Pith. sign in

REVIEW 3 major objections 3 minor 10 references

Grassmann tensor approach for two-dimensional QCD in the strong-coupling expansion

T0 review · 3 major / 3 minor · reviewed 2026-08-09 · deepseek-v4-flash

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

Pith's one-line read This paper computes two-dimensional QCD strong-coupling coefficients through order beta^3 with no higher-order contamination, then fits them to extract the critical chemical potential's beta expansion.

desk verdict Plausible algorithmic advance for strong-coupling tensor networks, but the central order-separation claim is asserted, not yet demonstrated. read the letter →

arxiv 2501.19192 v2 pith:6M7T4Z22 submitted 2025-01-31 hep-lat cond-mat.stat-mech

Thomas Samberger , Jacques Bloch , Robert Lohmayer This is my paper · ORCID
classification hep-latcond-mat.stat-mech
keywords strong-couplingexpansionlatticeQCDtensornetworkGrassmannvariableshigher-orderrenormalizationgroupchemicalpotentialchiralcondensatequarknumberdensity
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 paper sets out to show that two-dimensional lattice QCD with staggered quarks at nonzero chemical potential can be rewritten as a Grassmann tensor network, and that a modified higher-order tensor renormalization group computes the strong-coupling expansion of the partition function order by order. Its central claim is that the quark number density and chiral condensate can be obtained directly through order $\beta^3$, with no higher-order terms leaking into those coefficients. If true, the reported tensor data are exact strong-coupling coefficients, not contaminated approximations, and the fitted $\beta$-dependence of the critical chemical potential is a quantitative prediction for where the transitions occur. This matters because the strong-coupling expansion is one of the few controlled routes into the finite-density region of QCD where standard Monte Carlo sampling fails.

What carries the argument

The load-bearing object is the Grassmann higher-order tensor renormalization group (GHOTRG), a tensor-network contraction scheme that compresses coarse-grained tensors by singular-value decompositions while tracking Grassmann signs through auxiliary link variables. The paper adds a modified truncation: link indices are grouped according to the occupation of adjacent plaquettes, and separate SVD truncations are performed on each group so that contributions of different orders in $\beta$ are never mixed during coarsening. Plaquette contractions into links and links into sites are handled by site-dependent tensors, which is what allows all coefficients up to a given order to be computed simultaneously. Before any of this, all $SU(N_c)$ gauge integrals are performed exactly using generalized Weingarten functions, which eliminates all color degrees of freedom and keeps the initial bond dimension small.

What would settle it

Compute the $\beta^4$ coefficient of the quark number density with the same order-separating method: the claimed absence of contamination would be falsified if the $\beta^3$ coefficient changes when $n_{\max}$ is raised from 3 to 4, and the $\tanh$ extrapolation would be falsified if the cubic-parameter fit fails to predict the $\beta^4$ coefficient. A complementary check would compare the direct tensor coefficients at $\beta\le0.1$ with independent Monte Carlo or dual-representation data at the same lattice size and quark mass.

Watch

Extended reading notes

Core claim

The central claim is that the modified Grassmann tensor renormalization procedure computes the strong-coupling coefficients of the quark number density and chiral condensate through order $\beta^3$ with complete absence of higher-order terms infiltrating the result. Working at expansion order $n_{\max}=3$ on a $32\times32$ lattice with quark mass $m=0.5$ and bond dimension $D=135$, the paper reports coefficients $\rho_n(\mu)$ and $\langle \bar\psi\psi\rangle_n(\mu)$ for $n=0,\ldots,3$ that are stable when the truncation order is raised and consistent between neighboring bond dimensions. All coefficients are fitted simultaneously to a Fermi-Dirac-like $\tanh$ ansatz for the quark number density and an analogous $\tanh$ form for the chiral condensate, with the transition parameters expanded as power series in $\beta$ through third order. The fit reproduces the tensor data within the assumed errors, and the extracted critical chemical potentials for the two transitions agree within errors, with second- and third-order terms giving only small corrections. The paper's claim is that these fits remain reliable beyond $\beta=0.1$, where the direct expansion itself becomes unphysical.

Load-bearing premise

The conclusions rest on the assumption that the quark number density and chiral condensate are well described, over the fitted range, by the hyperbolic-tangent forms (8) and (11) with their parameters expanded as power series in $\beta$ through third order; if that functional form is wrong, the extracted critical chemical potential and its $\beta$ expansion are artifacts of the fit even though the underlying tensor coefficients may be correct.

Editorial extensions

If this is right

  • The coefficients $\rho_n(\mu)$ and $\langle \bar\psi\psi\rangle_n(\mu)$ for $n\le3$ are strong-coupling reference data at finite chemical potential that other computational approaches can be checked against.
  • The fitted expansions give a quantitative prediction that the critical chemical potential increases with the coupling $\beta$ in both the number-density and chiral transitions, with the two estimates agreeing within errors.
  • The method is stated to generalize straightforwardly to an arbitrary number of staggered quark flavors, extending the same order-separated tensor construction beyond the single-flavor case.
  • Because the direct expansion fails for $\beta>0.1$ while the $\tanh$ fits remain stable, the fit-based extrapolation is the paper's route to accessing larger couplings from strong-coupling data.

Reading between the lines

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

  • Beyond the paper, the same order-separating Grassmann tensor construction should transfer to other lattice gauge theories whose Boltzmann weight can be expanded and whose gauge integrals admit a Weingarten-type solution, making the method a general template for finite-density strong-coupling computations.
  • Beyond the paper, watching the fitted transition sharpness $a_\rho(\beta)$ and $a_{cc}(\beta)$ as $\beta$ grows could indicate whether the crossover sharpens into a genuine phase transition, a question the paper does not address.
  • Beyond the paper, the contamination-free coefficients from this small two-dimensional system are natural benchmark data for sign-problem-avoiding algorithms, since they carry no Monte Carlo statistical error or phase ambiguity.
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 / 3 minor

Summary. The manuscript presents a Grassmann tensor-network treatment of two-dimensional lattice QCD with staggered quarks in the strong-coupling expansion at finite chemical potential. The Boltzmann factor is expanded in the gauge and fermion actions, gauge fields are integrated exactly, and the partition function coefficients in powers of beta are evaluated with a modified GHOTRG that is claimed to separate orders in beta exactly. Results are shown for the quark number density and chiral condensate up to order beta^3 on a 32x32 lattice, together with tanh-based fits that extrapolate the transition parameters beyond beta=0.1. The paper is a proceedings contribution; details of the modified algorithm are deferred to Ref. [4].

Significance. If the order-separation claim is correct, the paper offers a practical route to exact-in-beta strong-coupling coefficients for 2d QCD at finite mu, with gauge degrees of freedom integrated out before tensor renormalization and with no sign problem. The coefficients rho_n(mu) and <psi psi>_n(mu) are computed, not fitted, and the consistency of n_max=3 with lower orders is encouraging. The main caveats are that the 'complete absence' of higher-order contamination is only asserted, and that the physical extrapolations rely on an ad hoc tanh ansatz rather than a derived beta expansion. The strength of the computation is its direct, non-perturbative-in-Grassmann method; its weakness is the lack of independent validation of the central algorithmic claim within this manuscript.

major comments (3)
  1. [Section 4 and abstract] The abstract and Section 4 state that the modified GHOTRG procedure computes the mu-dependent quark number density and chiral condensate 'up to order beta^3 with complete absence of higher-order terms infiltrating the result.' This is the load-bearing claim, but the manuscript does not demonstrate it. Section 4 describes the order-grouping and separate SVD truncations only verbally and defers the detailed proof and implementation to Ref. [4], which is 'in preparation'. Since GHOTRG with D=135 is an approximate algorithm, it is not self-evident that finite bond-dimension truncations performed in one order sector cannot mix different orders; the consistency check between n_max<3 and n_max=3 runs in Sec. 6.1 is internal to the same algorithm and does not exclude a common contamination. Consequently, the coefficients in Figs. 3 and 4, and all mu_c(beta) results in Fig. 5, inherit this unverified prerequisite. I ask for an independent validation, for example by comparing against a brute-force evaluation of the truncated strong-coupling expansion on a small lattice, or by demonstrating numerically that no terms of order beta^4 survive in rho_3(mu).
  2. [Section 6.1, Eqs. (8)-(9)] The 'valuable expansion in beta for the critical chemical potential' reported in Sec. 6.3 is not a derived strong-coupling quantity but an output of the tanh fit ansatz (8) with the power series (9). The fit parameters mu_c,rho,n and a_rho,n are introduced by hand, and the error bands in Fig. 4 and the extrapolation beyond beta=0.1 inherit the validity of this functional form. The paper does not test the ansatz against an alternative (e.g., the antisymmetric combination acknowledged in footnote 6), nor does it show how the extracted mu_c(beta) changes when n_max is increased from 3 to 4. Without such a stability check, the statement that 'the critical chemical potential remains quite accurate' for beta up to 1 is not supported.
  3. [Section 6.1, Figs. 3-4] All numerical results use a single lattice size (32x32) and a single bond dimension (D=135). No analysis is shown of the dependence of rho_n(mu) or <psi psi>_n(mu) on D or lattice volume, so the reader cannot tell whether the reported coefficients are converged with respect to the tensor-network truncation. This matters because the claim that the coefficients are free of higher-order contamination is tied to the exactness of the order separation; if the separation is only approximate at finite D, the truncation error in each sector has to be quantified.
minor comments (3)
  1. [Section 6.1] The choice of 1% of the maximal value of |rho_n(mu)| as the error for each data point is not motivated; the resulting covariance matrix and error bands in Figs. 4 and 5 are sensitive to this choice. Please state how this uncertainty estimate was calibrated.
  2. [Section 5] The introduction of the four special tensors and the 'impurity system' is not illustrated; a figure or explicit tensor diagram would help the reader follow the translational-invariance reduction.
  3. [Section 6.1, Fig. 3] The claim of 'excellent agreement' between tensor data and the simultaneous fit would be easier to assess with a residual plot or a quantitative goodness-of-fit statistic, especially because the error bars are assigned rather than estimated from the tensor computation.

Circularity Check

1 steps flagged · score 4.0 of 10

Central order-separation claim rests on an unverified self-citation; the fitted mu_c(beta) expansion is an honest fit output, not an independent prediction.

  1. self citation load bearing [Section 4 'Separation of different orders in beta' and Introduction (full details deferred to ref. [4]).]
    "For full details we refer to a forthcoming publication [4]. ... Therefore, we developed a modified GHOTRG procedure [4], where terms contributing to orders higher than nmax are avoided in each coarsening step. ... This method allows for a direct calculation of all coefficients of the strong-coupling expansion of Z_QCD up to order nmax simultaneously."

    The paper's load-bearing assertion, that the modified GHOTRG produces the mu-dependent coefficients up to order beta^3 'with complete absence of higher-order terms infiltrating the result,' is not demonstrated here; it is delegated to [4], a forthcoming paper by the same three authors. The only consistency check reported is between nmax<3 and nmax=3 runs of the same algorithm, which cannot exclude contamination by higher orders or leakage across SVD truncation sectors at finite bond dimension D=135. Every downstream result, including the error bands in Fig. 4 and the beta expansion of mu_c in Fig. 5, depends on this unverified premise, so the central claim reduces to an unverified self-citation.

full rationale

The strong-coupling coefficients rho_n(mu) and <bar psi psi>_n(mu) are computed from the theory by integrating out gauge and Grassmann fields and running GHOTRG, not obtained by fitting; the fit parameters mu_c,n and a_n are explicitly determined from those tensor data via the tanh ansatz (8)/(11) and are labeled as fits in Sec. 6. The resulting mu_c(beta) curve is the polynomial built from the fitted coefficients, so it is a fit output rather than a first-principles prediction; this is circularity-free because the paper says so. The ansatz limitations in footnotes 6 and 7 are acknowledged. The only genuine circularity concern is the central order-separation step in Sec. 4, which is justified solely by a same-author forthcoming reference [4] and is not independently verified in this manuscript. That warrants a moderate score of 4: some self-citation is load-bearing, but the presented tensor data and fits have independent content.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central computation of coefficients is a Taylor expansion of the theory with exact gauge integration, so the main free parameters are the fit parameters introduced in Section 6. The tanh ansatz is ad hoc to this paper. No new physical entities are postulated.

free parameters (3)
  • Fit parameters for quark number density ansatz = {mu_c_rho,0...3, a_rho,0...3} fitted to tensor coefficients
    The tanh ansatz (8) depends on mu_c_rho(beta) and a_rho(beta), each expanded as a power series in beta up to order 3; the eight coefficients are fitted simultaneously to the tensor data in Fig. 3.
  • Fit parameters for chiral condensate ansatz = {mu_c_cc,0...3, a_cc,0...3, b_cc,0...3} fitted to tensor coefficients
    The condensate ansatz (11) has three functions mu_c_cc, a_cc, b_cc, each expanded to order 3; twelve coefficients are fitted simultaneously to the tensor coefficients of Eq. (10).
  • Bond dimension D=135 for n_max=3 = 135
    The truncation dimension is chosen for the order-3 computation; convergence is not demonstrated for this value at n_max=3, and lower orders converge at smaller D.
assumptions (4)
  • domain assumption Strong-coupling expansion is performed by Taylor expanding the plaquette action and truncating at order n_max per plaquette, with a modified contraction to remove higher-order contamination.
    Section 4 states this expansion and the modified procedure; it is the basis for computing the beta coefficients of the partition function.
  • standard math Gauge field integrals are evaluated using generalized Weingarten functions given in Eq. (6), following Refs. [7,8].
    Section 3 uses these group-integral results to integrate out all gauge fields exactly.
  • ad hoc to paper The quark number density and chiral condensate are well described by the tanh ansatze (8) and (11), with parameters expanded as power series in beta up to the same order as the expansion.
    Adopted in Section 6 to model the mu dependence of the coefficients and to extrapolate beyond beta=0.1; the paper itself notes that a symmetrized combination would be a better ansatz.
  • domain assumption HOTRG/GHOTRG truncation with bond dimension D=135 is a controlled approximation to the exact tensor contraction.
    This numerical control parameter is used throughout; convergence is checked for n_max<3 but not for n_max=3 at D=135.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Grassmann tensor approach for two-dimensional QCD in the strong-coupling expansion." pith.science (2026). https://pith.science/paper/6M7T4Z22

@misc{pith2026250119192,
  author       = {Pith},
  title        = {Pith review of: Grassmann tensor approach for two-dimensional QCD in the strong-coupling expansion},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6M7T4Z22}},
  note         = {Machine review of arXiv:2501.19192}
}
abstract

We present a tensor-network approach for the strong-coupling expansion of two-dimensional QCD with staggered quarks at non-zero chemical potential. After expanding the Boltzmann factor in the gauge and fermion actions, all gauge fields can be integrated out exactly and the partition function can be evaluated using the Grassmann higher-order tensor renormalization group approach. The method is modified to compute the $\mu$ dependence of the quark number density and the chiral condensate up to order $\beta^3$ with complete absence of higher-order terms infiltrating the result. Although the expansion itself is only a good approximation to the full theory at small $\beta<0.1$, the range can be extended, by using judiciously chosen fits. Moreover, these fits also yield a valuable expansion in $\beta$ for the critical chemical potential.

Figures

Figures reproduced from arXiv: 2501.19192 by the authors.

Figure 1
Figure 1. Schematic contraction procedure for a two-dimensional tensor network. Each contraction step results in a new tensor network defined on a coarse lattice with half of the number of sites than before, but increased bond dimension of the tensor. 3. LQCD partition function as tensor network Here, we want to apply the HOTRG approach to the partition function of two-dimensional lattice QCD, given by 𝑍QCD = ∫ "Ö 𝑥 𝑑𝜓𝑥 𝑑𝜓¯ 𝑥… view at source ↗
Figure 2
Figure 2. Visualization of all contributions using the example 𝑛max = 2. Even though no plaquette has a contribution higher than 𝑛max = 2, the partition function contains terms of much higher order, since the expansion is done on every plaquette separately. These terms are avoided in our modified GHOTRG procedure. 5. Translational invariance Translational invariance is a symmetry of the system that can be used to reduce the n… view at source ↗
Figure 3
Figure 3. Results for the coefficients 𝜌𝑛 (𝜇) of the quark number density in the strong-coupling expansion up to order three. We observe excellent agreement between tensor data (blue crosses) and the simultaneous fit (solid lines). Dirac statistics,6 𝜌(𝛽, 𝜇) = 3 2  1 + tanh[𝑎𝜌 (𝛽) (𝜇 − 𝜇 c 𝜌 (𝛽))] , (8) which depends on a critical chemical potential 𝜇 c 𝜌 (𝛽) and a transition sharpness 𝑎𝜌 (𝛽). Using 𝜇 c 𝜌 (𝛽) = 𝑛∑︁max 𝑛=0 𝜇… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Quark number density (top) and chiral condensate (bottom), up to order three, as a function of 𝜇 for 𝑚 = 0.5, 𝑉 = 32 × 32, bond dimension 𝐷 = 135, and several 𝛽 values. The solid lines show the medians of the distributions of 𝜌(𝛽, 𝜇) and ⟨𝜓𝜓¯ ⟩ (𝛽, 𝜇). The error bands …
Figure 5
Figure 5. Figure 5: Critical chemical potential and transition sharpness versus coupling parameter 𝛽 for the transitions in the quark number density and the chiral condensate. The plots on the left show a close range 0 ≤ 𝛽 ≤ 0.1, while those on the right show a wider range up to 𝛽 = 1. pa…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

10 extracted references · 4 canonical work pages

  1. [4]

    Milde, Tensor networks for four-dimensional strong-coupling QCD , Master's thesis, University of Regensburg, 2023

    P. Milde, Tensor networks for four-dimensional strong-coupling QCD , Master's thesis, University of Regensburg, 2023

    work page 2023

  2. [1]

    write newline

    " write newline "" before.all 'output.state := FUNCTION blank.sep after.quote 'output.state := FUNCTION fin.entry output.state after.quoted.block = 'skip 'add.period if write newline FUNCTION new.block output.state before.all = 'skip output.state after.quote = after.quoted.block 'output.state := after.block 'output.state := if if FUNCTION new.sentence out...

  3. [2]

    Bloch and R

    J. Bloch and R. Lohmayer, Grassmann higher-order tensor renormalization group approach for two-dimensional strong-coupling QCD , https://doi.org/10.1016/j.nuclphysb.2022.116032 Nucl. Phys. B 986 (2023) 116032 [ https://arxiv.org/abs/2206.00545 2206.00545 ]

    work page arXiv 2023

  4. [3]

    Grassmann tensor-network method for strong-coupling QCD

    J. Bloch and R. Lohmayer, Grassmann tensor-network method for strong-coupling QCD , https://doi.org/10.22323/1.430.0004 PoS LATTICE2022 (2022) 004 [ https://arxiv.org/abs/2210.08935 2210.08935 ]

    work page Pith review arXiv 2022

  5. [5]

    Samberger, J

    T. Samberger, J. Bloch and R. Lohmayer, Tensor-network formulation of two-dimensional QCD in the strong-coupling expansion , in preparation, 2025

    work page 2025

  6. [6]

    Z.Y. Xie, J. Chen, M.P. Qin, J.W. Zhu, L.P. Yang and T. Xiang, Coarse-graining renormalization by higher-order singular value decomposition, https://doi.org/10.1103/physrevb.86.045139 Phys. Rev. B86 (2012) 045139 [ https://arxiv.org/abs/1201.1144 1201.1144 ]

    arXiv 2012

  7. [7]

    De Lathauwer, B

    L. De Lathauwer, B. De Moor and J. Vandewalle, A multilinear singular value decomposition, https://doi.org/10.1137/S0895479896305696 SIAM Journal on Matrix Analysis and Applications 21 (2000) 1253

    doi:10.1137/s0895479896305696 2000

  8. [8]

    Borisenko, S

    O. Borisenko, S. Voloshyn and V. Chelnokov, SU(N) polynomial integrals and some applications , https://doi.org/10.1016/S0034-4877(20)30015-X Rept. Math. Phys. 85 (2020) 129 [ https://arxiv.org/abs/1812.06069 1812.06069 ]

    arXiv 2020

Show all 10 references
  1. [9]

    Gagliardi and W

    G. Gagliardi and W. Unger, New dual representation for staggered lattice QCD , https://doi.org/10.1103/PhysRevD.101.034509 Phys. Rev. D 101 (2020) 034509 [ https://arxiv.org/abs/1911.08389 1911.08389 ]

    2020 arXiv

  2. [10]

    Shimizu and Y

    Y. Shimizu and Y. Kuramashi, Grassmann tensor renormalization group approach to one-flavor lattice Schwinger model , https://doi.org/10.1103/PhysRevD.90.014508 Phys. Rev. D90 (2014) 014508 [ https://arxiv.org/abs/1403.0642 1403.0642 ]

    2014 arXiv

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.