the full vertex reasonably well at small coupling but deviates significantly at large coupling. If correct, this means strong-coupling results obtained from naive SD truncations, 3PI effective actions, or the Ball-Chiu ansatz carry a gauge bias that can only be assessed by comparing across gauges.

Core claim

The central claim is that gauge invariance—quantified by how well the non-perturbative vertex satisfies the Ward identity—is significantly violated in both major continuum methods for strong-coupling QED. Using a 12-component vertex basis, the authors solve the truncated SD equations and the three-loop 3PI equations in Landau gauge, and compare the longitudinal components of the computed vertices with the Ward-identity prediction. The violation grows with coupling and reaches about 30% at α=5; the 3PI vertex is slightly but consistently closer to the Ward identity than the SD vertex. The Ball-Chiu ansatz, though gauge-consistent by construction, agrees with the fully computed vertices only a

What carries the argument

The machinery is the decomposition of the QED three-point vertex into 12 basis tensors (8 transverse, 4 longitudinal) whose scalar dressing functions are obtained by projecting the SD or 3PI integral equations. The gauge-invariance test uses the Ward identity, which fixes the three longitudinal components λ1–λ3 in terms of the fermion two-point functions; comparing the computed λi with those expressions measures the gauge violation. The Ball-Chiu ansatz is used as the standard gauge-consistent but transverse-blind baseline. The numerical simplification algorithm that cancels spurious kinematic denominators is essential: it makes the self-consistent equations tractable at all.

Load-bearing premise

The calculation assumes that setting the vertex component λ4 to zero does not distort the Ward-identity check, even though the truncated vertices being tested are not Ward-consistent and could in principle generate a non-zero λ4 that feeds back into the other components.

What would settle it

Compute the same SD and 3PI vertices in Feynman gauge (or another covariant gauge) and compare a physical quantity such as the fermion condensate or the critical coupling for chiral symmetry breaking with the Landau-gauge value. If the physical quantity is unchanged despite the roughly 30% Ward-identity violation, the paper's implication that the methods are gauge-biased at strong coupling is falsified. A second check is to repeat the Landau-gauge calculation keeping λ4 in the truncated system and see whether the measured violation changes significantly.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • At strong coupling, a naive SD truncation with a bare four-point function yields vertices whose gauge-invariance violation is already visible at the level of the Ward identity, reaching roughly 30% at α=5.
  • The 3PI effective action is not exactly gauge invariant, but it is consistently a little closer to satisfying the Ward identity than the SD approach at every coupling tested.
  • The Ball-Chiu ansatz, despite satisfying the Ward identity by construction, is not a faithful representation of the full non-perturbative vertex once the coupling is large; its transverse components and their momentum dependence deviate substantially.
  • The two-point dressing functions A, B, and Π computed from all these methods agree fairly well even at α=2, so the vertex-level gauge violation does not necessarily destroy the reliability of two-point quantities.
  • The negligible θ-dependence of the computed vertex dressing functions is consistent with using simpler, θ-independent ansätze in the weak-coupling regime.

Where Pith is reading between the lines

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

  • An obvious next check is to run the same systems in Feynman gauge; the key question is whether the roughly 30% Ward-identity violation shows up as a shift in a physical observable such as the fermion condensate or the chiral-symmetry-breaking critical coupling. If it does not, the practical gauge bias is smaller than the raw number suggests.
  • The paper's choice to enforce λ4=0 by hand sits on top of truncated equations that are not Ward-consistent. Since λ4 can be generated by the truncation and can feed into the λ1–λ3 equations, the measured 30% could be sensitive to this input constraint. A direct calculation with λ4 included would settle whether the violation is intrinsic or partly an artifact.
  • The comparison of σ2 and τ2 shows structural differences in a component the BC ansatz ties to the fermion self-energy. If this component contributes significantly to the vertex in strong-coupling applications, previous BC-based strong-coupling studies in QED3 may carry a systematic vertex error of the kind quantified here.
  • The denominator-simplification algorithm, which reduced the weight of singular kinematic factors by up to a factor of nine, is a transferable technique for other tensor-decomposed vertex equations and could make otherwise intractable non-perturbative vertex calculations feasible.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 4 minor

Summary. The paper studies the gauge invariance of non-perturbative vertex prescriptions in three-dimensional QED. It solves coupled integral equations for the fermion and photon two-point functions together with the full vertex, using both a truncated Schwinger-Dyson (SD) hierarchy and a 3PI effective action, and compares the resulting longitudinal vertex components against the Ward identity. It also compares the full vertex with the Ball-Chiu (BC) ansatz. The main quantitative finding is that at the largest coupling α=5 the Ward identity is violated by approximately 30% for both 3PI and SD schemes, with 3PI slightly better, and that the BC ansatz agrees reasonably at small coupling but deviates more at large coupling. The paper also reports one-loop checks and grid-stability tests.

Significance. If the quantitative claim is correct, the paper provides a useful, computationally demanding benchmark for vertex truncations in strongly coupled gauge theories. The symbolic reduction, the one-loop Ward identity validation (Table II), the grid-stability checks (Table III), and the comparison with the widely used BC ansatz are concrete strengths. The qualitative conclusion—that Ward-identity violation grows with coupling and is slightly smaller for 3PI than for SD—is defensible. However, the exact 30% number is sensitive to an imposed input constraint and to an unspecified weighting, as detailed below, so the quantitative headline should be treated with caution pending further analysis.

major comments (3)
  1. [§2.C] The statement 'we have set λ4 to zero' is an input constraint, not a result. The exact Ward identity forces λ4=0, but the truncated vertices under test are not Ward-consistent; λ4 may be generated by the truncation and, through the full middle vertex Γ^μ(k+r,p+r) in eq. (2.18), contributes to the projections onto L1–L3. Thus the computed λ1–λ3, and the resulting 0.305/0.322 WI violations at α=5, describe a modified truncation, not the pure 3PI/SD scheme. The magnitude of this bias is not estimated. Please either release the λ4=0 constraint (include a λ4 equation) or demonstrate that its effect on λ1–λ3 is negligible.
  2. [Table IV / §4.C] The 'weighted average' column is not reproducible. The weights are said to be 'calculated from the results in fig. 5', but no numerical values or formula are given, and the headline ~30% number is exactly this weighted average. Because λ3 has a relative difference of ~1.0–1.16 at α=5, the weighted average could change substantially under alternative, equally plausible weights. Provide the explicit weights and show sensitivity of the conclusion to the weighting choice.
  3. [§2.B/§2.C] The SD method is defined by two different vertex integral equations—one with the third vertex bare and one with the second vertex bare—used respectively in (2.9) and (2.10). It is not stated which of these two vertices is used for the longitudinal components reported in Table IV, nor whether the two variants give different λ_i. Without this, the 'SD' row of the table is ambiguous. Please specify the algorithm concretely, or report both variants.
minor comments (4)
  1. [Eq. (2.1)] Typo: q^μ = k^μ − p^ν should read q^μ = k^μ − p^μ.
  2. [Eq. (2.15)] The definition of T_7 contains a bracket structure that is hard to parse; consider writing it with explicit antisymmetrized indices or a clearer notation.
  3. [§4.C] The text should define the 'weights' used for the weighted average in Table IV explicitly, rather than only referencing Fig. 5.
  4. [References] Reference [25] should be formatted consistently (e.g., 'Zel'dovich').

Circularity Check

0 steps flagged

No circular derivation: Ward-identity and Ball-Chiu checks compare independently computed quantities; λ4=0 is a stated validity limitation, not a construction that forces the result.

full rationale

The central numerical claim is the gauge-invariance check in Sec. 4.C, Table IV: the longitudinal dressing functions λ1–λ3 obtained from the self-consistent vertex integral equations (2.16)–(2.18) are compared with the closed-form Ward-identity expressions (2.22a)–(2.22c) computed from the fermion functions A and B. Neither side is fitted to the other; both are outputs of the same truncated system, but the Ward-identity expressions are a different functional of A and B. The one-loop limit is cross-checked against the independent analytic results of Appendix A (Table II). The Ball-Chiu comparison in Sec. 4.D is also independent: the transverse components τi of the 3PI/SD vertices are solved from their integral equations, while the BC components σi are constructed from A and B via (2.26) and (2.28). Nothing is tuned to produce the reported ~30% violation. The self-citations ([7], [17], [23], [26]) support standard 3PI/setup statements and are corroborated by non-author references ([6], [18]); they are not used to forbid alternatives or to supply the numerical result. The paper itself states the main limitation: Sec. 2.C imposes λ4=0 because the exact Ward identity would require it, even though the truncated vertices are not Ward-consistent. This could bias the quantitative violation and its magnitude is not estimated, but it is an input assumption, not a circular reduction: the λ1–λ3 comparison does not equal the λ4 constraint by construction. The paper also explicitly limits the physical interpretation, noting in Sec. 5 that a direct check in a different gauge is still in progress. These are correctness/validity caveats, not circularity.

Axiom & Free-Parameter Ledger

3 free parameters · 6 axioms · 0 invented entities

No fitted constants appear; the couplings α∈{0.64,1,2,5} are physical inputs, and the main hand-set numbers are the bare mass m=1, the photon regulator, and the weighting of the WI-violation measure. The significant ledger items are truncation axioms and the λ4=0 constraint.

free parameters (3)
  • fermion mass in cutoff units (m/Λ) = 1
    Set by hand in Sec. 3; the paper postpones mass dependence. The Ward-identity violation could depend on this value.
  • photon regulator δ = 1e-5
    Added to the photon dressing to avoid a small-momentum instability; its effect on the 30% figure is not quantified.
  • WI-violation averaging weights = not specified; based on fig. 5 magnitudes
    The reported ~30% is a weighted average of λ_i relative differences with weights from average magnitudes; a different weighting, e.g. unweighted, changes the headline number. This is a chosen aggregation.
axioms (6)
  • domain assumption QED3 with four-component spinors, bare mass and momentum cutoff is a valid testbed for the truncations
    Used throughout; physical implications for phase transitions are inferred from this model.
  • domain assumption Truncating the SD hierarchy at a bare four-point vertex neglects no other terms at that order
    Sec. 1; the claim quantifies this truncation's gauge violation.
  • domain assumption The three-loop 3PI effective action preserves gauge invariance to the level of the truncation
    Sec. 1; relies on refs [6,7] and motivates interpreting 3PI as the better vertex.
  • standard math The 12-term Ball-Chiu basis (2.15) spans the vertex and yields dressing functions finite at q^2=0 and Δ=0
    Sec. 2.B; relies on [1,19].
  • domain assumption Landau-gauge transverse and longitudinal dressing functions decouple in the equations for two-point functions and τ_i
    Sec. 2.D; the numerical scheme exploits this decoupling.
  • ad hoc to paper λ4=0 is imposed
    Sec. 2.C; exact WI implies λ4=0, but the truncated vertices being tested are not gauge-consistent, so this constraint may bias the diagnostic.

pith-pipeline@v1.3.0-alltime-deepseek · 16455 in / 17958 out tokens · 187373 ms · 2026-08-02T09:59:42.593322+00:00 · methodology

0 comments
read the original abstract

We study the gauge invariance of different continuum methods to include non-perturbative effects in gauge theories. We work with three dimensional quantum electrodynamics and implement vertices using two different methods: a set of coupled Schwinger-Dyson (SD) integral equations, and the self-consistent equations obtained from the 3-particle irreducible (3PI) effective action. We work in Landau gauge and assess the extent to which results are gauge invariant by checking how well the Ward identity is satisfied. Our results show that there is a fairly significant violation of the Ward identity at large coupling, although the 3PI effective theory is slightly better than the SD vertex. We also compare the results of both calculations with the commonly used Ball-Chiu ansatz and show that the agreement of the ansatz with both non-perturbative vertices is fairly good at small coupling but deviates more significantly at large coupling. We compare results for the two point functions of the theory and discuss the possible implications for phase transitions.

Figures

Figures reproduced from arXiv: 2606.27213 by A.R. Frey, B.A. Meggison, M.E. Carrington.

Figure 1
Figure 1. Figure 1: FIG. 1: propagator corrections [PITH_FULL_IMAGE:figures/full_fig_p008_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2: The vertex correction [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3: Fermion and photon dressing functions using different vertex methods with [PITH_FULL_IMAGE:figures/full_fig_p023_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4: Fermion and photon dressing functions using different vertex methods with [PITH_FULL_IMAGE:figures/full_fig_p024_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5: The average value of the vertex dressing functions for [PITH_FULL_IMAGE:figures/full_fig_p024_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6: Transverse dressing functions with [PITH_FULL_IMAGE:figures/full_fig_p025_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7: Longitudinal dressing functions with [PITH_FULL_IMAGE:figures/full_fig_p026_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8: The dressing functions ˆσ [PITH_FULL_IMAGE:figures/full_fig_p028_8.png] view at source ↗

Review history (2 revisions) →

discussion (0)

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

Reference graph

Works this paper leans on

40 extracted references · 1 linked inside Pith

  1. [1]

    Vertex functions are notoriously difficult to calculate and the problem is commonly avoided by introducing some simplified ansatz to represent the vertex

    INTRODUCTION The purpose of this paper is to study the gauge invariance of different non-perturbative methods that are commonly used to calculate vertex functions in strongly coupled field theories. Vertex functions are notoriously difficult to calculate and the problem is commonly avoided by introducing some simplified ansatz to represent the vertex. The...

  2. [2]

    We work in three dimensional Euclidean space

    NOT A TION We use natural units (ℏ=c= 1). We work in three dimensional Euclidean space. Greek letters denote Lorentz indices of three-dimensional vectors. We use a representation of three four-dimensional gamma matrices that satisfy the commutation relation{γ µ, γν}= 2δ µν. A three point vertex function depends on two external momenta because of momentum ...

  3. [3]

    SIMPLIFICA TION AND NUMERICAL METHOD As discussed in sec. 2 C, the most important aspect of the numerical calculation is the method used to simplify the set of self-consistent equations that we are trying to solve, before attempting to solve the self-consistent equations numerically. The basic problem is the appearance in denominators of factors of the fo...

  4. [4]

    Collect terms with the same powers of all factors exceptp 2,k 2 andk·p

  5. [5]

    Apply Mathematica’s Simplify function and identify combinations ofp 2,k 2, andk·p that give eitherq 2 or ∆

  6. [6]

    Repeat steps 1 and 2 until the number of terms of the integrand is not reduced. Note that it is necessary to repeat the process because a simplified term has a different power ofq 2 or ∆ than the terms it was made from (like in (3.1)), which means that in step 1 the new term will be collected with a different set and might again find partners it can be co...

  7. [7]

    Replacek·pwith 16 1 2 (k2 +p 2 −q 2) and Expand

    Perform the collection in step 1 then select terms where the denominator has at least one factor ofq 2 and the numerator has at least one factor ofk·p. Replacek·pwith 16 1 2 (k2 +p 2 −q 2) and Expand. If the length of the integrand does not increase, keep the new expression, if it does, revert to the original. The new expression (if kept) will have at wor...

  8. [8]

    Repeat step 4 until the length of the integrand is no longer reduced

  9. [9]

    Perform steps 4 and 5 but with ∆ in the denominator instead ofq 2 and (k·p) 2 in the numerator instead ofk·p, and replace (k·p) 2 withk 2p2 −∆

  10. [10]

    For example, for the integral equation that givesτ 2, the original expression has 629,221 terms and each of them have at least one power ofq 2 or ∆ in the denominator

    Go back to step 1 and repeat the entire process until the length of the integrand does not change. For example, for the integral equation that givesτ 2, the original expression has 629,221 terms and each of them have at least one power ofq 2 or ∆ in the denominator. The number with each of these factors is shown in table I in the column labelled ‘before.’...

  11. [11]

    Preliminary One of the main goals of this work is to compare the extent to which the 3PI and SD vertex functions satisfy the Ward identity

    RESUL TS A. Preliminary One of the main goals of this work is to compare the extent to which the 3PI and SD vertex functions satisfy the Ward identity. This gives an indication of the extent to which each method can be trusted. We will also test how well the Ball-Chiu vertex ansatz agrees with the non-perturbative vertex functions. This is important infor...

  12. [12]

    We have solved the equations of motion obtained from the three loop 3PI effective action

    CONCLUSIONS We have studied different methods to calculate non-perturbative vertex dressing functions in strongly coupled QED in three Euclidean dimensions. We have solved the equations of motion obtained from the three loop 3PI effective action. We have also solved the coupled set of QED SD equations truncated with a bare four-point vertex. We have compa...

    2023

  13. [13]

    J. S. Ball and T. W. Chiu, Phys. Rev. D22, 2542 (1980)

    1980

  14. [14]

    C. D. Roberts and A. G. Williams, Prog. Part. Nucl. Phys.33, 477 (1994)

    1994

  15. [15]

    Binosi and J

    D. Binosi and J. Papavassiliou, Phys. Rept.479, 1 (2009)

    2009

  16. [16]

    E. A. Calzetta, Int. J. Theor. Phys.43, 767 (2004)

    2004

  17. [17]

    Reinosa and J

    U. Reinosa and J. Serreau, Annals Phys.325, 969 (2010)

    2010

  18. [18]

    Arrizabalaga and J

    A. Arrizabalaga and J. Smit, Phys. Rev. D66, 065014 (2002). 30

    2002

  19. [19]

    M. E. Carrington, G. Kunstatter and H. Zaraket, Eur. Phys. J. C42, 253 (2005)

    2005

  20. [20]

    M. J. Brown and I. B. Whittingham, Phys. Rev. D95, 025018 (2017)

    2017

  21. [21]

    M. Q. Huber, Phys. Rept.879, 1 (2020)

    2020

  22. [22]

    D. C. Curtis and M. R. Pennington, Phys. Rev. D42, 4165 (1990)

    1990

  23. [23]

    D. C. Curtis and M. R. Pennington, Phys. Rev. D48, 4933 (1993)

    1993

  24. [24]

    Kizilersu and M

    A. Kizilersu and M. R. Pennington, Phys. Rev. D79, 125020 (2009)

    2009

  25. [25]

    Goecke, C

    T. Goecke, C. S. Fischer and R. Williams, Phys. Rev. B79, 064513 (2009)

    2009

  26. [26]

    Kızılers¨ u, T

    A. Kızılers¨ u, T. Sizer, M. R. Pennington, A. G. Williams and R. Williams, Phys. Rev. D91, 065015 (2015)

    2015

  27. [27]

    Eichmann, J

    G. Eichmann, J. M. Pawlowski and J. M. Silva, Phys. Rev. D104, 114016 (2021)

    2021

  28. [28]

    Albino, A

    L. Albino, A. Bashir, A. J. Mizher and A. Raya, Phys. Rev. D106, 096007 (2022)

    2022

  29. [29]

    M. E. Carrington, Eur. Phys. J. C35, 383 (2004)

    2004

  30. [30]

    Berges, Phys

    J. Berges, Phys. Rev. D70, 105010 (2004)

    2004

  31. [31]

    Kizilersu, M

    A. Kizilersu, M. Reenders and M. R. Pennington, Phys. Rev. D52, 1242 (1995)

    1995

  32. [32]

    Eichmann, [arXiv:2603.00804 [hep-ph]]

    G. Eichmann, [arXiv:2603.00804 [hep-ph]]

    Pith/arXiv arXiv

  33. [33]

    A. I. Davydychev, P. Osland and L. Saks, Phys. Rev. D63, 014022 (2001)

    2001

  34. [34]

    M. E. Carrington, C. S. Fischer, L. von Smekal and M. H. Thoma, Phys. Rev. B94, 125102 (2016)

    2016

  35. [35]

    M. E. Carrington, A. R. Frey and B. A. Meggison, Phys. Rev. D107, 056012 (2023)

    2023

  36. [36]

    Williams, C

    R. Williams, C. S. Fischer and W. Heupel, Phys. Rev. D93, 034026 (2016)

    2016

  37. [37]

    Zel’Dovich, Ia B, Sov. Phys. JETP6(6), 1184–1186 (1958)

    1958

  38. [38]

    M. E. Carrington and E. Kovalchuk, Phys. Rev. D77, 025015 (2008)

    2008

  39. [39]

    Kuipers, T

    J. Kuipers, T. Ueda, J. A. M. Vermaseren and J. Vollinga, Comput. Phys. Commun.184, 1453-1467 (2013)

    2013

  40. [40]

    Bermudez, L

    R. Bermudez, L. Albino, L. X. Guti´ errez-Guerrero, M. E. Tejeda-Yeomans and A. Bashir, Phys. Rev. D95, 034041 (2017). 31

    2017