Transfer of abelian model structures to equivariant categories and homotopy squares

Guoliang Tang; Li Liang; Liping Li; Rongmin Zhu; Zhenxing Di

Transfer of abelian model structures to equivariant categories and homotopy squares

T0 review · 1 major / 5 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 Finite group actions with invertible order transfer abelian model structures to equivariant categories, and the homotopy categories agree up to retracts.

desk verdict A substantial transfer theorem for Hovey triples to equivariant categories, with a genuinely new comparison result; the main ideas are sound, but one step in Prop. 2.11 is wrong as written and needs a small repair before the proof is complete. read the letter →

arxiv 2608.08141 v2 pith:5RIRQS6K submitted 2026-08-08 math.RT

Zhenxing Di , Liping Li , Li Liang , Guoliang Tang , Rongmin Zhu This is my paper · ORCID

classification math.RT MSC 18G2518G8018G65

keywords abelianmodelstructureequivariantcategoryHoveytripleFrobeniusfunctortriangleequivalenceuptoretractsPGFmodulesskewgroupringhomotopysquare

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

This paper establishes a lifting theorem: when a finite group G acts on a Grothendieck category A with enough projectives and |G| is invertible, any cofibrantly generated hereditary Hovey triple whose classes are suitably G-invariant induces the same kind of Hovey triple on the equivariant category A^G. It then proves that the homotopy category of the lifted model structure is triangle equivalent, up to retracts, to the equivariantization of the original homotopy category, with a genuine triangle equivalence when a certain stable category is idempotent complete. The paper also shows that derived functors of Quillen adjunctions are compatible with this comparison, producing a commutative square of homotopy categories. This gives a general mechanism for passing abelian model structures, and their homotopy theories, through equivariantization, with concrete consequences for modules over skew group rings and PGF modules.

What carries the argument

The engine is the transfer theorem for Hovey triples along a faithful Frobenius functor (Theorem 2.16), specialised to the Frobenius pair Ind: A ⇄ A^G : U. A Frobenius pair is an adjunction whose right adjoint is also a left adjoint; the symmetry lets one control both cotorsion pairs of a Hovey triple simultaneously. In the equivariant setting, |G|-invertibility makes the counit Ind∘U → id split, which turns the epimorphism condition of the transfer criterion into a tautology, so the transferred classes (C^G, W^G, F^G) inherit cofibrant generation and heredity. The comparison of homotopy categories is carried by factoring Ho(γ_A^G) through the stable categories St_{ω^G}(C^G ∩ F^G) and St_ω(C ∩ F)^G, where the comparison is a triangle equivalence up to retracts.

What would settle it

Take A = Mod(Z) with G = C2 acting by the sign automorphism, so |G| = 2 is not invertible, and let M be the PGF Hovey triple on Mod(Z). Compute the lifted triple (C^G, W^G, F^G) on Mod(Z)^G and test whether it satisfies the completeness of the two cotorsion pairs required for a Hovey triple; if it does, the |G|-invertible hypothesis is unnecessary. A more direct test is to find one equivariant object X for which the counit short exact sequence 0 → ker ε_X → Ind U X → X → 0 does not split; Lemma 3.1, and hence the proof of the transfer theorem, would fail at that point.

Watch

Extended reading notes

Core claim

The central claim is Theorem A: under the |G|-invertible hypothesis, if any two of the three classes C, W, F of a cofibrantly generated hereditary Hovey triple are G-invariant, then (C^G, W^G, F^G) is again a cofibrantly generated hereditary Hovey triple on A^G, and the induced functor Ho(γ_A^G): Ho(M^G) → Ho(M)^G is a triangle equivalence up to retracts, becoming a genuine triangle equivalence when St_{ω^G}(C^G ∩ F^G) is idempotent complete. Theorem B adds that for a two-sided Quillen adjunction satisfying faithfulness, invariance, and weak-equivalence conditions on unit and counit cokernels and kernels, the square comparing L(F^G) with L(F)^G commutes up to natural isomorphism and the horizontal functors are triangle equivalences. The paper verifies the hypotheses for PGF Hovey triples, obtaining a comparison functor Ho(PGF(RG)) → Ho(PGF(R))^G that is a triangle equivalence up to retracts, and constructs homotopy squares from Frobenius bimodules and from stable equivalences of adjoint type.

Load-bearing premise

The whole argument depends on the order of the group being invertible in the category: that single hypothesis makes the counit of the induction-forgetful adjunction split, and without the split the lifted triple is not guaranteed to satisfy the completeness conditions of a Hovey triple.

Editorial extensions

If this is right

For any finite group action with |G| invertible, a cofibrantly generated hereditary abelian model structure whose classes are G-invariant transfers to the equivariant category, so equivariant homological algebra inherits the model structure.
The homotopy categories Ho(M^G) and Ho(M)^G coincide up to retracts; when the relevant stable category is idempotent complete, they are genuinely triangle equivalent.
Total derived functors commute with equivariantization: L(F^G) and L(F)^G sit in a commutative square, so equivariant derived functors can be computed either before or after passing to fixed points.
For rings, the right orthogonal class PGF(RG)^⊥ equals (PGF(R)^⊥)^G, and Ho(PGF(RG)) is triangle equivalent up to retracts to Ho(PGF(R))^G.
Stable equivalences of adjoint type between base rings lift to the skew group rings, giving triangle equivalences of the associated equivariant PGF homotopy categories.

Reading between the lines

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

The |G|-invertibility condition appears to be the real dividing line: without it, the split counit used in the transfer proof is unavailable, so one should expect genuinely different behaviour, and testing whether some weaker averaging condition suffices would be a natural next step.
The "up to retracts" caveat is not cosmetic; any failure of idempotent completeness in the equivariant stable category would make the comparison fail to be a full equivalence, so examples with non-idempotent-complete stable categories could separate the two notions.
The transfer machinery is stated for Grothendieck categories, but its core transfer theorem is purely about Frobenius functors; applying it to other Frobenius adjunctions, such as induction between module categories over Frobenius extensions, may yield analogous lifting results without any group action.
The PGF illustration suggests the same square should exist for Gorenstein projective, injective, and flat Hovey triples; the paper states the verification is analogous, so those cases are a promising place to test whether the hypotheses can be relaxed.

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.

Desk Editor's Note

read the letter

The paper is worth taking seriously. Its real contribution is a transfer theorem for cofibrantly generated hereditary Hovey triples along Frobenius functors, then applied to equivariant categories. Theorem A, comparing the homotopy category of the lifted model structure with the equivariantization of the original homotopy category up to retracts, is a genuine extension of the Chen/Elagin/Sun results from derived and stable categories to abelian model structures. Theorem B, the homotopy square compatibility for derived functors, is also new and is the kind of result people will want to cite. The PGF module corollaries give concrete payoff, and the authors are honest about which results are imported from Gillespie, Sun, and Chen. The stress-test note is correct. In Prop. 2.11, proof of the inclusion ^\perp H^{-1}(D) ⊆ H^{-1}(C), the displayed chain uses the identity ^\perp((^\perp F(D))^\perp) = ^\perp F(D). That identity is not valid in general; the double left orthogonal can be strictly larger. However, the repair is immediate: condition (a) already gives F(S) ⊆ ^\perp F(D), so for X in ^\perp(F(S)^\perp) one directly gets Ext^1_B(X, F(D))=0 for all D in D, hence X in ^\perp F(D). The reverse inclusion is fine. So the central claim of Prop. 2.11 is likely true, but the written proof is incomplete at exactly the point that feeds Theorem 2.16 and Theorem A(1). A referee should ask for that fix. Other soft spots are proportionate. The split counit in Lemma 3.1 depends on |G| being invertible; that is a real restriction, but it is stated plainly and is also needed for Sun's triangulated equivariant structure. Theorem B's hypotheses are heavy, especially the condition on coker(eta_X) and ker(epsilon_Y) and the faithfulness assumptions, but the statements are explicit and the applications to Frobenius bimodules and stable equivalences of adjoint type show the conditions are not empty. The proofs are long and depend on many imported results, so I could not fully check every diagram. The citation pattern looks healthy; I did not see self-citation inflation or fitting. Who gets value: anyone working on abelian model structures, Gorenstein homological algebra, or equivariant categories. The paper deserves a serious referee, not a desk reject. I would send it to someone who can check Section 2 carefully and the compatibility square in Theorem B. With the Prop. 2.11 repair, I expect the central results to stand.

Referee Report

1 major / 5 minor

Summary. The paper develops a transfer theorem for abelian model structures along Frobenius functors (Theorem 2.16 and 2.18) and applies it to the equivariant category A^G of a Grothendieck category A equipped with a finite group action with |G| invertible. Its Theorem A states that if two of the three classes of a cofibrantly generated hereditary Hovey triple are G-invariant, then the equivariant classes form such a triple, and the induced functor Ho(A^G) -> Ho(A)^G is a triangle equivalence up to retracts, with a genuine equivalence under an idempotent completeness condition. Theorem B gives conditions under which a Quillen adjunction between two such categories induces a commutative square relating L(F^G) and L(F)^G, with both horizontal functors triangle equivalences. Section 5 illustrates the results for PGF Hovey triples over skew group rings, Frobenius bimodules, and stable equivalences of adjoint type.

Significance. If the main theorems are correct, the paper provides a general and useful framework for equivariantizing abelian model structures, complementing the transfer results of Hovey, Gillespie, Sun, and Chen. The paper is explicit about which results are imported and provides detailed proofs for the new lifting lemmas; the PGF corollaries give concrete, checkable applications. The principal caveats are the essential hypothesis that |G| is invertible and the reliance on Sun's theorem for the triangulated comparison; these are stated clearly. The identified gap in Proposition 2.11 is localized and repairable, so the overall contribution remains valuable.

major comments (1)

[Section 2, Proposition 2.11] The proof of the implication (⇐) contains the displayed chain ^⊥H^{-1}(D)=^⊥(F(S)^⊥)⊆^⊥((^⊥F(D))^⊥)=^⊥F(D)=H^{-1}(C). The middle equality ^⊥((^⊥F(D))^⊥)=^⊥F(D) is not valid for arbitrary classes in an abelian category, since the double left orthogonal can be strictly larger. This step is used to prove the inclusion ^⊥H^{-1}(D)⊆H^{-1}(C), which is needed to identify (^⊥H^{-1}(D), H^{-1}(D)) with the cotorsion pair cogenerated by F(S). The gap is repairable directly from condition (a): since F(S)⊆^⊥F(D), every F(D) with D∈D lies in F(S)^⊥=H^{-1}(D); hence for X∈^⊥(F(S)^⊥) one has Ext^1_B(X,F(D))=0 for all D, so X∈^⊥F(D)=H^{-1}(C). I recommend replacing the invalid equality by this direct argument and checking the surrounding identifications in Theorem 2.16 and Theorem A(1), which rest on this proposition.

minor comments (5)

[General] The title and running header contain the typo 'EQUIV ARIANT'; the word should be 'EQUIVARIANT'.
[Section 2.10 and Proposition 2.11] The notation λ_X in the statement of Proposition 2.11 is introduced via the short exact sequence for objects of B, but the variable name is reused; writing the sequence explicitly for each object appearing in conditions (b) and (c) would remove ambiguity.
[Section 4, proof of Theorem 4.11] In Step 1 the symbol Q_A^G is used both for the equivariant cofibrant replacement functor and for the underlying object Q_A^G X; clarifying that these are the same object-level notation would help the reader follow the long diagram chase.
[Section 5.3, Lemma 5.19] The proof of Lemma 5.19 is condensed and relies on facts (1) and (2) that are only cited; expanding the verification of these two facts, or at least giving precise references with the exact adjunction isomorphisms, would improve readability.
[References] Reference [15] is cited as a preprint without a year; if a published version exists, it should be updated.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the transfer and comparison results rest on external theorems, with no self-citation or fitted-input reduction.

full rationale

The paper's central derivation chain is self-contained in the relevant sense. Theorem A(1) uses the transfer criterion Theorem 2.16, which is proved from cotorsion-pair manipulations, and applies it to the Frobenius pair Ind ⟂ U; the lifted triple M^G is defined by preimage classes U^{-1}(C), U^{-1}(W), U^{-1}(F), and the proof verifies the hypotheses using the splitness of the counit supplied by Lemma 3.1, a genuine condition requiring invertibility of |G|. This is a sufficient condition, not a restatement of the conclusion. Theorem A(2) invokes Sun's external theorem for the canonical triangulated structure on equivariant categories, and Theorem A(3) assembles the comparison functor from Hovey's localization theorem, Gillespie's stable-category description, and Sun's comparison theorem [36, Theorem 3.14]; these are external benchmarks, not results derived from the paper's own conclusions. The PGF illustration uses the external complete hereditary cotorsion pair of Šaroch and Št'ovíček. No parameter is fitted, no quantity called a prediction is defined in terms of the claimed output, and no self-citation is load-bearing; in fact, no self-citations by the present authors appear. One non-circular correctness concern exists: the proof of Proposition 2.11 uses the identity ^⊥((^⊥F(D))^⊥) = ^⊥F(D), which is not valid in general for arbitrary classes in an abelian category. This is a genuine gap in the written proof, but the skeptical repair shows that the needed inclusion follows without the invalid identity, so the theorem is not reduced to its own assumptions. The gap is therefore a correctness issue, not circularity.

Assumptions & free parameters 0 free parameters · 7 assumptions · 0 invented entities

The paper introduces no free parameters or new postulated objects. It constructs comparison functors (e.g., Theta and St(K^G)) from existing ingredients, and it relies on standard theorems plus explicitly stated structural hypotheses.

assumptions (7)

standard math Hovey's correspondence (Theorem 1.6) between abelian model structures and Hovey triples.
Used throughout to translate between model structures and triples of classes (C,W,F) of objects.
standard math Gillespie's results on homotopy categories of hereditary Hovey triples (Theorems 1.12 and 1.14).
Provides localization functor and triangle equivalence between stable category and homotopy category.
standard math Sun's theorem that admissible G-actions on triangulated categories yield canonical pre-triangulated structures when |G| is invertible (Lemma 3.8, [36]).
Used to triangulate Ho(M)^G and St(C∩F)^G in Theorem A(2).
domain assumption A is a Grothendieck category with enough projectives and |G| is invertible in A.
Ensures A^G is Grothendieck with generators and the counit of Ind ⊣ U splits (Lemma 3.1).
domain assumption M = (C,W,F) is a cofibrantly generated hereditary Hovey triple and any two of C,W,F are G-invariant.
The main hypotheses of Theorem A; used to verify the transfer conditions in Theorem 2.16.
domain assumption For Theorem B: F,H are faithful, both (F,H) and (H,F) are Quillen adjunctions, coker(η_X) ∈ W_A for X ∈ C_A, and ker(ε_Y) ∈ W_B for Y ∈ F_B.
These are the extra conditions needed for the horizontal functors to be triangle equivalences.
standard math For PGF illustrations: (PGF(R), PGF(R)^⊥) is a complete hereditary cotorsion pair cogenerated by a set (Saroch-Stovicek).
Gives the PGF Hovey triple on Mod(R).

how reviews work

0 comments

Cite this review

Pith. "Pith review of Transfer of abelian model structures to equivariant categories and homotopy squares." pith.science (2026). https://pith.science/paper/5RIRQS6K

@misc{pith2026260808141,
  author       = {Pith},
  title        = {Pith review of: Transfer of abelian model structures to equivariant categories and homotopy squares},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5RIRQS6K}},
  note         = {Machine review of arXiv:2608.08141}
}

abstract

Let $G$ be a finite group acting on a Grothendieck category $\mathcal{A}$ with enough projectives, such that $|G|$ is invertible in $\mathcal{A}$. We prove a general lifting theorem for abelian model structures from $\mathcal{A}$ to its equivariant category $\mathcal{A}^G$, and establish a triangle equivalence up to retracts between the corresponding homotopy categories. We also construct a commutative square whose horizontal functors are triangle equivalences and whose vertical comparison functors are triangle equivalences up to retracts. This square relates derived functors on the lifted equivariant model categories to the equivariantizations of the derived functors on the original homotopy categories. In the module category setting, we illustrate the above results using the PGF Hovey triples, and apply them to homotopy squares induced by a Frobenius bimodule and by a stable equivalence of adjoint type.

Discussion (0). Continue with ORCID to comment.

Pith tools

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