Topology in 2D non-Abelian Lattice Gauge Theories

Philip Rouenhoff; Stephan Durr

REVIEW 2 major objections 6 minor 7 references

Topology in 2D non-Abelian Lattice Gauge Theories

T0 review · 2 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read This paper constructs explicit q-instanton configurations in 2D U(2) lattice gauge theory and argues they are the global action minima in their topological sectors.

desk verdict Explicit U(2) instanton configurations that saturate the action bound, with one unproved but true inequality as the load-bearing step; a solid proceedings paper after a minor revision. read the letter →

arxiv 2411.11593 v1 pith:J36W7DL6 submitted 2024-11-18 hep-lat

Stephan Durr , Philip Rouenhoff This is my paper · ORCID

classification hep-lat PACS 11.15.Ha

keywords latticegaugetheoryU(2)topologicalchargeinstantongradientflowfreezingWilsonactiontwo-dimensional

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 claims that in two-dimensional $U(2)$ lattice gauge theory every topological sector with integer charge $q$ contains an explicit link configuration, Eq. (7), whose Wilson action is $N_x N_t (1 - \cos(\pi q/(N_x N_t)))$ and that this is the minimal action possible in that sector. These are lattice instantons with the unusual feature that the action density is uniform over all plaquettes. If the claim is right, gradient flow inside a fixed sector always lands on one of these configurations, giving a concrete picture of the phase space and of why topological freezing occurs. The paper also constructs for arbitrary $N_c$ a family of 'special configurations' with uniform action density, labeled by $(q,z)$, which coincide with the sector minima for $N_c=2$ only when $q$ is even.

What carries the argument

The construction is based on the 2D $U(1)$ instanton of Eq. (6), a set of links whose phases wind once around the torus, with extra $SU(2)$-valued defect factors inserted on the last $x$- and $t$-slices. The vectors $\vec u,\vec v\in\mathbb R^3$ and their orthogonality or parallel constraints are the mechanism that cancels the $\mathbb Z_2$ ambiguity left by taking the square root of the $U(1)$ phase, so that the corner plaquette matches all others. The result is a configuration whose untraced plaquette is the same group element everywhere, hence uniform action density, which the paper identifies as the requirement for a local action minimum in two dimensions. Gradient flow serves as the test: the excess action above Eq. (8) decays to zero as flow time increases, connecting the constructed solutions to thermalized configurations.

What would settle it

Test the bound directly on a small lattice: minimize the Wilson action over $U(2)$ link configurations with fixed topological charge, say $q=1$ on a $4\times4$ or $8\times8$ lattice, using simulated annealing or many random gradient-flow starts. Any configuration whose action falls strictly below $N_x N_t (1 - \cos(\pi q/(N_x N_t)))$ disproves the central claim; a proof of the inequality for all $q$ and all lattice sizes would confirm it.

Watch

Extended reading notes

Core claim

The central claim is that Eq. (7) gives exact $q$-instanton configurations in 2D $U(2)$ lattice gauge theory: for any integer $q$, take horizontal links $e^{-i t \pi q/(N_x N_t)}$ with an extra $SU(2)$ factor $\exp(i\vec u\cdot\vec\sigma)$ on the last $x$-slice, and vertical links $e^{i x \pi q/N_x}$ with $\exp(i\vec v\cdot\vec\sigma)$ on the last $t$-slice; require $\vec u\perp\vec v$ and $|\vec u|=|\vec v|=\pi/2$ for odd $q$, or $\vec u\parallel\vec v$ for even $q$. With these constraints every untraced plaquette takes the same value $e^{i\pi q/(N_x N_t)}$, so the action density is exactly uniform. The paper asserts the resulting action $S/\beta = N_x N_t(1-\cos(\pi q/(N_x N_t)))$ is the lower bound for charge $q$, and it uses gradient flow to show thermalized configurations evolve toward these solutions. The 'special configurations' of Eq. (9) have uniform action density for all $N_c$ but only match the instanton action for $N_c=2$ when $q$ is a multiple of 2; for other $q$ a single-link perturbation in certain color directions makes them flow down to the true sector minimum.

Load-bearing premise

The construction's status as the sector minimum rests on the unproved inequality that every $U(2)$ lattice configuration with topological charge $q$ has Wilson action at least $N_x N_t (1 - \cos(\pi q/(N_x N_t)))$; if that bound fails, the configurations of Eq. (7) may exist but would not be the minimal-action fields in their sector.

Editorial extensions

If this is right

Gradient flow in a fixed topological sector ends at a gauge-transformed copy of the explicit configuration (7), so the low-action part of each sector has a single known attractor rather than an unknown landscape.
The action formula (8) supplies an analytic value for the minimum action in each sector, so the gap between neighboring sectors gives a lower bound on the barrier height relevant to topological freezing.
For $N_c=2$, the 'special configurations' (9) include the true instantons only for even $q$; a tiny single-link perturbation in the $\sigma_1$ or $\sigma_2$ color direction makes gradient flow descend from a special configuration to the sector minimum.
The instantons constructed here have uniform action density, unlike instantons in four dimensions where the density is localized; this is a direct property of the solutions themselves.

Reading between the lines

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

If the lower bound (8) is a genuine inequality, it likely comes from a topological identity, a lattice analogue of a Bogomol'nyi bound; finding a sum-of-positive-terms proof would simultaneously establish the claim and suggest how the construction generalizes to $N_c\ge3$.
The plateaus seen when a single link is perturbed suggest the special configurations are saddle points or quasi-stationary states that organize the flow toward the sector minimum; mapping their basin structure could turn topological freezing into a rare-event problem with known transition states.
A natural numerical test beyond the paper: measure the flow time needed to reach Eq. (7) from thermalized configurations as the lattice spacing shrinks; if it diverges, the 'trivialization' within a sector is only practical at finite lattice spacing.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

The genuinely new thing here is the explicit U(2) instanton construction, Eq. (7): link fields with topological charge q and uniform action density that saturate the lower bound Eq. (8). The SU(2) dressing vectors u and v, with the odd/even q constraints, are a real addition over the U(1) formula from Smit and Vink. I verified the algebra myself: every untraced plaquette equals exp(i pi q / (Nx Nt)) times the identity, so the action density is exactly the claimed bound. That is a clean, checkable result, and the Monte Carlo comparison against the Bonati-Rossi analytic plaquette and susceptibility is honest. Those formulas come from the literature, not from a fit to the data, and the agreement in Table 1 is good. The secondary construction, the family of special configurations (9) for arbitrary N_c, is also new and interesting, even though the paper correctly notes it does not give q-instantons for q not a multiple of N_c. The gradient-flow figures are suggestive and the discussion of plateaus at the special-configuration action levels is a nice observation. The paper is candid that the local-minimum status of those configurations is open, and that the N_c >= 3 instanton problem is unsolved. The soft spot is structural, and the stress-test note puts it exactly right: Eq. (8) is asserted in Sec. 2 without proof or citation, and the claim that (7) are global minima of their charge sector rests entirely on it. The bound is true. A one-line convexity argument on the determinant phase of each plaquette gives exactly S/beta >= V (1 - cos(pi q / V)), but the manuscript does not supply that argument. As written, the conclusion that these are exact q-instanton configurations is not fully supported by the text alone, even though the mathematics is sound. That is a missing derivation, not a flaw in the result. A referee should ask for the few lines. The gradient-flow evidence that thermalized configurations flow to these minima is numerical and suggestive, not a proof, which the paper itself does not overstate. This is a conference proceedings from Lattice 2024. It deserves a serious referee, not because it resolves any big open question, but because the explicit U(2) instanton family is a useful, checkable tool for future work on topological freezing in 2D. I would cite it if I worked on this toy model, and I would bring it to a reading group focused on lattice topology. Recommend: send to peer review; require the proof of Eq. (8) in the revision.

Referee Report

2 major / 6 minor

Summary. The manuscript investigates topological sectors in two-dimensional U(N_c) lattice gauge theory on a torus, with N_c=2 as the main case. After comparing Monte Carlo data for the average plaquette and topological susceptibility with analytic formulas from refs. [4,5], it constructs explicit link configurations (7) whose plaquettes are all equal to exp(i pi q/(N_x N_t)) times the identity, and asserts that these configurations saturate the action lower bound (8). It also introduces a one-parameter family of uniform-action 'special configurations' (9) and studies their behavior under gradient flow, concluding that they flow to the sector minimum when perturbed in certain color directions. The central claim is that (7) are exact q-instantons for 2D U(2) lattice gauge theory.

Significance. The explicit construction (7) is elegant and potentially useful for understanding topological freezing and for benchmarking algorithms. Its uniform action density and exact saturation of the conjectured lower bound are striking. The paper is also honest in its numerical comparisons: the analytic formulas (4)-(5) come from the literature and no parameter is fitted to produce the central claim. The gradient-flow studies in Secs. 3-4 are exploratory and are largely framed as such. If the inequality (8) is supplied with a proof, the main result is sound and would be a useful contribution to the lattice-topology literature.

major comments (2)

[Sec. 2, Eq. (8)] The lower bound S/beta >= N_x N_t (1 - cos(pi q/(N_x N_t))) is stated without proof or citation, and the conclusion in Sec. 5 that (7) are exact q-instantons, i.e., global minima of the Wilson action in their topological sector, rests entirely on this inequality. Without it, (7) would only be a homogeneous low-action configuration. The bound is in fact true (for example, by writing each plaquette as e^{i delta_n} times an SU(2) part, bounding Re Tr(1-U_box) below by 2(1-cos(delta_n/2)), and using convexity), but the manuscript should provide this derivation or a precise reference. Please add it before the paper claims minimality.
[Sec. 2, around Eq. (7)] The statement that 'each untraced plaquette takes the same value e^{i pi q/(N_x N_t)}' is asserted but not demonstrated. In particular, the corner plaquette at (x,t)=(N_x,N_t) contains the product of the two SU(2) dressing factors and a U(1) phase e^{-i pi q}; the cancellation for odd q relies on the u perpendicular v constraint with |u|=|v|=pi/2. A short explicit computation of the corner plaquette would make the construction self-contained and remove any doubt about the claimed uniform action density.

minor comments (6)

[Eqs. (6)-(7)] The Kronecker deltas are typeset ambiguously (e.g., 'delta_{t, N_t}' appears as '𝛿𝑡, 𝑁𝑡'); please use delta_{t,N_t} and delta_{x,N_x} throughout.
[Eq. (4)] The phrase 'withthe2Dconvention' should read 'with the 2D convention'; there are several missing spaces in the extracted text, so a careful proofread of the final PDF is needed.
[Sec. 2, bullet after Eq. (7)] For even q, 'require only u parallel v' leaves the magnitudes of u and v unspecified; the following paragraph states they may be chosen freely, but the bullet should say so explicitly.
[Abstract] 'With the help of gradient flow we derive instanton-like solutions' is misleading, since (7) is constructed analytically; gradient flow is used only as a check.
[Sec. 3, after Eq. (9)] The first use of z should state z in Z explicitly, since the range matters for the 'infinite tower' statement.
[Eq. (6)] The label S^{SU(1)}_{inst} is odd because the gauge group is U(1); rename to S^{U(1)}_{inst}.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: the instanton construction and Monte Carlo comparisons are self-contained and externally anchored, with the unproved action bound being a missing proof rather than a circular input.

full rationale

The paper's central new result is the explicit construction (7) of q-instanton configurations for 2D U(2) lattice gauge theory. This construction is derived directly from the definitions of the topological charge (1), the plaquette (2), and the Wilson action (3), and it is verified by direct computation: every untraced plaquette of (7) equals exp(i π q/(N_x N_t)), so the action (8) follows by substitution. No parameter is fitted to the Monte Carlo data to produce this construction; the data in Table 1 and Fig. 1 are compared with, not used to fit, the external analytic formulas (4)-(5) from Refs. [4,5]. The gradient-flow tests in Figs. 3, 5, and 6 compare thermalized or perturbed configurations against the explicit action formula (8) and do not supply the minimized quantity as an input. The only potentially load-bearing step is the inequality stated as the lower bound (8), which is asserted without derivation and is used to justify that (7) is globally minimal in its charge sector rather than merely low-action. This is a gap in proof, not circularity: the bound is not fitted to data, is not imported from the same authors' prior work, and is not defined in terms of the target configurations. It is an independent mathematical claim that could be proven by a convexity argument, so its absence affects rigor but not the circularity status of the derivation. The paper also honestly reports open questions, including whether the special configurations (9) are local minima, and does not disguise these as established results. There is no self-citation load-bearing chain, no renaming of known results as new organization, and no fitted input relabeled as a prediction. Accordingly, no significant circularity is present, and the appropriate score is 0.

Assumptions & free parameters 1 free parameters · 3 assumptions · 0 invented entities

The central construction depends on the unproved action lower bound (8) and on standard properties of the topological charge. No constants are fitted to data; the continuous parameters in the even-q instanton family do not affect the central claim.

free parameters (1)

SU(2) dressing vectors u, v in Eq. (7) for even q
Continuous parameters that label distinct minimal-action configurations for even topological charge q; the paper notes they cannot be removed by gauge transformations. Their values do not affect the action or charge, so the central claim does not depend on them.

assumptions (3)

domain assumption Wilson action lower bound Eq. (8) for fixed topological charge q
Invoked in Sec. 2 to establish that (7) is of minimal action; not proved or cited in the paper.
standard math Integer-valuedness of the topological charge definition (1)
Standard for U(N) on the torus; used throughout the paper.
domain assumption Gauge equivalence of all odd-q instanton choices in (7)
The paper states this without proof in Sec. 2; it is used to argue uniqueness up to gauge for odd q.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Topology in 2D non-Abelian Lattice Gauge Theories." pith.science (2026). https://pith.science/paper/J36W7DL6

@misc{pith2026241111593,
  author       = {Pith},
  title        = {Pith review of: Topology in 2D non-Abelian Lattice Gauge Theories},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/J36W7DL6}},
  note         = {Machine review of arXiv:2411.11593}
}

abstract

In two dimensions, $U(N_c)$ gauge theories exhibit a non-trivial topological structure, while $SU(N_c)$ theories are topologically trivial. Hence, for $G = U(N_c)$ the phase space is divided into topological sectors, characterized by a topological index (a.k.a. ``topological charge''). These sectors are separated by action barriers, which diverge if the lattice spacing is taken small, resulting in an algorithmic problem known as ``topological freezing''. We study these theories in various box sizes and at various couplings. With the help of gradient flow we derive instanton-like solutions for 2D $U(N_c)$ theory with a specific focus on the case of $N_c = 2$.

Figures

Figures reproduced from arXiv: 2411.11593 by the authors.

**Figure 1.** The action density and topological susceptibility of 2D 𝑈(2) theory as given in Tab. 1. In the limit 𝛽 → ∞ we find that 𝑠wil/𝑔 2 → 1/2 and 𝜒top/𝑔 2 → 0.02533 = 1/(2𝜋) 2 . 2. Global minima per topological sector In 2D 𝑈(1) theory formulas for instanton configurations, i.e. field configurations of minimal action and fixed topological charge 𝑞 ∈ Z, are known as [7] 𝑈𝑥 (𝑥, 𝑡) = e −i𝑡 2𝜋𝑞 𝑁𝑥 𝑁𝑡 , 𝑈𝑡(𝑥, 𝑡) = e i𝑥 2𝜋𝑞 𝑁𝑥 𝛿… view at source ↗

**Figure 2.** Left: A visualization of (7), the instanton-like solution for 2D 𝑈(2) theory with topological charge 𝑞. Right: the same configuration in maximal tree gauge. Unity matrices are shown as dashed lines. 3 [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** Left: Action of the 𝑈(2) instanton configurations (7) as well as the lower bound (8). Right: excess of the action density of five thermalized configurations (𝛽 = 6.0) over the bound (8) plotted against the gradient flow time 𝜏 = 𝜌 · 𝑁stout with two step sizes 𝜌. Square lattices of 𝐿/𝑎 = 32 are used in both cases. All vertical links are unity 1, except for those at 𝑡 = 𝑁𝑡 , which are e i𝑥 𝜋𝑞 𝑁𝑥 exp(i𝑣®𝜎® ). Despite t… view at source ↗

Figures from the paper (4 more)

**Figure 4.** Figure 4: The action of (9), evaluated for various (𝑞, 𝑧)-combinations (𝑁𝑐 = 2). The lower bound is Eq. (8). 3. Special topological configurations For 2D 𝑈(𝑁𝑐) theory one may derive further configurations of homogeneous action density 𝑈𝑥 (𝑥, 𝑡) = e −i𝑡 2𝜋𝑞 𝑁𝑐 1 𝑁𝑥 𝑁𝑡 exp −i𝑡 2…

**Figure 5.** Figure 5: The action of the “special configuration” (9) with 𝑁𝑐 = 2 and (𝑞, 𝑧) = (1, 5) under gradient flow, after each link has been multiplied with a random 𝑈(2)-element of step size 𝜀, for 𝑁𝑥 = 𝑁𝑡 = 32. The lower bound is Eq. (8). Without perturbation the configuration (9) se…

**Figure 6.** Figure 6: The action of the “special configuration” (9) with 𝑁𝑥 = 𝑁𝑡 = 32, 𝑁𝑐 = 2 and (𝑞, 𝑧) = (1, 5) under gradient flow, after an arbitrary link has been multiplied with a 𝑈(2)-element of the form exp(i𝜀𝜎𝑗/2). The grey, dotted lines show the action of (9) for 𝑞 = 1 and various…

**Figure 7.** Figure 7: Left: a link of a “special configuration” (9) with 𝑁𝑥 = 𝑁𝑡 = 32, 𝑁𝑐 = 2 and (𝑞, 𝑧) = (1, 5) is perturbed by a 𝑈(2)-element of step size 𝜀1 and Δ𝑆 is measured in units of machine precision 𝜀machine. Right: after a perturbation 𝜀1 = 10−3 along an algebra direction 𝑗1, a …

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

7 extracted references · 5 canonical work pages

[1]

Baig and J

M. Baig and J. Colet,Monte Carlo Analysis of the SU(2) x U(1) and U(2) Lattice Gauge Theories at Different Space-time Dimensionalities, Phys. Lett. B176, 149-152 (1986)

work page 1986
[2]

On the U(2) Lattice Gauge Theory

C. Roiesnel,On the U(2) lattice gauge theory, [arXiv:hep-lat/9509092 [hep-lat]]

work page Pith review arXiv
[3]

B. C. Hall,Lie Groups, Lie Algebras, and Representations: An Elementary Introduction, Springer International Publishing, Switzerland, 2015

work page 2015
[4]

The topological susceptibility of two-dimensional $U(N)$ gauge theories

C. Bonati and P. Rossi,Topological susceptibility of two-dimensional𝑈(𝑁) gauge theories, Phys. Rev. D99, no.5, 054503 (2019) [arXiv:1901.09830 [hep-lat]]

work page Pith review arXiv 2019
[5]

C.BonatiandP.Rossi, Topologicaleffectsincontinuumtwo-dimensional 𝑈(𝑁)gaugetheories, Phys. Rev. D100, no.5, 054502 (2019) [arXiv:1908.07476 [hep-th]]

arXiv 2019
[6]

Hirasawa, A

M. Hirasawa, A. Matsumoto, J. Nishimura and A. Yosprakob,Tensor renormalization group and the volume independence in 2D U(N) and SU(N) gauge theories, JHEP12, 011 (2021) [arXiv:2110.05800 [hep-lat]]

arXiv 2021
[7]

Smit and J

J. Smit and J. C. Vink,Remnants of the Index Theorem on the Lattice, Nucl. Phys. B286, 485-508 (1987). 8

work page 1987

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [1]

Baig and J

M. Baig and J. Colet,Monte Carlo Analysis of the SU(2) x U(1) and U(2) Lattice Gauge Theories at Different Space-time Dimensionalities, Phys. Lett. B176, 149-152 (1986)

work page 1986

[2] [2]

On the U(2) Lattice Gauge Theory

C. Roiesnel,On the U(2) lattice gauge theory, [arXiv:hep-lat/9509092 [hep-lat]]

work page Pith review arXiv

[3] [3]

B. C. Hall,Lie Groups, Lie Algebras, and Representations: An Elementary Introduction, Springer International Publishing, Switzerland, 2015

work page 2015

[4] [4]

The topological susceptibility of two-dimensional $U(N)$ gauge theories

C. Bonati and P. Rossi,Topological susceptibility of two-dimensional𝑈(𝑁) gauge theories, Phys. Rev. D99, no.5, 054503 (2019) [arXiv:1901.09830 [hep-lat]]

work page Pith review arXiv 2019

[5] [5]

C.BonatiandP.Rossi, Topologicaleffectsincontinuumtwo-dimensional 𝑈(𝑁)gaugetheories, Phys. Rev. D100, no.5, 054502 (2019) [arXiv:1908.07476 [hep-th]]

arXiv 2019

[6] [6]

Hirasawa, A

M. Hirasawa, A. Matsumoto, J. Nishimura and A. Yosprakob,Tensor renormalization group and the volume independence in 2D U(N) and SU(N) gauge theories, JHEP12, 011 (2021) [arXiv:2110.05800 [hep-lat]]

arXiv 2021

[7] [7]

Smit and J

J. Smit and J. C. Vink,Remnants of the Index Theorem on the Lattice, Nucl. Phys. B286, 485-508 (1987). 8

work page 1987