Pith. sign in

REVIEW 3 major objections 4 minor 22 references

Classification of symmetric fusion categories over $\mathbb{R}$

T0 review · 3 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash

Pith's one-line read Every symmetric fusion category over the real numbers is equivalent to a semi-linear super representation category of a finite $\mathbb{Z}_2$-graded super group.

desk verdict A solid, genuinely new real analogue of Deligne's classification, with one local error in Section 6.3 and one load-bearing proof omission in Section 3 that both need fixing before it is fully rigorous. read the letter →

arxiv 2608.04940 v1 pith:JIHCCKJL submitted 2026-08-05 math.QA math.CTmath.RT

classification math.QAmath.CTmath.RT MSC 18M2018M0518M1518M25
keywords symmetricfusioncategoriessupergroupsZ2-gradedsemi-linearrepresentationsGaloisdescentrealformsequivariantizationfiberfunctors
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 establishes the real-number analogue of the classical classification of symmetric fusion categories over algebraically closed fields of characteristic zero: every symmetric fusion category over $\mathbb{R}$ is equivalent to $\mathrm{sRep}_{\mathbb{C}/\mathbb{R}}(G,z,s)$ for some finite super group $(G,z)$ equipped with a grading $s:G\to\mathbb{Z}_2$ separating unitary from anti-unitary elements. Such an object is the category of finite-dimensional complex super vector spaces on which even elements of $G$ act $\mathbb{C}$-linearly and odd elements act anti-linearly, with the central element $z$ acting as fermion parity, so the result says that every finite real quantum symmetry, bosonic or fermionic, unitary or time-reversal-like, is concrete. The proof uses Galois descent over $\mathbb{C}/\mathbb{R}$: a real category is recovered from its complexification by equivariantization with respect to a semi-linear $\mathbb{Z}_2$-action, and the paper classifies these actions on the complex categories $\mathrm{sRep}_{\mathbb{C}}(K,z)$ as $\mathbb{Z}_2$-graded group extensions $1\to K\to G\to\mathbb{Z}_2\to 1$. A further Tannaka-Krein type theorem identifies real symmetric multi-fusion categories with finite groupoids carrying a $\mathbb{Z}_2\times B\mathbb{Z}_2$-action, and the paper determines exactly which real categories admit fiber functors to $\mathrm{Vec}_{\mathbb{R}}$ or $\mathrm{sVec}_{\mathbb{R}}$.

What carries the argument

The carrying mechanism is Galois descent over $\mathbb{C}/\mathbb{R}$, summarized as the symmetric monoidal equivalence of 2-categories $2\mathrm{Vec}_{\mathbb{C}}^{\mathbb{Z}_2}\simeq 2\mathrm{Vec}_{\mathbb{R}}$ (Theorem 4.14): real forms of a finite semisimple complex category are precisely semi-linear $\mathbb{Z}_2$-actions, i.e. anti-linear symmetric monoidal autoequivalences $J$ with $J^2\cong 1$. The classification step is the computation, for each finite super group $(K,z)$, of all such actions on $\mathrm{sRep}_{\mathbb{C}}(K,z)$: Section 6.2 shows that a semi-linear $\mathbb{Z}_2$-action is the same datum as a $\mathbb{Z}_2$-graded extension $1\to K\to G\to\mathbb{Z}_2\to 1$ with $z$ central, and equivariantization produces $\mathrm{sRep}_{\mathbb{C}/\mathbb{R}}(G,z,s)$. In the reconstruction half, the carrying object is the groupoid $\widetilde{\mathcal{F}}_{\mathcal{C}}$ of $\mathbb{R}$-linear fiber functors $\mathcal{C}\to\mathrm{sVec}_{\mathbb{C}}$, equipped with the action of the 2-group $\mathrm{Aut}_{\mathbb{R}}(\mathrm{sVec}_{\mathbb{C}})\simeq\mathbb{Z}_2\times B\mathbb{Z}_2$; Theorem 6.21 is the reconstruction $\mathcal{C}\simeq\mathrm{Fun}_{\mathbb{Z}_2\times B\mathbb{Z}_2}(\widetilde{\mathcal{F}}_{\mathcal{C}},\mathrm{sVec}_{\mathbb{C}})$.

What would settle it

Compute, for the super group $(K,z)=(\mathbb{Z}_2,1)$, i.e. for $\mathrm{sVec}_{\mathbb{C}}$, all semi-linear $\mathbb{Z}_2$-actions and their equivariantizations; the paper's classification predicts exactly two actions, corresponding to the extensions $\mathbb{Z}_2\times\mathbb{Z}_2$ and $\mathbb{Z}_4$, and a third action or a third equivariantization would contradict Theorem 6.17.

Watch

Extended reading notes

Core claim

The central claim is Theorem 6.17: every symmetric fusion category over $\mathbb{R}$ is equivalent to $\mathrm{sRep}_{\mathbb{C}/\mathbb{R}}(G,z,s)$ for some finite super group $(G,z)$ with a group homomorphism $s:G\to\mathbb{Z}_2$ satisfying $s(z)=0$. A semi-linear super representation is a finite-dimensional complex super vector space $V$ with a $\mathbb{C}$-linear or anti-linear action of $G$ according as $s(g)=0$ or $1$, such that $z$ acts by the parity operator, and the category is symmetric monoidal and real-linear rather than complex-linear in general. The reduction is by complexification: a real symmetric fusion category becomes a symmetric fusion category over $\mathbb{C}$, which the complex classification (Theorem 3.11) identifies with $\mathrm{sRep}_{\mathbb{C}}(K,z)$; the remaining choice is a semi-linear $\mathbb{Z}_2$-action, and Sections 6.1-6.3 show these actions correspond exactly to $\mathbb{Z}_2$-graded extensions of $(K,z)$, with equivariantization producing $\mathrm{sRep}_{\mathbb{C}/\mathbb{R}}(G,z,s)$. The paper further proves Theorem 6.22, an opposite equivalence between the 2-groupoid of real symmetric multi-fusion categories and finite groupoids equipped with a $\mathbb{Z}_2\times B\mathbb{Z}_2$-action, realized by the groupoid of $\mathbb{R}$-linear fiber functors into $\mathrm{sVec}_{\mathbb{C}}$.

Load-bearing premise

The whole classification rests on the assumption that every way of adding an anti-unitary symmetry to a complex super-representation category can be written, up to isomorphism, as complex conjugation followed by an ordinary unitary symmetry, a step the paper states as a proposition whose proof is omitted.

Editorial extensions

If this is right

  • Every finite bosonic or fermionic symmetry of a quantum system over the real numbers appears as $\mathrm{sRep}_{\mathbb{C}/\mathbb{R}}(G,z,s)$; there are no real symmetric fusion categories outside this family.
  • The explicit data are finite super groups with a $\mathbb{Z}_2$-grading, so listing all real symmetric fusion categories is equivalent to listing $\mathbb{Z}_2$-graded extensions of the finite super groups $(K,z)$ appearing over $\mathbb{C}$.
  • Theorems 6.23 and 6.24 characterize Tannakian versus super-Tannakian real categories: they are $\mathrm{Rep}_{\mathbb{C}/\mathbb{R}}(K\rtimes\mathbb{Z}_2^T)$ and $\mathrm{sRep}_{\mathbb{C}/\mathbb{R}}((K,z)\rtimes\mathbb{Z}_2^T)$ respectively, with $K$ (resp. $(K,z)$) carrying a $\mathbb{Z}_2$-action.
  • Theorem 6.22 says a real symmetric multi-fusion category and its concrete fiber-functor groupoid determine each other completely, giving a reconstruction result for real categories analogous to the complex Tannaka-Krein theorem.
  • Conjecture 6.26 proposes the same shape $\mathrm{sRep}_{E/F}(G,z,s)$ for general characteristic-zero fields, with $E/F$ a finite Galois extension and $G$ graded by $\mathrm{Gal}(E/F)$, so the method is expected to extend beyond $\mathbb{C}/\mathbb{R}$.

Reading between the lines

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

  • An independent computation of the 2-group of semi-linear autoequivalences of $\mathrm{sRep}_{\mathbb{C}}(K,z)$ for small $(K,z)$, such as $\mathrm{sVec}_{\mathbb{C}}$ or $\mathrm{Rep}_{\mathbb{C}}(\mathbb{Z}_2)$, would both test the omitted Proposition 3.15 and give explicit generators for the $\mathbb{Z}_2\times B\mathbb{Z}_2$-action appearing in the reconstruction theorem.
  • The same descent framework, using the semi-linear $\mathbb{Z}_2$-action machinery of Section 4, could compute real forms of non-symmetric fusion categories even though the complex classification of those categories is not known; the paper restricts attention to the symmetric case.
  • In physics terms, the grading $s$ records whether a symmetry is time-reversal-like and the central element $z$ records fermion parity, so Theorem 6.17 gives a normal form for any finite symmetry group of a fermionic system with anti-unitary symmetries; this application is not developed in the paper.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 4 minor

Summary. The paper claims a complete classification of symmetric fusion categories over the real numbers: every such category is equivalent to sRep_{C/R}(G,z,s) for some finite Z2-graded super group (Theorem 6.17). The proof proceeds by complexification, using a Galois-descent equivalence 2Vec_C^{Z2} ≃ 2Vec_R (Theorem 4.14), Deligne's classification over C, and a computation of semi-linear Z2-actions on sRep_C(K,z) in terms of Z2-graded group extensions. A second main result (Theorem 6.22) is a Tannaka-Krein style opposite equivalence between symmetric multi-fusion categories over R and finite groupoids with a Z2 × BZ2-action. The paper also characterizes which real symmetric fusion categories admit fiber functors to Vec_R or sVec_R.

Significance. If the technical gaps identified below are repaired, this is a substantial contribution. It provides the expected real analogue of Deligne's theorem and identifies all real symmetric fusion categories as representation categories of finite super groups with an anti-unitary grading, unifying examples such as Vec_R, Vec_H, and sVec_R. The Galois-descent framework is appropriate, and the argument is not circular: it imports Deligne's algebraically closed classification and reduces the real case to descent data. The explicit computation of fiber functors and the reconstruction theorem are additional strong results. The paper is also careful with separability and Morita-theoretic details, which is a genuine strength.

major comments (3)
  1. [6.3, first paragraph] The displayed equivalence sRep_C(G,z) ≃ sRep_{C/R}((G,z)×Z2^T) is false. In sRep_C(G,z) the unit endomorphism algebra is C, whereas in sRep_{C/R}((G,z)×Z2^T) the anti-linear Z2^T factor acts on the unit by complex conjugation, forcing End(1) ≅ R. The Ω_C ≅ C case must instead be represented by a triple with trivial grading s, so that sRep_{C/R}(G,z,s) = sRep_C(G,z). This is repairable, but as written the proof of Theorem 6.17 contains an incorrect identification in a load-bearing case.
  2. [3.3, Proposition 3.15] Proposition 3.15 is asserted with the proof omitted ('we omit the details'), yet it is load-bearing. It is used, through Lemma 6.3 and Proposition 6.13, to reduce arbitrary semi-linear Z2-actions on sRep_C(K,z) to Z2-graded super group extensions, which is exactly the step that yields Theorem 6.17. The manuscript should either provide the full parallel proof or cite a precise reference where the super-case equivalence is proved.
  3. [6.1, Lemma 6.3] The claim that the proof of Lemma 6.3 is 'completely identical' to Lemma 6.1 is not sufficiently justified. Unlike the bosonic case, the proof requires the omitted Proposition 3.15, and it also needs to check that the isomorphism F∘conj ≃ conj∘F can be chosen coherently for all C-linear autoequivalences F of sRep_C(K,z). Because this lemma is the bridge between semi-linear Z2-actions and group extensions, a complete proof should be spelled out.
minor comments (4)
  1. [6.4, after Proposition 6.20] For s trivial, the statement that 'tilde F_C is equivalent to Z2 × BG' should be clarified as a disjoint union of two copies of BG, rather than a product groupoid with two objects, to avoid confusion with the later homotopy quotient.
  2. [Examples 4.12 and 5.11] The category Vec_H is used without definition; please define it explicitly as the category of finite-dimensional modules over the quaternion algebra H.
  3. [Example 5.14] The isomorphism H ⊗_C H ≃ M_2(C) is terse; spelling out the chosen complex structure on H would make the fusion rule H ⊗_C H ≃ C^{⊕4} transparent.
  4. [Section 6.5, Conjecture 6.26] The assertion that a field with finite-degree algebraic closure is either algebraically closed or real closed is the Artin-Schreier theorem; adding a reference would be helpful.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the main derivation imports Deligne's external theorem and proves the descent/action classification internally; the sole self-citation is non-load-bearing.

full rationale

The claimed derivation is not circular. Theorem 6.17 is obtained by (i) the Galois descent equivalence of Theorem 4.14, which is proved in Section 4.3 via Lemmas 4.16–4.17 and Proposition 4.18; (ii) Deligne’s external classification over C (Theorems 3.9–3.11); and (iii) an internal computation of semi-linear Z2-actions on sRep_C(K,z) in Sections 6.1–6.3. The reduction of arbitrary semi-linear actions to Z2-graded group extensions is a theorem, not a notational convention, and it rests on Proposition 3.15, whose proof is sketched as parallel to Proposition 3.14. An omitted or compressed proof is a correctness risk, not circularity. No parameter is fitted, and no class is reverse-engineered from the target result: the categories sRep_{C/R}(G,z,s) are defined first and then shown to exhaust all symmetric fusion categories over R. The only self-citation, [HXZ24] in Remark 6.16, supplies an alternative homotopy-quotient description of Vec_{C/R} and is not load-bearing for the main argument. I also note, as a correctness concern rather than a circularity, that the Ω_C≃C case in Section 6.3 appears to identify sRep_C(G,z) with sRep_{C/R}((G,z)×Z2^T), which would change the unit endomorphism algebra from C to R; this is a special-case identification error, not a circular dependence of the descent framework. Overall, the derivation is self-contained relative to its stated external inputs, so the circularity score is 0.

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

The central claim rests on standard background theorems: Deligne's theorem over C, connectivity of fiber functor groupoids, Morita theory for fusion categories, and the Grothendieck construction relating monoidal functors to group extensions. No free parameters or invented entities are introduced. The Galois descent theorem is proven in the paper and therefore not listed as an axiom.

assumptions (4)
  • standard math Deligne's classification of symmetric fusion categories over algebraically closed fields of characteristic zero: every such category is equivalent to sRep_k(G,z) (Theorem 3.11).
    Imported from [Del90, Del02]; used in Section 6.3 to reduce the real classification to semi-linear Z2-actions on sRep_C(K,z).
  • standard math The groupoid of fiber functors C -> sVec_k is connected (Theorem 3.13).
    Imported from [EGNO15, Theorem 9.9.26]; used to prove Propositions 3.14 and 3.15 and to compute autoequivalence 2-groups.
  • standard math Monoidal functors Gamma -> Aut(BK) correspond to group extensions 1 -> K -> G -> Gamma -> 1 (Section 6.2).
    Well-known Grothendieck construction; assumed without proof and used to translate Z2-graded extensions into actions.
  • standard math Standard Morita theory for fusion categories (Theorems 2.16, 2.18, 2.20).
    Imported from [DSPS20], [KZ17], and [ENO10]; used in the proof of Galois descent Theorem 4.14.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Classification of symmetric fusion categories over $\mathbb{R}$." pith.science (2026). https://pith.science/paper/JIHCCKJL

@misc{pith2026260804940,
  author       = {Pith},
  title        = {Pith review of: Classification of symmetric fusion categories over $\mathbbR$},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JIHCCKJL}},
  note         = {Machine review of arXiv:2608.04940}
}
abstract

We show that every symmetric fusion category over $\mathbb{R}$ is equivalent to the category of finite-dimensional semi-linear representations of a $\mathbb{Z}_2$-graded finite super group. The proof uses Galois descent for tensor categories over $\mathbb{C}/\mathbb{R}$, reducing the classification to semi-linear $\mathbb{Z}_2$-actions on symmetric fusion categories over $\mathbb{C}$. As a further structural result, we establish a Tannaka-Krein type correspondence between symmetric fusion categories over $\mathbb{R}$ and finite groupoids with a $\mathbb{Z}_2 \times \mathrm{B} \mathbb{Z}_2$-action. This gives a complete real analogue of Deligne's classification result.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

22 extracted references · 7 canonical work pages

  1. [1]

    Tannakian categories in positive characteristic

    Kevin Coulembier. Tannakian categories in positive characteristic . Duke Mathematical Journal, 169(16), 2020. arXiv:1812.02452 https://arxiv.org/abs/1812.02452

  2. [2]

    Compact semisimple 2-categories

    Thibault D \' e coppet. Compact semisimple 2-categories . Transactions of the American Mathematical Society, 2023. arXiv:2111.09080 https://arxiv.org/abs/2111.09080

  3. [3]

    D \' e coppet

    Thibault D. D \' e coppet. Rigid and separable algebras in fusion 2-categories . Advances in Mathematics, 419:108967, 2023. arXiv:2205.06453 https://arxiv.org/abs/2205.06453

  4. [4]

    P. Deligne. Cat \' e gories tannakiennes . In The Grothendieck Festschrift, Volume II , volume 87 of Progress in Mathematics , pages 111--195. Birkhäuser Boston, 1990

  5. [5]

    P. Deligne. Cat\' e gories tensorielles . Moscow Mathematical Journal, 2(2):227--248, 2002

  6. [6]

    D \' e coppet, Peter Huston, Theo Johnson-Freyd, Dmitri Nikshych, David Penneys, Julia Plavnik, David Reutter, and Matthew Yu

    Thibault D. D \' e coppet, Peter Huston, Theo Johnson-Freyd, Dmitri Nikshych, David Penneys, Julia Plavnik, David Reutter, and Matthew Yu. The classification of fusion 2-categories . arXiv preprint, 2024. arXiv:2411.05907 https://arxiv.org/abs/2411.05907

  7. [7]

    Deligne and J

    P. Deligne and J. S. Milne. Tannakian categories . In Lecture Notes in Mathematics , pages 101--228. Springer Berlin Heidelberg, 1982

  8. [8]

    The Witt group of non-degenerate braided fusion categories

    Alexei Davydov, Michael Müger, Dmitri Nikshych, and Victor Ostrik. The Witt group of non-degenerate braided fusion categories . Journal für die reine und angewandte Mathematik (Crelles Journal), 2013(677), 2013. arXiv:1009.2117 https://arxiv.org/abs/1009.2117

Show all 22 references
  1. [9]

    Dualizable tensor categories , volume 268 of Memoirs of the American Mathematical Society

    Christopher Douglas, Christopher Schommer-Pries, and Noah Snyder. Dualizable tensor categories , volume 268 of Memoirs of the American Mathematical Society . American Mathematical Society, 2020. arXiv:1312.7188 https://arxiv.org/abs/1312.7188

  2. [10]

    Descent and forms of tensor categories

    Pavel Etingof and Shlomo Gelaki. Descent and forms of tensor categories . International Mathematics Research Notices, 2012(13):3040--3063, 2011. arXiv:1102.0657 https://arxiv.org/abs/1102.0657

  3. [11]

    Tensor categories

    Pavel Etingof, Shlomo Gelaki, Dmitri Nikshych, and Victor Ostrik. Tensor categories . American Mathematical Society, Providence, RI, 2015

  4. [12]

    Fusion categories and homotopy theory

    Pavel Etingof, Dmitri Nikshych, and Victor Ostrik. Fusion categories and homotopy theory . Quantum Topology, pages 209--273, 2010. arXiv:0909.3140 https://arxiv.org/abs/0909.3140

  5. [13]

    Etingof and V

    P. Etingof and V. Ostrik. Finite tensor categories . Moscow Mathematical Journal, 4(3):627--654, 2004. arXiv:math/0301027 https://arxiv.org/abs/math/0301027

  6. [14]

    The 2-character theory of finite 2-groups

    Mo Huang, Hao Xu, and Zhi-Hao Zhang. The 2-character theory of finite 2-groups . arXiv preprint, 2024. arXiv:2404.01162 https://arxiv.org/abs/2404.01162

  7. [15]

    Joyal and R

    A. Joyal and R. Street. Braided tensor categories . Advances in Mathematics, 102(1):20--78, 1993

  8. [16]

    Semisimple and separable algebras in multi-fusion categories

    Liang Kong and Hao Zheng. Semisimple and separable algebras in multi-fusion categories . arXiv preprint, 2017. arXiv:1706.06904 https://arxiv.org/abs/1706.06904

  9. [17]

    Categories of quantum liquids II

    Liang Kong and Hao Zheng. Categories of quantum liquids II . Communications in Mathematical Physics, 405(9), 2024. arXiv:2107.03858 https://arxiv.org/abs/2107.03858

  10. [18]

    Module categories, weak Hopf algebras and modular invariants

    Victor Ostrik. Module categories, weak Hopf algebras and modular invariants . Transformation Groups, 8(2):177--206, 2003. arXiv:math/0111139 https://arxiv.org/abs/math/0111139

  11. [19]

    On symmetric fusion categories in positive characteristic

    Victor Ostrik. On symmetric fusion categories in positive characteristic . Selecta Mathematica, 26(3), 2020. arXiv:1503.01492 https://arxiv.org/abs/1503.01492

  12. [20]

    Fusion categories over non-algebraically closed fields

    Sean Sanford. Fusion categories over non-algebraically closed fields . Journal of Algebra, 663:316--351, 2025. arXiv:2401.02354 https://arxiv.org/abs/2401.02354

  13. [21]

    Finite quantum groupoids and inclusions of finite type

    Kornél Szlachányi. Finite quantum groupoids and inclusions of finite type . In Mathematical Physics in Mathematics and Physics: Quantum and Operator Algebraic Aspects , pages 393--407. American Mathematical Society, 2001. arXiv:math/0011036 https://arxiv.org/abs/math/0011036

  14. [22]

    Adjointable monoidal functors and quantum groupoids

    Kornél Szlachányi. Adjointable monoidal functors and quantum groupoids . In Hopf Algebras in Noncommutative Geometry and Physics , pages 291--308. CRC Press, 2005. arXiv:math/0301253 https://arxiv.org/abs/math/0301253

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.