Monopoles, duality, and energy splitting in Chern-Simons QED$_3$

\'Eric Dupuis; Shai M. Chester

REVIEW

Monopoles, duality, and energy splitting in Chern-Simons QED$_3$

Shai M. Chester, Éric Dupuis

High Energy Physics - Theory · Submitted 29 Sep 2026

Pith's verdict: A genuinely new subleading k-neq-0 fermionic monopole calculation whose 1-7% duality matches are impressive, but the load-bearing closed-form kernels are pattern-verified rather than proven, and the m=0 failure inside the same formalism is unresolved.

T0 · AI review

How this review was made: An AI referee (deepseek/deepseek-v4-flash) read the paper on September 30, 2026 and answered a fixed list of questions about it. T0 is the starting level every paper gets. The levels describe how deeply Pith has checked the paper, never the authors.

T0 AI review An AI read the paper and answered a fixed list of questions.
T1 Audit Automatic checks that the references and files the paper relies on exist and match.
T2 Compute The paper's code or calculations re-run with the same results.
T3 Formal statement The claims written out in a formal math language a computer can read.
T4 Proof Those claims proved, and the proof checked by computer.

How Pith checks papers →

· 3 major objections · 5 minor · Ask about this paper · The compiled record · Read the full review →

Challenge this review Comment (0) Read on arXiv ↗

Cite ▾

arXiv ID2609.37795 (version 2)

Pith IDLNER3TYQ, Pith's permanent record of this paper and its review

Citation\'Eric Dupuis, Shai M. Chester. "Monopoles, duality, and energy splitting in Chern-Simons QED$_3$." arXiv:2609.37795 (2026). https://pith.science/paper/LNER3TYQ

Cite this reviewBibTeX for Pith's review of this paper

Claim paper

Why Pith says this

Highlight any text to comment on it.

Editor's note

read the letter

Two things to know. First, this is real new work: the subleading correction to monopole scaling dimensions in fermionic QED3 with nonzero Chern-Simons level did not exist before. Prior subleading results covered k=0 QED3 [22] and k-neq-0 sQED3 [19,24], so the fermionic case with the CP-violating kernels and the m=2,3 comparisons is a genuine extension, not a repackaging. Second, the numerical evidence comes out stronger than I expected. The m=1 matches -- q=1/2 at 2.09556 against the exact SO(3) current value 2, q=1 at 5.27538 against the fuzzy sphere 5.1079/5.3435, and higher q against sQED3 within roughly 1% -- are quantitatively tested claims against independent benchmarks. m=2 and m=3 land at 5% and 7%. That is the paper's best evidence and it is good evidence.

The main soft spot is exactly what the stress test flags: the closed-form matter kernels (3.25)-(3.33) are asserted by pattern matching over finitely many q and j-tilde values, and every table entry and both large-q fits pass through them. The authors disclose this plainly ("by looking at many values of q we found a compact formulation") and they add independence checks: the Appendix A small-distance expansion compared against the spectral decomposition fixes the homogeneous coefficient C_q, so the kernels are not hanging on a single unverified number. Still, a mis-parameterized Pochhammer ratio at an unexplored q would silently shift digits. I would not call this disqualifying -- it is the established pipeline for this program -- but a referee should demand an out-of-sample check on a handful of q values not used in the pattern search.

The m=0 failure is the second genuine concern. The 46-60% errors against the critical O(2) values sit inside the same formalism as the accepted matches, and the observation that the subleading term alone matches O(2) while the leading term ruins it is unexplained. The authors say so themselves. It is honestly reported and structurally different (filled lowest Landau level, leading term dominates), but it does cap how much confidence the m=1-3 agreement earns.

The q^0 results are fit outputs with empirical error bars only, and the authors themselves flag that the value may not survive the N,k to finite extrapolation -- indeed if it did, it would contradict the m=1 duality. So the universal-superfluid-number-for-parity-breaking claim is real but bracketed.

Who this is for: anyone testing 3d bosonization dualities or computing monopole data. It deserves a serious referee; the main thing to push on is the kernel closed forms. Send it out.

Referee report

Significance

If the computation is correct, this is the first dynamical field-theoretic check of the spectrum of the m=1 (k=-3/2 QED3, SO(3)-enhanced) duality, a regime where lattice methods fail due to the sign problem; the q=1/2 result 2.09556 being within ~5% of the exact current dimension 2 is suggestive, and the q=1 value 5.27538 sits between the two fuzzy-sphere predictions (5.1079, 5.3435). The claim that the parity-breaking QED3 large-charge sector still exhibits the parity-preserving superfluid q^0 coefficient, while sQED3 at κ≠0 does not, is a striking and falsifiable observation. Strengths I credit explicitly: the calculation follows the established pipeline of [18,22,23,27] with explicit cutoffs, zeta regularization, and stated numerical precision; the duality benchmarks (fuzzy sphere [17], lattice [37], bootstrap [38,49], and the large-charge universal value [44,45]) are external and independent; the paper is transparent about its pattern-matching step and supplies the Mathematica notebook; and the m=0 failure and the q^0 tension with the m=1 duality are acknowledged rather than hidden.

Major objections (3)

[§3.3, Eqs. (3.25)-(3.33)] The closed-form matter kernels are pattern-matched conjectures, not proved: the text states they were found 'by looking at many values of q' (CP-preserving) and 'by computing the kernel for many values of q and j-tilde until a pattern emerged' (CP-violating). Every digit in Tables 1-6 and every coefficient in the fits (4.2)-(4.3) is evaluated from these formulas, so a failure of the pattern at any unprobed q or filled-shell configuration would silently corrupt the central claims. The mitigations in the paper are real and should be stated more forcefully (the large-ℓ,ω asymptotics in Appendix A, (A.15)-(A.20), constrain the q-dependence independently, and the smoothness of the large-q fits over q=4..150 is empirical evidence against failure in that range), but I ask for three additions: (i) disclose the actual ranges of q and j-tilde over which the patterns were explicitly verified; (ii) provide a direct numerical evaluation of the kernels by the algorithmic integration of §3.3 at a sampling of q not used in the pattern search, with emphasis on the filled-shell thresholds (the red entries of Tables 2-3, e.g. q=1 for κ=5/2 and q=1/2,3 for κ=7/2); and (iii) state in the text that the closed forms are numerically validated conjectures for all q rather than proved identities. Items (i)-(ii) are within reach with the notebook the authors already provide.
[§4, Table 4 (m=0)] The same machinery that matches to 1-7% for m=1,2,3 fails at 46-60% for the m=0 duality with the critical O(2) model, and the authors' own observation that dropping the leading term leaves a subleading correction matching O(2) to high accuracy while the leading term ruins the match shows that the regime dependence of the expansion is not understood. The paper offers two speculations (complete filling of the lowest Landau level; subleading term much larger than leading, suggesting diverging large-N expansion) but resolves neither. Because Tables 1-3 are formed from the same Δ(0)+Δ(1) sum, I ask for a quantitative diagnostic: tabulate the ratio Δ(1)/Δ(0) and the sign pattern of Δ(0) across m=0,1,2,3, and analyse why the leading term is so damaging in the m=0 case (e.g., whether the Casimir piece in (2.16) is anomalously small there). A dedicated discussion of this failure would substantially raise confidence that the m=1-3 agreements are not fortuitous. The authors acknowledge this: “We note that the subleading term is also much bigger than the leading term, unlike the other cases we considered, so it could also simply be the case that the largeNexpansion diverges badly in this case.”
[§4, Eqs. (4.2)-(4.3) and Figures 1-2] The q^0 claim for QED3 rests on fits whose quoted q^0 ranges from -0.0937(10) to -0.096(9); the κ=5/2 and κ=7/2 cases are consistent with the universal -0.09372 only at ~10% relative precision, so the headline assertion that QED3 'takes the same nonzero value' is supported but loosely for two of the four κ values. For sQED3 the claimed zeros (0.0001(5), 0.0000(7), 0.0000(8)) are much tighter. Please report, for each fit: the number of data points and q-range, the number of fit parameters, a reduced chi-squared or equivalent goodness-of-fit measure, and the sensitivity of q^0 to trimming the lowest q values and to dropping the q^{-3/2}, q^{-2} terms. Also state how the 'empirical systematic uncertainties' are obtained. These diagnostics matter because the symmetry of the result (same nonzero q^0 for QED3 at all κ vs zero for sQED3) is a central, and physically surprising, claim.

Minor objections (5)

[Tables 1-3 captions] The 'average relative error of just one percent' claim applies to q≥3/2; the two smallest charges in Table 1 have errors 14.7% (q=1/2) and 7.4% (q=1), which are the headline comparisons (emergent current and fuzzy sphere). Please state the q-range over which the one-percent statement holds, or quote the average including the low-q entries.
[§2.1 and Tables 2-3 captions] Typos: 'the partition funciton' (Sec 2.1); 'red denotes filled the filled first nonzero energy shell' (Tables 2-3 captions). Also the caption of Table 1 defines Δ_{f,b} = Δ^{(0)}_{f,b} + Δ^{(1)}_{f,b} with subscripts that are not all introduced in the table body; please align the notation.
[Figure 1] The axis/legend labeling 'Δ(1)_f (q) (q)' and the fit labels f^f_0(q)..f^f_3(q) are confusing, in part because the same symbol f is used for the fermionic fits and for Δ^{(1)}_f. Please relabel the fits (e.g., g_m(q)) and remove the duplicated '(q)'.
[Footnotes 16-17] The sign corrections to N_{q,j} and the iσ_2 term relative to [18] are important for anyone building on the published Green's function; please state the corrected signs explicitly in the main text (or in a short appendix) rather than only in a footnote, and verify that the notebook implements the corrected signs.
[§3, paragraph after Eq. (3.8)] The log β/β term and the log N term related to O(N^{-1}) degeneracy breaking are deferred to [18] with no formula here; a one-sentence reminder of their form (or a pointer to the exact equation in [18]) would let a reader reconstruct the full free-energy expansion without consulting the earlier paper.

How Pith checks papers → · Full audit of this paper →

The paper

The paper claims

The subleading large-$N,k$ expansion of monopole scaling dimensions in QED$_3$, taken to $N=1$, gives $2.09556$ for the $q=1/2$ monopole at $k=-3/2$ — within a few percent of the exact $2$ an emergent SO(3) current would require — matches…

The reading

Building on the large-$N,k$ expansion of monopole scaling dimensions in Chern-Simons-matter theories, this paper computes the subleading correction for QED$_3$ with $N$ fermions and Chern-Simons level $k$ at fixed $k/N$, then extrapolates to $N=1$ to test the 3d bosonization dualities. The $q=1/2$ monopole at $k=-3/2$ lands at $2.09556$, close to the exact value $2$ that an emergent SO(3) symmetry would demand, and the $q=1$ monopole lands at $5.27538$, bracketed by the two fuzzy-sphere predictions $5.1079$ and $5.3435$. Higher charges match the dual scalar-theory values at about one percent average error for the $m=1$ duality, and at five to seven percent for the $m=2,3$ dualities. Separate large-$q$ fits give a $q^0$ term near the universal superfluid value $-0.0937$ for QED$_3$ even though parity is broken, while the same term vanishes for scalar QED$_3$ at nonzero $k$.

What carries the argument

The computation proceeds through the thermal free energy on $S^2\times S^1_\beta$ with $4\pi q$ magnetic flux, and the state-operator map identifies the $\beta$-independent term with the monopole scaling dimension. Because the Chern-Simons term induces a gauge charge proportional to $q$, the naive vacuum must be dressed with matter modes drawn from progressively higher Landau levels as $\kappa=k/N$ increases, and the subleading correction is carried by a set of closed-form matter kernels, equations (3.25) through (3.33), that combine a CP-preserving piece with filling-fraction-dependent CP-violating pieces all assembled from the same $3j$-symbol ingredient. The result is an infinite sum and integral that is made finite by subtracting a computed asymptotic tail and adding back its zeta-regularized value.

Extended reading notes

Core claim

The central claim is that the subleading large-$N,k$ expansion of monopole scaling dimensions, evaluated at $N=1$, dynamically confirms the 3d bosonization dualities: the $q=1/2$ monopole of QED$_3$ at $k=-3/2$ lands at $2.09556$, close to the value $2$ of the extra current needed to enhance the $U(1)$ topological symmetry to $SO(3)$; the $q=1$ monopole lands at $5.27538$, bracketing the two fuzzy-sphere values $5.1079$ and $5.3435$; and monopoles with $q\ge 3/2$ match the dual sQED$_3$ at $k=2$ to about one percent average error. The same comparison works for the $m=2,3$ dualities at roughly five and seven percent average error, and fails for the $m=0$ duality to the critical $O(2)$ model. The large-$q$ limit also yields a $q^0$ term near the universal superfluid value $-0.0937$ for QED$_3$ at every $k$ considered — survival of the parity-preserving coefficient in a parity-breaking theory — and a vanishing $-q^0$ term for sQED$_3$ at $k\neq 0$, as expected for a free-fermion large-charge EFT.

Load-bearing premise

The closed-form matter kernels (3.25) through (3.33) are asserted to hold for every charge $q$ after being verified on finitely many values, and every table value, every duality match, and both large-$q$ fits are computed through these formulas, so a single untested $q$ where the pattern fails would corrupt the central claims.

Editorial extensions

If this is right

The $q=1/2$ value $2.09556$ at extrapolated $N=1$, $k=-3/2$ supports the predicted enhancement of the $U(1)$ topological symmetry to $SO(3)$ with an emergent conserved current of dimension $2$.
The $q=1$ value landing between the fuzzy-sphere pair $5.1079$ and $5.3435$ supplies a dynamical, gauge-theoretic check of that non-perturbative model, independent of whether the calculation picks out spacetime spin zero or two.
The $m=2,3$ matches at five and seven percent average error corroborate the effective-description view that one scalar coupled to two $U(1)$ gauge fields at $k=(m+1)/m$ reproduces the monopole data of QED$_3$ at $k=-(m+1/2)$.
A $q^0$ coefficient near $-0.0937$ in QED$_3$ at nonzero $k$ indicates that the superfluid large-charge EFT survives parity breaking for this class of theories.
The vanishing $q^0$ for sQED$_3$ at $k\neq 0$ is consistent with the seed duality mapping $k=1$ sQED$_3$ to a free fermion, whose large-charge EFT has no $q^0$ term.

Reading between the lines

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

The two $q^0$ results harbour a tension the authors flag: if the $q^0$ coefficient is universal and $N$-independent, then QED$_3$ (superfluid, nonzero) and sQED$_3$ (free-fermion-like, zero) cannot both be the correct IR description at finite $N=1$, nonzero $k$, so either the $N=1$ extrapolation or the universality premise must break somewhere.
The $m=0$ failure, where the subleading term alone matches the $O(2)$ data far better than the leading-plus-subleading sum, suggests the large-$N$ expansion itself is degrading for the completely filled lowest Landau level rather than implying the duality is wrong.
A testable extension would be a fuzzy-sphere or lattice computation of monopole dimensions at $q>1$ for the $k=-3/2$ theory, to see whether QED$_3$'s better agreement at $q=1/2,1$ persists at higher charge.

Original abstract

We compute the scaling dimensions of monopole operators in QED$_3$ with $N$ 2-component complex fermions and Chern-Simons level $k$ to subleading order in the limit where $k,N$ are large and $k/N$ is fixed, which generalizes previous results for scalar QED$_3$ (sQED$_3$) coupled to $N$ complex scalars. In both cases, at leading large $N$ many monopole operators are degenerate, and we compute their energy splitting. We use these results to check the duality between QED$_3$ with $N=1,k=-3/2$ and sQED$_3$ with $N=1,k=2$, whose $U(1)$ monopole symmetry is enhanced to $SO(3)$. We find the scaling dimensions in each dual theory match to high precision for all charges, the lowest charge spin 1 monopoles are close to 2 as expected for the emergent $SO(3)$ current, and the degenerate second lowest charge monopoles match a prediction from a fuzzy sphere calculation, including the energy splitting. We also find similar evidence for dualities between QED$_3$ with $N=1,k=-(2m+1)/2$ and one scalar coupled to two $U(1)$ gauge fields, which has an effective description as sQED$_3$ with $N=1,k=\frac{m+1}{m}$ and $q_s=mq_f$ for $m>1$. Finally, we also find similar evidence for Son's duality between QED$_3$ with $N=1,k=0$ and a free fermion, and clarify a previously discussed match between sQED$_3$ with $N=k=1$ dual to a free fermion.

The compiled record

The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-09-30: 1 request, 5,577 input tokens.

What the abstract itself states, as probabilities: claims first 0.04 · states limitation 0.24 · human participants 0.01 · controlled comparison 0.15 · randomized assignment 0.01 · states quantitative result 0.97 · compares with prior results 0.98 · mentions code or data availability 0.02.

The abstract, sentence by sentence

arXiv listing abstract v1. Results and the main claim are underlined; show every role.

We compute the scaling dimensions of charge $q$ monopole operators in QED$_3$ with $N$ two-component complex fermions and Chern-Simons level $k$, to subleading order in the limit where $k,N$ are large and $k/N$ is fixed. We use this and previous results for scalar QED$_3$ (sQED$_3$) to check the duality between QED$_3$ with $N=1,k=-3/2$ and sQED$_3$ with $N=1,k=2$, whose $U(1)$ monopole symmetry is enhanced to $SO(3)$. We find that the lowest charge monopole value for QED$_3$ is close to 2 as expected for the emergent $SO(3)$ current, the second lowest charge monopole matches a prediction from a fuzzy sphere calculation, while higher $q$ monopoles match the corresponding sQED$_3$ values with relative error of just one percent. We also find similar evidence for dualities between QED$_3$ with $N=1,k=-(2m+1)/2$ and one scalar coupled to two $U(1)$ gauge fields, which has an effective description as sQED$_3$ with $N=1,k=\frac{m+1}{m}$ and $q_s=mq_f$ for $m>1$. Finally, by fitting many values of $q$ we show that the $q^0$ term at large $q$ for QED$_3$ takes the same nonzero value that appears in the universal EFT for parity preserving $U(1)$ theories, even though parity is broken for $k\neq0$. For sQED$_3$, a similar fit gives zero $q^0$ term for $k\neq0$, unlike the universal nonzero value previously observed for $k=0$.

background contribution method result interpretation main claim

What the classifier reports for each sentence. The number is the probability it gave the label.
#	sentence	role	the authors' own work	states the main conclusion	strength
s0	We compute the scaling dimensions of charge $q$ monopole operators in QED$_3$ with $N$ two-component complex fermions and Chern-Simons level $k$, to subleading order in the limit where $k,N$ are large and $k/N$ is fixed.	contribution 1.00	0.97	0.10	asserted 1.00
s1	We use this and previous results for scalar QED$_3$ (sQED$_3$) to check the duality between QED$_3$ with $N=1,k=-3/2$ and sQED$_3$ with $N=1,k=2$, whose $U(1)$ monopole symmetry is enhanced to $SO(3)$.	contribution 0.65	0.79	0.08	asserted 0.90
s2	We find that the lowest charge monopole value for QED$_3$ is close to 2 as expected for the emergent $SO(3)$ current, the second lowest charge monopole matches a prediction from a fuzzy sphere calculation, while higher $q$ monopoles match the corresponding sQED$_3$ values with relative error of just one percent.	result 1.00	0.97	0.43	asserted 1.00
s3	We also find similar evidence for dualities between QED$_3$ with $N=1,k=-(2m+1)/2$ and one scalar coupled to two $U(1)$ gauge fields, which has an effective description as sQED$_3$ with $N=1,k=\frac{m+1}{m}$ and $q_s=mq_f$ for $m>1$.	result 0.94	0.91	0.11	asserted 0.86
s4	Finally, by fitting many values of $q$ we show that the $q^0$ term at large $q$ for QED$_3$ takes the same nonzero value that appears in the universal EFT for parity preserving $U(1)$ theories, even though parity is broken for $k\neq0$.	result 0.99	0.96	0.24	asserted 1.00
s5	For sQED$_3$, a similar fit gives zero $q^0$ term for $k\neq0$, unlike the universal nonzero value previously observed for $k=0$.	result 0.99	0.92	0.07	asserted 1.00

What kind of paper the classifier reports this is: theoretical result 0.96 · evidence from simulation or computation 0.61.

What was examined, and what was not

sources read

arXiv listing abstract v1

not examined

full text
figures and tables
references and citing sentences
relations to other papers

levels run

P abstract

not run

S sections
T body sentences
C citations
R cross-paper relations

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. Every number in the measurements table is the authors', checked by code. Record as JSON

Figures

**Figure 1.** Subleading monopole scaling dimensions ∆

**Figure 2.** Subleading monopole scaling dimensions ∆

Figures reproduced from arXiv:2609.37795 by the authors.

Discussion

Disagree with the review, or know something it missed? Leave a comment, highlight any passage above to comment on it, or challenge this review formally.