REVIEW 2 major objections 6 minor 29 references
Tensor Product CFTs and One-Character Extensions
T0 review · 2 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash
Pith's one-line read One-character CFT extensions reduce to a small S-invariant basis.
desk verdict First-wave but genuinely useful: S-invariant polynomial bases give closed-form candidate characters for one-character extensions at c>24, with the basis-generation claim unproved and the enumeration possibly incomplete. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The machinery is an S-invariant basis: a small set of S-invariant homogeneous polynomials $P_i$ in the seed characters. Because S-invariance is imposed as a polynomial identity, once the characters are expressed in this basis the condition is automatic, and the problem reduces to solving the linear Diophantine equation that matches the degree of a monomial in the $P_i$ with the tensor power $N$. Constant S-invariant polynomials (like $P_2=2$ for $A_{1,1}$) generate the quotient ideal, which is why the final answers are far more compact than the original degree-$N$ polynomials.
What would settle it
For $A_{1,1}$ at $c=24$, solve the S-invariance condition directly on the $q$-series of the ansatz; if any S-invariant character with non-negative integer $q$-series coefficients does not match $P_1^3+(-42+p)P_2^4$ for integer $p$, the polynomial basis method is incomplete.
Extended reading notes
Core claim
For each seed CFT studied (the [2,0] WZW theories $A_{1,1}$, $A_{2,1}$, $G_{2,1}$, $D_{4,1}$, $F_{4,1}$, $E_{6,1}$, $E_{7,1}$, the Ising model $M(4,3)$, and the infinite classes $D_{r,1}$ and $B_{r,1}$), the paper discovers a basis of S-invariant homogeneous polynomials $P_i$ in the seed characters. Every one-character extension character of $C^{\otimes N}$ is then a polynomial in these $P_i$, and the allowed monomials are exactly the non-negative integer solutions of a degree equation such as $8a+6b=N$ for $A_{1,1}$. The constant S-invariant polynomials generate the ideal of relations, so the extension characters belong to the quotient ring of the character ring. As examples, the character of the monster CFT is $P_1^3 - 744$ where $P_1$ is the degree-16 S-invariant polynomial of Ising characters, and the paper reproduces all 28 entries of Schellekens' list that are tensor-product extensions. It also rules out extension characters at several low central charges for $G_{2,1}$ and $F_{4,1}$, and conjectures six infinite series of new one-character extension CFTs.
Load-bearing premise
The enumeration assumes that imposing S-invariance as a polynomial identity in the character ring is equivalent to imposing it on the $q$-series, which is only true when the character ring has no polynomial relations; if relations exist, the polynomial condition is stronger and may miss valid extension characters.
Editorial extensions
If this is right
- For each studied seed, one-character extension characters can be written in closed form as polynomials in the $P_i$ at arbitrarily high central charge.
- The monster CFT character is explicitly $P_1^3-744$ with $P_1$ a degree-16 S-invariant polynomial of Ising characters.
- The method reproduces the known $c=24$ one-character CFTs that arise as tensor-product extensions, validating the approach at the classified central charge.
- For $G_{2,1}$ and $F_{4,1}$, several low central charges admit no admissible extension character, so those extensions are ruled out.
- The conjectured six and four infinite series of one-character extensions provide testable targets for future CFT classification at $c>24$.
Reading between the lines
- The polynomial-level S-invariance condition is stronger than equality of $q$-series when the seed characters obey polynomial relations, so the enumeration of admissible extension characters may be incomplete.
- The same basis construction should work for other seed CFTs with rational characters, giving a general algebraic recipe for one-character extensions.
- The constant S-invariant polynomials are analogues of classical polynomial relations among modular characters, suggesting hidden character-ring relations for other rational CFTs.
- If the conjectured infinite series correspond to genuine CFTs, they would supply infinitely many one-character theories at $c>24$; checking the full CFT data beyond characters is the next step.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper develops an algebraic method for obtaining the characters of one-character (holomorphic) CFTs as one-character extensions of tensor products C^⊗N of a fixed seed theory C. For seeds that are [2,0] WZW theories (A1,1, A2,1, G2,1, D4,1 and their coset partners E7,1, E6,1, F4,1), the Ising model M(4,3), A4,1, and the infinite classes D_{r,1} and B_{r,1}, the authors build an ansatz as a homogeneous polynomial in the characters with integer conformal dimension, impose the T-transformation and S-invariance as a polynomial identity (eq. (18)), solve the resulting overdetermined linear system, and reorganize all answers in terms of a small set of S-invariant polynomials {P1, P2, P3}. This yields a Diophantine shortcut (e.g., 8a + 6b = N for A1,1, eq. (40)) that reproduces the direct computation, giving compact closed forms up to c = 128 for A1,1 and A2,1 and uniform r-independent formulas for all D_{r,1} and B_{r,1}. Flagship results include the monster character written as P1^3 − 744 in Ising characters (eq. (106)), agreement with the Schellekens list at c = 24, non-existence rulings at selected large central charges for G2,1 and F4,1, and conjectured six-plus-four infinite families of one-character extensions. The paper is explicit that it works at the level of admissible characters (the 'first wave') and does not claim a proof of existence of the corresponding CFTs for c > 24.
Significance. If the completeness of the enumeration is granted, this is a substantial and useful technical contribution. The linear-algebra core is transparent and reproducible from the published q-series data, and the spot checks are clean (e.g., eq. (22) evaluates to j^{1/3} with leading coefficient 1 + 248q, and the c = 24 outputs match the relevant Schellekens entries). The paper is honest about its main caveat, and the Schellekens list is used only as external calibration, so the c = 24 agreement is a genuine check rather than a circular fit. The derivation that the ansatz monomial set is r-independent for the D_{r,1} and B_{r,1} families (Section 4.3) is a genuine proof, and the quotient-ring observation about constant S-invariant polynomials (Section 5) is conceptually nice, connecting to known identities such as the Rogers-Ramanujan relations. The non-existence rulings and the infinite-family conjectures are falsifiable predictions for future classification work. The main caveat to the significance is that the completeness claims ('every k', 'all r', and the non-existence rulings) rest on an unproved generation statement for the S-invariant bases, as detailed in the major comments.
major comments (2)
- [§3.1–3.4, §4.2–4.4; eqs. (40), (58), (107), (121), (133), (145), (157), (165)] The central organizational claim is that every one-character extension character of C^⊗N is a polynomial in a fixed small set of S-invariant polynomials {P1, P2, P3}, so that the full enumeration reduces to a Diophantine degree equation such as 8a + 6b = N for A1,1 (eq. (40)). The evidence offered is the match between this shortcut and the direct linear-system computation for k ≤ 16 (tables 4–7) and s ≤ 6 (tables 10–14); the text states 'we find a S-invariant basis' (Introduction; Section 3.1) but gives no invariant-theoretic proof that the discovered polynomials generate the graded subring of S-invariant polynomials whose monomials have integer conformal dimensions. The generation claim is nontrivial: for A1,1 the full invariant ring of the S-action has Hilbert series 1/((1−t)(1−t^2)) (generator degrees 1 and 2 over the splitting field), while the commensurate subring relevant to extension characters contains ℚ[P1, P2^4] with Hilbert series 1/((1−t^8)(1−t^24)); P2 itself (eq. (31)) is not commensurate, and only powers P2^{4j} appear in table 4. The paper provides no computation showing that the commensurate invariant subring equals ℚ[P1, P2^4]. Because the universal statements ('every k', 'all r'), the conjectured six-plus-four infinite series, and especially the non-existence rulings for G2,1 and F4,1 at c = 224, 280, 392, 448 and 416, 520, 728, 832 (Section 3.3) test only the handful of candidates from eq. (58), a missing generator at some higher N would invalidate the closure of the tables and those conclusions. The directly computed cases (e.g., N = 20, 40 for G2,1 and F4,1) are solid; the risk is concentrated in the shortcut-based cases. The first-wave caveat of the Introduction properly protects the step from admissible characters to genuine CFTs, but it does not protect the completeness of the character-level enumeration. The fix is concrete: compute the Hilbert series of the commensurate S-invariant subring (a finite-group invariant ring, so a Molien-series computation) and verify degree by degree that it matches the monomial count from the Diophantine equations; alternatively, restrict all completeness and non-existence claims to the range of N that was computed directly.
- [§2, eq. (18); §5; eqs. (34), (103), (166)] The paper imposes S-invariance as an equality of homogeneous polynomials (eq. (18)) and solves the resulting linear system, but the physical condition on the extension character is equality of the resulting q-series. When the seed characters satisfy polynomial relations that hold only at the level of q-series, as the paper itself shows with P2 = 2 for A1,1 (eq. (34)), P3 = 0 and P2 = 1 for the Ising model (Section 4.2), and the constant values in eqs. (119), (131), (143), (155) and (166), the polynomial-level condition is strictly stronger than the q-series-level condition, so the solution space of eq. (18) can be a proper subset of the space of q-series-S-invariant polynomials in the ansatz class. Concretely, a polynomial P in the ansatz space with P(Sχ) − P(χ) lying in the nonzero relation ideal generated by such constant relations would pass the q-series test while failing eq. (18), and the paper does not show that no such P exists in the claimed ranges. The parameter counts in the computed range match the admissible characters of appendix A (e.g., [k/3] free parameters for A1,1), which is reassuring evidence, but the enumeration is extended to all k and all r without proving that the ansatz space intersects the relation ideal trivially in the relevant degrees. Relatedly, Section 5 claims that extension characters are 'properly' elements of the quotient ring generated by the ideal of constant S-invariant polynomials; that description also requires the exhibited relations to generate the full relation ideal of the character ring, which is not proved. The authors should either prove these equivalences (by computing the relation ideal and its intersection with the ansatz space) or state the completeness claims at the q-series level as an additional assumption.
minor comments (6)
- [§2 (after eq. (18))] 'In anycase' should read 'in any case'.
- [Tables 4–14] The printed entries are heavily abbreviated, with many q-series coefficients shown only as '...'; the authors should state whether the full polynomials and q-series are available in ancillary files, so that an independent reader can reproduce the admissibility inequalities.
- [§3.3, footnote 11] The external corroboration of the G2,1/F4,1 non-existence claims, received from B. Rayhaun and referencing [28,29], plays a substantive role in the conclusions and should be moved into the main text rather than left in a footnote.
- [§5, eq. (167)] The Rogers-Ramanujan discussion would benefit from defining G(q) and H(q) and citing the classical identities, since the connection is otherwise undecidable for a reader who does not already know the Lee-Yang character identities.
- [§3.1–3.3] The notation P^{C}_{c(1)=8k} overloads the symbol C (the seed theory) with the superscript label, which makes passages such as 'for every two-character CFT that we perform in section 2, we find two S-invariant polynomial basis' (Introduction) unnecessarily hard to parse; a consistent notation for the seed and for the extension character would help.
- [§3.4, closing paragraph] The argument that the number of S-invariant polynomials equals the number of seed characters 'because any two sets of bases should have the same cardinality' is explicitly speculative (the text says 'perhaps'), and since this is exactly the unproved generation step flagged in the major comments, the sentence should be removed or replaced by a reference to a proof.
Circularity Check
No significant circularity: extension characters are derived from S-invariance constraints on tensor-product ansatze, with external calibration only at c=24.
full rationale
The paper's central derivation chain is self-contained. For each seed CFT, the authors form an ansatz character as a homogeneous polynomial of the tensor-product characters, restrict to identity-commensurate monomials, and solve the overdetermined linear system imposed by S-invariance (eqs. 10, 17, 18). The S-invariant polynomials P1, P2, P3 are discovered from these solutions and used only to rewrite the solutions compactly. The Diophantine shortcut (e.g., eq. 40: 8a+6b=N for A1,1) is a degree-counting enumeration of monomials in the already-found S-invariant polynomials; the same answers are explicitly checked against the full linear-system procedure for the reported N values, so the shortcut is not a fitted parameter renamed as a prediction. The Schellekens list is used only to select which of the infinitely many admissible characters at c=24 correspond to genuine CFTs, which is external calibration. The monster character expression in eq. (106) fixes the free parameter p3 by the known value j-744, so it is a representation of a known character rather than a derived prediction. Claims for c>24 and for the infinite classes are explicitly labeled as surmises/conjectures. The unproved assertion that the discovered P_i generate the full S-invariant subalgebra is a completeness gap that could affect exhaustiveness, but it is not circular: no target character is defined in terms of P_i, and no equation in the paper reduces to its own input by construction.
Assumptions & free parameters
free parameters (2)
- Free parameters p_i / p~_i in extension characters =
Unfixed for c > 24; [k/3] parameters per central charge 8k; pinned only at c = 24 by matching the Schellekens list
- Admissible-character parameters N1, N2, ... (appendix A) =
Non-negative integers satisfying nested inequalities
assumptions (4)
- domain assumption The seed CFTs (A1,1, A2,1, G2,1, D4,1, F4,1, E6,1, E7,1, A4,1, Ising, D_{r,1}, B_{r,1}) have the stated characters, S-matrices, and q-series.
- standard math MMS framework: n characters are linearly independent solutions of an [n,l] MLDE; a one-character CFT has central charge a multiple of 8 and an S-invariant character.
- ad hoc to paper S-invariance of the ansatz can be imposed as a polynomial identity in the characters (eq. 18), ignoring polynomial relations among them.
- ad hoc to paper Character-level admissibility, together with c = 24 Schellekens matches, is a sufficient basis for the c > 24 existence conjectures.
Cite this review
Pith. "Pith review of Tensor Product CFTs and One-Character Extensions." pith.science (2026). https://pith.science/paper/M6CWC33I
@misc{pith2026241210112,
author = {Pith},
title = {Pith review of: Tensor Product CFTs and One-Character Extensions},
year = {2026},
howpublished = {\url{https://pith.science/paper/M6CWC33I}},
note = {Machine review of arXiv:2412.10112}
}
abstract
We study one-character CFTs obtained as one-character extensions of the tensor products of a single CFT $\mathcal{C}$. The motivation comes from the fact that $28$ of the $71$ CFTs in the Schelleken's list of $c = 24$ CFTs are such CFTs. We study for $\mathcal{C}$ : (i) any two-character WZW CFT with vanishing Wronskian index, (ii) the Ising CFT, (iii) the infinite class of $D_{r,1}$ CFTs and the $A_{4,1}$ CFT. The characters being $S$-invariant homogenous polynomials of the characters of $\mathcal{C}$, when organised in terms of a $S$-invariant basis, take compact forms allowing for closed form answers for high central charges. We find a $S$-invariant basis for each of the CFTs studied. As an example, one can find an explicit expression for the character of the monster CFT as a degree-$48$ polynomial of the characters of the Ising CFT. In some CFTs, some of the $S$-invariant polynomials of characters compute, after using the $q$-series of the characters, to a constant value. Hence, the characters of one-character extensions are more properly elements of the quotient ring of polynomials (of characters) with the ideal needed for the quotient, generated by $S$-invariant polynomials that compute to a constant. In some cases, we are able to rule out the existence of one-character extension CFTs. In other cases, we predict their existence. We are able to conjecture a discrete set of six and four infinite series of one-character extension CFTs.
Reference graph
Works this paper leans on
-
[1]
Infinite Conformal Symmetry in Two-Dimensional Quantum Field Theory,
A. A. Belavin, A. M. Polyakov and A. B. Zamolodchikov, “Infinite Conformal Symmetry in Two-Dimensional Quantum Field Theory,” Nucl. Phys. B 241 (1984) 333
work page 1984
-
[2]
Nonabelian Bosonization in Two-Dimensions,
E. Witten, “Nonabelian Bosonization in Two-Dimensions,” Commun. Math. Phys. 92 (1984), 455-472
work page 1984
-
[3]
P. Di Francesco, P. Mathieu and D. Senechal, “Conformal Field Theory,” Springer-Verlag, New York, 1997
work page 1997
- [4]
-
[5]
Twenty-five years of two-dimensional rational conformal field theory
J. Fuchs, I. Runkel and C. Schweigert, “Twenty-five years of two-dimensional rational con- formal field theory,” J. Math. Phys. 51 (2010), 015210 [arXiv:0910.3145 [hep-th]]
work page Pith review arXiv 2010
-
[6]
An Introduction to conformal field theory,
M. R. Gaberdiel, “An Introduction to conformal field theory,” Rept. Prog. Phys. 63 (2000), 607-667 [arXiv:hep-th/9910156 [hep-th]]
arXiv 2000
-
[7]
Rationality in Conformal Field Theory,
G. Anderson and G. W. Moore, “Rationality in Conformal Field Theory,” Commun. Math. Phys. 117 (1988), 441
work page 1988
-
[8]
Differential Equations for Conformal Characters in Moduli Space,
T. Eguchi and H. Ooguri, “Differential Equations for Conformal Characters in Moduli Space,” Phys. Lett. B 203 (1988), 44 85
work page 1988
Show all 29 references
-
[9]
On the Classification of Rational Conformal Field Theories,
S. D. Mathur, S. Mukhi and A. Sen, “On the Classification of Rational Conformal Field Theories,” Phys. Lett. B 213 (1988), 303-308
1988
-
[10]
Reconstruction of Conformal Field Theories From Modular Geometry on the Torus,
S. D. Mathur, S. Mukhi and A. Sen, “Reconstruction of Conformal Field Theories From Modular Geometry on the Torus,” Nucl. Phys. B 318 (1989), 483-540
1989
-
[11]
Differential Equations For Rational Conformal Characters,
S. G. Naculich, “Differential Equations For Rational Conformal Characters,” Nucl. Phys. B 323 (1989), 423-440
1989
-
[12]
On 2d Conformal Field Theories with Two Characters,
H. R. Hampapura and S. Mukhi, “On 2d Conformal Field Theories with Two Characters,” JHEP 01 (2016), 005 [arXiv:1510.04478 [hep-th]]
2016 arXiv
-
[13]
Cosets of Meromorphic CFTs and Mod- ular Differential Equations,
M. R. Gaberdiel, H. R. Hampapura and S. Mukhi, “Cosets of Meromorphic CFTs and Mod- ular Differential Equations,” JHEP 04 (2016), 156 [arXiv:1602.01022 [hep-th]]
2016 arXiv
-
[14]
Two-dimensional RCFT’s without Kac-Moody symmetry,
H. R. Hampapura and S. Mukhi, “Two-dimensional RCFT’s without Kac-Moody symmetry,” JHEP 07 (2016), 138 [arXiv:1605.03314 [hep-th]]
2016 arXiv
-
[15]
Universal RCFT Correlators from the Holomorphic Boot- strap,
S. Mukhi and G. Muralidhara, “Universal RCFT Correlators from the Holomorphic Boot- strap,” JHEP 02 (2018), 028 [arXiv:1708.06772 [hep-th]]
2018 arXiv
-
[16]
Towards a Classification of Two-Character Rational Conformal Field Theories,
A. R. Chandra and S. Mukhi, “Towards a Classification of Two-Character Rational Conformal Field Theories,” JHEP 04 (2019), 153 [arXiv:1810.09472 [hep-th]]
2019 arXiv
-
[17]
Curiosities above c = 24,
A. R. Chandra and S. Mukhi, “Curiosities above c = 24,” SciPost Phys. 6 (2019) no.5, 053 [arXiv:1812.05109 [hep-th]]
2019 arXiv
-
[18]
Classification of RCFT from Holomorphic Modular Bootstrap: A Status Report,
S. Mukhi, “Classification of RCFT from Holomorphic Modular Bootstrap: A Status Report,” [arXiv:1910.02973 [hep-th]]
1910 arXiv
-
[19]
Meromorphic C = 24 conformal field theories,
A. N. Schellekens, “Meromorphic C = 24 conformal field theories,” Commun. Math. Phys. 153 (1993), 159-186 [arXiv:hep-th/9205072 [hep-th]]
1993 arXiv
-
[20]
Three-Dimensional Gravity Revisited,
E. Witten, “Three-Dimensional Gravity Revisited,” [arXiv:0706.3359 [hep-th]]
-
[21]
A mass formula for unimodular lattices with no roots,
O. D. King, “A mass formula for unimodular lattices with no roots,” Mathematics of Computation 72 (2003), 839 [math/0012231]
2003 arXiv
-
[22]
Holomorphic modular bootstrap revisited,
J. Kaidi, Y. H. Lin and J. Parra-Martinez, “Holomorphic modular bootstrap revisited,” JHEP 12 (2021), 151 [arXiv:2107.13557 [hep-th]]
2021 arXiv
-
[23]
Classifying three-character RCFTs with Wronskian index equalling 0 or 2,
A. Das, C. N. Gowdigere and J. Santara, “Classifying three-character RCFTs with Wronskian index equalling 0 or 2,” JHEP 11 (2021), 195 [arXiv:2108.01060 [hep-th]]. 86
2021 arXiv
-
[24]
Bootstrapping fermionic rational CFTs with three characters,
J. B. Bae, Z. Duan, K. Lee, S. Lee and M. Sarkis, “Bootstrapping fermionic rational CFTs with three characters,” JHEP 01 (2022), 089 [arXiv:2108.01647 [hep-th]]
2022 arXiv
-
[25]
New meromorphic CFTs from cosets,
A. Das, C. N. Gowdigere and S. Mukhi, “New meromorphic CFTs from cosets,” JHEP 07 (2022), 152 [arXiv:2207.04061 [hep-th]]
2022 arXiv
-
[26]
Meromorphic cosets and the classification of three- character CFT,
A. Das, C. N. Gowdigere and S. Mukhi, “Meromorphic cosets and the classification of three- character CFT,” JHEP 03 (2023), 023 [arXiv:2212.03136 [hep-th]]
2023 arXiv
-
[27]
Wronskian Indices and Rational Conformal Field Theories,
A. Das, C. N. Gowdigere and J. Santara, “Wronskian Indices and Rational Conformal Field Theories,” JHEP 04 (2021), 294 [arXiv:2012.14939 [hep-th]]
2021 arXiv
-
[28]
Commutative algebras in Fibonacci categories,
T. Booker and A. Davydov, “Commutative algebras in Fibonacci categories,” Journal of Algebra 355 (2012) 176
2012
-
[29]
Equivalence Relations on Vertex Operator Algebras, II: Witt Equivalence and Orbifolds,
S. M¨ oller and B. C. Rayhaun, “Equivalence Relations on Vertex Operator Algebras, II: Witt Equivalence and Orbifolds,” [arXiv:2410.18166 [hep-th]]. 87
Reviewed August 11, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.