Homotopical presentation of categories

Seunghun Lee

REVIEW 3 major objections 5 minor 1 cited by

Homotopical presentation of categories

T0 review · 3 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 The paper establishes a sufficient condition on a functor between small categories for it to generate a homotopical presentation of the universal model category of its target.

desk verdict Alleged circularity in the proof is a misreading; the dependency loop is actually acyclic, and the paper is a solid contribution needing only a small fix in Lemma 4.4. read the letter →

arxiv 2411.15790 v1 pith:HS5NTN4E submitted 2024-11-24 math.CT

Seunghun Lee This is my paper · ORCID

classification math.CT MSC 18N4014C20

keywords universalmodelcategoryhomotopicalpresentationQuillenequivalenceleftBousfieldlocalizationadmissiblesubcategorycommacalculusofrightfractionssimplicialpresheaves

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 gives a criterion for a functor between small categories to present the universal model category of its target by generators and relations. If the source category has finite limits, the functor preserves them, is essentially surjective on objects, and each associated comma category contains a cofiltered admissible subcategory with a lifting property, then localizing the source's universal model category at the corresponding morphisms yields a Quillen equivalence with the target's. This matters because it turns a homotopical statement about model categories into finitary categorical data that can be checked object by object. The criterion recovers known characterizations: equivalences of categories coincide with universal equivalences, and localizations admitting a calculus of right fractions induce Quillen equivalences.

What carries the argument

The load-bearing object is the $F$-admissible subcategory: for each object $d$ of $C$, one looks at the comma category $C_d$ whose objects are morphisms $e\to d$ in $C$ that $F$ sends to isomorphisms in $B$, and a subcategory $A_d$ is $F$-admissible when it is cofiltered and $F$ has the $A_d$-lifting property, meaning every morphism $Fd\to Fe$ in $B$ can be traced back through an object of $A_d$. Such subcategories are used to prove that the restricted functor $F_d|_{A_d}$ is initial, and that is what makes the filtered colimits over the comma categories compute correctly. The proof then runs on the standard Quillen-equivalence recognition criterion: show the derived counit is an isomorphism and the derived unit is a weak equivalence.

What would settle it

Let $F:C\to \Delta[0]$ be the unique functor from the two-object poset $C=\{0<1\}$, which has finite limits, to the terminal category; the theorem predicts that localizing $U(C)$ at the single non-identity map yields a model category Quillen equivalent to $U(\Delta[0])$. Computing the homotopy category of that localization directly would settle whether the criterion holds in this simplest case.

Watch

Extended reading notes

Core claim

The central claim is Theorem 1.4: for small categories $C$ and $B$ and a functor $F:C\to B$ with $C$ finitely complete and $F$ preserving finite limits, $F$ generates a homotopical presentation of $B$ with generator $C$ and relation $S_A$ whenever $F$ is essentially surjective on objects and, for every object $d$ of $C$, there is an $F$-admissible subcategory $A_d$ of the comma category $C_d$. Concretely, the induced Quillen adjunction $F_*: L_{S_A}(U(C)) \rightleftarrows U(B): F^*$ is a Quillen equivalence, so the homotopy theory of $U(B)$ is presented by $C$ with the morphisms in $S_A$ inverted. The proof shows that under these hypotheses the derived counit is an isomorphism and the derived unit is a weak equivalence after localization. This gives a direct model-categorical route to results that are often treated with infinity-categorical machinery.

Load-bearing premise

The load-bearing premise is that the source category $C$ has all finite limits and the functor $F$ preserves them.

Editorial extensions

If this is right

The theorem reduces the existence of a small presentation to a check on comma categories: essentially surjective on objects plus, for each $d$, an $F$-admissible subcategory $A_d$ of $C_d$.
Corollary 1.6 characterizes the presentation using all $F$-inverted morphisms: it holds exactly when $F$ is essentially surjective and has the $C_d$-lifting property for every $d$.
Corollary 1.9 identifies equivalences of categories with universal equivalences under the finite-limit hypothesis, so the universal model category reflects weak equivalences on that class of categories.
Corollary 1.12 gives a model-categorical version of the calculus of right fractions: for a fractionable set $\Sigma$, $L_\Sigma(U(C))$ is Quillen equivalent to $U(C[\Sigma^{-1}])$.
The examples use the criterion to produce presentations for sheaves and topological spaces, and show that $U(-)$ does not preserve Quillen equivalences in general.

Reading between the lines

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

A practical corollary the author leaves implicit is that the criterion turns the search for a Quillen equivalence into a finite-limit-plus-lifting check on comma categories, so one can hunt for presentations by looking for cofiltered subcategories instead of constructing model structures.
The same admissible-subcategory pattern should dualize to co-universal model categories and left fractions, and the paper states a dual version in Proposition 5.9; testing whether the lifting condition can be relaxed to filtered admissible subcategories on colimit comma categories would be a natural next step.
The paper's closing remark points toward algebraic geometry: ampleness of a divisor could be characterized by whether a certain functor's comma categories admit admissible subcategories, which would make positivity a homotopical-presentation question.

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

Dear colleague,

I read Lee's 'Homotopical presentation of categories' with the referee's concerns in hand. The alleged circularity between Corollary 1.12 and Corollary 1.6 does not hold up. Corollary 1.12 invokes only the (1)->(2) direction of Corollary 1.6, which follows directly from Theorem 1.4. The converse direction (2)->(1) of Corollary 1.6 uses Corollary 1.12 afterwards. The dependency graph is acyclic: Theorem 1.4 => Cor 1.6(1->2) => Cor 1.12 => Cor 1.6(2->1). So the referee's strongest objection is a misreading.

What is actually new is Theorem 1.4: a criterion for a finite-limit-preserving functor F:C->B, essentially surjective, with F-admissible subcategories, to generate a homotopical presentation of U(B). The proof is careful and largely self-contained, with the key Lemma 3.11 showing F_d restricted to an admissible subcategory is initial. The three corollaries are known results in infinity-category clothing, explicitly acknowledged as such, but the direct model-categorical proofs here have independent value, particularly for someone who wants to avoid quasi-category technology. The examples (sheaf localization, sSet/Top, nerve) are instructive.

The real soft spot is small. In Lemma 4.4, the cofibrant replacement of F*Y is taken in L_S_A(UpC). But the proof uses that this replacement is an objectwise weak equivalence in UpC. A cofibrant replacement in the localized category need not be an objectwise equivalence, since the weak equivalences are coarser. The fix is to choose a cofibrant replacement in UpC, which is also cofibrant in the localization (cofibrations are unchanged) and gives an objectwise weak equivalence. This is an immediate repair and does not affect the statement of Theorem 1.4. Similarly, Remark 4.7 promises a direct proof of Corollary 1.12 that would bypass the result, but it isn't supplied; since the proof via Cor 1.6 is logically sound, this is only a matter of presentation.

The blanket finite-limit assumption on C and preservation by F is strong, but the author explicitly notes in Remark 1.10 that the characterization is not optimal, pointing to Cauchy completion. That is honest.

Overall: the main theorem is new, the proofs are mostly rigorous, and the flaws are minor. I'd send this to peer review with a request for the small fix in Lemma 4.4. Worth citing for the criterion and the direct proofs. Bring to reading group if you're working on presentations of model categories.

Referee Report

3 major / 5 minor

Summary. The paper gives a criterion for a functor F:C→B between small categories to generate a small presentation of the universal model category U(B) in the sense of Dugger. Under the assumptions that C has finite limits, F preserves finite limits, F is essentially surjective, and for each d∈C there exists an F-admissible subcategory A_d of C_d, Theorem 1.4 establishes a Quillen equivalence between L_{S_A}U(C) and U(B). The proof goes through a sequence of lemmas on comma categories, initial functors, filtered colimits, and the derived unit and counit of the induced adjunction. The paper derives three corollaries: a characterization of equivalences of categories by universal equivalences (Corollary 1.9), a Quillen equivalence for localizations satisfying the calculus of right fractions (Corollary 1.12), and a discussion of when a functor generates a presentation. It closes with examples involving sheaf categories, simplicial sets and topological spaces, and categories.

Significance. If the main theorem is correct, it provides a useful, fairly general presentation theorem for universal model categories, with direct model-categorical proofs of statements often obtained via ∞-category theory. The proof of Theorem 1.4 is self-contained and detailed, and the paper is careful about the finite-limit hypotheses. The alleged circularity between Corollary 1.12 and Corollary 1.6 is a misreading: Corollary 1.12 uses only the (1)⇒(2) direction of Corollary 1.6, which is proved from Theorem 1.4, so the dependency is acyclic. I found no error in the main theorem's proof beyond the clarification needed in Lemmas 4.4 and 4.5. However, one of the examples, Example 5.6, is invalid, and there are several local gaps in proofs that need to be addressed.

major comments (3)

[§5, Example 5.6] Example 5.6 is not valid as written. The example asserts that Sing is fully faithful for the adjunction |•|:sSet⇄Top:Sing; as an ordinary functor, Sing is faithful but not full, since full faithfulness holds only after deriving, not on the nose. Consequently Proposition 5.3 cannot be applied to this adjunction. Moreover, the claim that the localization map (5.11) is not injective is unsupported: the two endpoint inclusions Δ0→Δ1 are not identified by localizing at morphisms whose geometric realization is a homeomorphism. Please replace this example with a correct one or remove it, and do not use it as evidence that Up(−) does not preserve Quillen equivalences.
[§4, Lemmas 4.4 and 4.5] In Lemma 4.4 and in Lemma 4.5, the proofs use the assertion that the cofibrant replacement pF*Y of F*Y is a weak equivalence in Up(C). This is not automatic for an arbitrary cofibrant replacement taken in L_{S_A}(Up(C)); the filtered-colimit argument only applies when pF*Y is an objectwise weak equivalence. The proof should explicitly choose pF*Y to be a cofibrant replacement in Up(C), which is also a cofibrant replacement in L_{S_A}(Up(C)), and note this at the point where (4.11) and the corresponding part of Lemma 4.5 are used.
[§3, Lemma 3.14] Lemma 3.14 is stated without proof and is used in the proof of Corollary 1.6 to deduce the Cd-lifting property from fullness of the functor H. This is a load-bearing point for the converse direction of Corollary 1.6. The lemma is a consequence of Lemma 3.15(2) applied to the subcategories A^Σ_d, since Σ⊆C_d, but the argument is not given. Please add a proof or a precise reference.

minor comments (5)

[§1, proof of Corollary 1.9] The sentence 'Thus Corollary 1.6 reduces to Corollary 1.9' appears to have the direction reversed; the surrounding argument shows that Corollary 1.9 follows from Corollary 1.6. Please rephrase to make the logical flow clear.
[§4, Remark 4.7] Remark 4.7 states that Corollary 1.12 can be proved directly using Theorem 1.4 and Lemma 3.18, but no direct proof is given. Either supply the promised proof or remove the claim.
[§3, proof of Lemma 3.11] The proof of Lemma 3.11 is hard to follow because diagram (3.17) uses the symbol e1 for several different objects. Please rewrite the proof with distinct labels for the objects involved in the cofilteredness and equalizer steps.
[Throughout] There are several grammatical and typographical errors: for example, 'F have Ad-lifting property' in Definitions 3.7 and 3.8 should be 'F has'; Remark 4.8 'homotopical pretension' should presumably be 'homotopical presentation'; the header of the paper contains a mis-spaced 'PRESENT A TION'. Please correct these.
[§1, Remark 1.7] Remark 1.7 states that SC admits the calculus of right fractions without proof; this fact is used in the proof of Corollary 1.6. Please add a reference or a one-sentence justification.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the dependency between Corollary 1.6 and Corollary 1.12 is acyclic.

full rationale

The main theorem (Theorem 1.4) is proved internally from Lemmas 4.1-4.4, whose proofs use only the finite-limit assumptions, cofilteredness of comma categories (Lemma 3.6), initiality of admissible restrictions (Lemma 3.11), and standard model-category facts. The apparent mutual dependence flagged in the reader's take is a misreading: Section 4 states that the (1)->(2) direction of Corollary 1.6 follows from Theorem 1.4, and that proof is completed before Corollary 1.12 is proved. Corollary 1.12 then applies that already-established direction to the localization functor. The remaining direction (2)->(1) of Corollary 1.6 is proved later using Lemma 4.5, Lemma 4.6, and Corollary 1.12, closing an acyclic dependency chain: Theorem 1.4, then Corollary 1.6(1=>2), then Corollary 1.12, then Corollary 1.6(2=>1). Corollary 1.9 is terse, saying that Corollary 1.6 reduces to it, but it is not load-bearing for the main theorem. Remark 4.7 claims a direct proof of Corollary 1.12 without giving it, but this omission is harmless because Corollary 1.12 already has a complete proof via Corollary 1.6(1=>2). No fitted inputs are renamed as predictions, no load-bearing self-citation chain is used, and no theorem is assumed in its own proof.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

No free parameters or invented entities appear; the paper is a pure mathematics proof. The axioms are standard results from the model-category and localization literature, plus the domain assumption of finite-limit preservation that the theorem states explicitly.

assumptions (5)

standard math The Bousfield-Kan model structure on simplicial presheaves U(C) is a left proper, simplicial, cofibrantly generated model category.
Invoked in Section 2.2 and used throughout; cited from [BK72] and [Dug01b].
standard math For every small category C and set S of morphisms in C, the left Bousfield localization L_S U(C) exists and inherits a left proper simplicial model structure from U(C).
Used in Definition 1.1 and the proof of Theorem 1.4; cited from [Hir03].
standard math Dugger's universal model category U(B) and the associated Quillen adjunction (1.3) give the notion of small presentation; a Quillen equivalence as in (1.1) presents U(B).
The framework of the central claim; cited from [Dug01b].
standard math If Sigma admits the calculus of right fractions in a category C with finite limits, the localization functor C -> C[Sigma^{-1}] preserves finite limits and the localized category has finite limits.
Used in the proof of Corollary 1.12 and Proposition 5.3; cited from [GZ67] and [KS06].
domain assumption C has finite limits and F preserves finite limits.
Assumed in Theorem 1.4 and all corollaries; used to prove C_d cofiltered and to commute with equalizers and pullbacks. The author notes in Remark 1.10 that the equivalence characterization is not optimal without it.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Homotopical presentation of categories." pith.science (2026). https://pith.science/paper/HS5NTN4E

@misc{pith2026241115790,
  author       = {Pith},
  title        = {Pith review of: Homotopical presentation of categories},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HS5NTN4E}},
  note         = {Machine review of arXiv:2411.15790}
}

abstract

We give a criterion for a functor $F:C\rightarrow B$ between small categories to generate a small presentation of the universal model category $U(B)$ in the sense of Dugger.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

Q-divisor and Ampleness
math.AG 2024-11 reject novelty 5.0 of 10

The paper proves that ampleness of a divisor is equivalent to every subvariety being a cosupport of a multiplier ideal sheaf in the divisor's rational multiples, and equivalently to a model-categorical presentation of...

Reference graph

Works this paper leans on

24 extracted references · 17 canonical work pages · cited by 1 Pith paper

[1]

Artin, Grothendieck topologies, Harvard university, 1962, Notes on a Seminar

M. Artin, Grothendieck topologies, Harvard university, 1962, Notes on a Seminar

1962
[2]

thesis, Harvard University, 2019, arXiv:2004.12937 [math.CT] https://arxiv.org/abs/2004.12937

Reid William Barton, A model 2-category of enriched combinatorial premodel categories, Ph.D. thesis, Harvard University, 2019, arXiv:2004.12937 [math.CT] https://arxiv.org/abs/2004.12937

arXiv 2019
[3]

Différentielle Catég

Francis Borceux and Dominique Dejean, Cauchy completion in category theory, Cahiers Topologie Géom. Différentielle Catég. 27 (1986), no. 2, 133–146

work page 1986
[4]

A. K. Bousfield and D. M. Kan, Homotopy limits, completions and localizations, Lecture notes in mathematics, vol. 304, Springer, 1972

work page 1972
[5]

Brown, Abstract homotopy theory and generalized sheaf cohomology, Trans

Kenneth S. Brown, Abstract homotopy theory and generalized sheaf cohomology, Trans. Amer. Math. Soc. 186 (1973), 419--458

work page 1973
[6]

180, Cambridge University Press, 2019

Denis-Charles Cisinski, Higher categories and homotopical algebra, Cambridge Studies in Advanced Mathematics, no. 180, Cambridge University Press, 2019

2019
[7]

Daniel Carranza, Chris Kapulkin, and Zachery Lindsey, Calculus of fractions for quasicategories, arXiv:2306.02218 [math.AT]

work page Pith review arXiv
[8]

Daniel Dugger, Combinatorial model categories have presentations, Adv. Math. 164 (2001), no. 1, 177--201

work page 2001

Show all 24 references

[9]

, Universal homotopy theories, Adv. in Math. 164 (2001), no. 1, 144--176

2001
[10]

Alexander Grothendieck, Sur quelques points d'algèbre homologique, Tohoku Math. J. 9 (1957), no. 2, 119–221

1957
[11]

Gabriel and M

P. Gabriel and M. Zisman, Calculus of fractions and homotopy theory, 1st ed., Ergebnisse der Mathematik und ihrer Grenzgebiete, vol. 35, Springer, 1967

1967
[12]

24, 875--958

Simon Henry, Weak model categories in classical and constructive mathematics, Theory and A pplications of C ategories 35 (2020), no. 24, 875--958

2020
[13]

Hirschhorn, Model categories and their localizations, Mathematical Surveys and Monographs, vol

Philip S. Hirschhorn, Model categories and their localizations, Mathematical Surveys and Monographs, vol. 99, American Mathematical Society, 2003

2003
[14]

, Notes on homotopy colimits and homotopy limits https://math.mit.edu/ psh/notes/hocolim.pdf , available at author's hompage, 2014

2014
[15]

63, American Mathematical Society, 1999

Mark Hovey, Model categories, Mathematical Surveys and Monographs, vol. 63, American Mathematical Society, 1999

1999
[16]

André Joyal, https://ncatlab.org/joyalscatlab/published/Model+structures+on+Cat Model structures on C at , available at Joyal's C at L ab
[17]

332, Springer-Verlag, 2006

Masaki Kashiwara and Pierre Schapira, Categories and sheaves, Grundlehren der Mathematischen Wissenschaften [Fundamental Principles of Mathematical Sciences], vol. 332, Springer-Verlag, 2006

2006
[18]

5, Springer, 1998

Saunders Mac Lane, Categories for the working mathematician, 2st ed., Graduate Texts in Mathematics, vol. 5, Springer, 1998

1998
[19]

36, 1163--1180

Seunghun Lee, Building a model category out of cofibrations and fibrations: the two out of three property for weak equivalence, Theory and A pplications of C ategories 30 (2015), no. 36, 1163--1180

2015
[20]

, Q-divisor and ampleness, prerprint, 2024

2024
[21]

Mathoverflow, https://mathoverflow.net/questions/18744/is-model-structure-on-catset-unique/46471\#46471 Is model structure on CatSet unique ? , 2010

2010
[22]

Peter May and Kate Ponto, More concise algebraic topology, Chicago lectures in mathematics, The university of Chicago press, 2012

J. Peter May and Kate Ponto, More concise algebraic topology, Chicago lectures in mathematics, The university of Chicago press, 2012

2012
[23]

43, Springer-Verlag, 1967

Daniel Quillen, Homotopical algebra, Lecture notes in mathematics, no. 43, Springer-Verlag, 1967

1967
[24]

Charles Rezk, A model category for categories https://ncatlab.org/nlab/files/Rezk_ModelCategoryForCategories.pdf , available at ncatlab.org, 1996

1996

Pith tools

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

[1] [1]

Artin, Grothendieck topologies, Harvard university, 1962, Notes on a Seminar

M. Artin, Grothendieck topologies, Harvard university, 1962, Notes on a Seminar

1962

[2] [2]

thesis, Harvard University, 2019, arXiv:2004.12937 [math.CT] https://arxiv.org/abs/2004.12937

Reid William Barton, A model 2-category of enriched combinatorial premodel categories, Ph.D. thesis, Harvard University, 2019, arXiv:2004.12937 [math.CT] https://arxiv.org/abs/2004.12937

arXiv 2019

[3] [3]

Différentielle Catég

Francis Borceux and Dominique Dejean, Cauchy completion in category theory, Cahiers Topologie Géom. Différentielle Catég. 27 (1986), no. 2, 133–146

work page 1986

[4] [4]

A. K. Bousfield and D. M. Kan, Homotopy limits, completions and localizations, Lecture notes in mathematics, vol. 304, Springer, 1972

work page 1972

[5] [5]

Brown, Abstract homotopy theory and generalized sheaf cohomology, Trans

Kenneth S. Brown, Abstract homotopy theory and generalized sheaf cohomology, Trans. Amer. Math. Soc. 186 (1973), 419--458

work page 1973

[6] [6]

180, Cambridge University Press, 2019

Denis-Charles Cisinski, Higher categories and homotopical algebra, Cambridge Studies in Advanced Mathematics, no. 180, Cambridge University Press, 2019

2019

[7] [7]

Daniel Carranza, Chris Kapulkin, and Zachery Lindsey, Calculus of fractions for quasicategories, arXiv:2306.02218 [math.AT]

work page Pith review arXiv

[8] [8]

Daniel Dugger, Combinatorial model categories have presentations, Adv. Math. 164 (2001), no. 1, 177--201

work page 2001