Continuous Hochschild Cohomology and Formality

Patrick Antweiler

REVIEW 3 major objections 3 minor 11 references

Continuous Hochschild Cohomology and Formality

T0 review · 3 major / 3 minor · reviewed 2026-08-03 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The continuous Hochschild cohomology of the Dolbeault algebra of any complex manifold is computed by the HKR map, and its degree-2 part is the generalized-complex deformation complex.

desk verdict Interesting framework and a plausible Dolbeault formality, but the foundation is false and the de Rham theorem has a concrete counterexample; this paper is not ready. read the letter →

arxiv 2512.21090 v2 pith:SFOTWZMS submitted 2025-12-24 math.QA math-phmath.CTmath.MP

Patrick Antweiler This is my paper

classification math.QAmath-phmath.CTmath.MP MSC 16E4018G8053D1832G0553D55

keywords continuousHochschildcohomologycontraderivedcategoriescompletelocallyconvexdg-algebrasformalityDolbeaultalgebrageneralizedcomplexgeometrymatrixfactorisationsdeformationtheory

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 builds a homological setting for complete locally convex (curved) dg-algebras—contraderived categories, in which products rather than sums are the primitive operation—and shows that in this setting the Hochschild cohomology of algebras of smooth functions, forms, and Dolbeault forms is computed by an explicit continuous Hochschild complex. Its central result is a formality theorem for the Dolbeault algebra of a complex manifold: the HKR map from holomorphic polyvector fields with Dolbeault-form coefficients to continuous Hochschild cochains is a quasi-isomorphism, and it extends to an L∞-quasi-isomorphism of dg-Lie algebras. Consequently, the second continuous Hochschild cohomology of the Dolbeault algebra is the deformation complex that controls deformations of the manifold as a generalized complex manifold, namely H^0(X,Λ^2 T_X) ⊕ H^1(X,T_X) ⊕ H^2(X,O_X). The same machinery yields formality for smooth functions and the de Rham algebra, and computes continuous Hochschild cohomology for several families of matrix factorisations.

What carries the argument

The carrier of the argument is the continuous contraderived category of a complete locally convex dg-algebra A: the Verdier quotient of the dg-category of all dg-modules by the class of contraacyclic modules, equivalently the homotopy category of graded-projective dg-modules. The key identity is that the completed projective tensor product commutes with arbitrary products of complete locally convex spaces, so arbitrary products of graded-free modules remain graded-free; this makes the two-sided bar construction a graded-projective resolution of A over its enveloping algebra. The continuous Hochschild complex obtained by taking Hom out of that resolution carries a canonical B∞-algebra structu

What would settle it

Take X = C^n and compute the degree-2 cohomology of the continuous Hochschild complex of the Dolbeault algebra A(C^n) directly from continuous multilinear cochains, following the paper's explicit local resolution; the theorem predicts the HKR map induces an isomorphism onto ΛT^{1,0}C^n⊗A(C^n), so any extra 2-cocycle modulo coboundaries would refute the main claim. Alternatively, test Lemma A.2 on the natural map (∏_k E_k) ⊗̂ N → ∏_k(E_k ⊗̂ N) by comparing continuous duals; if the duals differ, a lemma used throughout the framework is false.

Watch

Extended reading notes

Core claim

The paper's central claim is that for every complex manifold X, the continuous Hochschild complex of the Fréchet Dolbeault dg-algebra A(X) is L∞-quasi-isomorphic—a homotopy-coherent equivalence of dg-Lie algebras—to the complex ΛT^{1,0}X ⊗ A(X)[-1], whose sections are holomorphic polyvector fields with coefficients in Dolbeault forms. The comparison is the HKR map, which sends a polyvector field to the continuous Hochschild cochain obtained by contraction and multiplication. The proof is local first: an explicit graded-free resolution of A(U) as a bimodule over A(U×U) is constructed for open U⊂C^n, and the HKR map is shown to be a quasi-isomorphism after applying Hom. The local statement is

Load-bearing premise

The argument rests on Lemma A.2, the claim that the completed projective tensor product commutes with arbitrary direct products of complete locally convex spaces; if that identity fails in general, the graded-projective model of the contraderived category, and with it the sheaf-theoretic formality proofs, no longer go through.

Editorial extensions

If this is right

For a complex manifold X, continuous Hochschild cohomology in degree 2 is H^0(X,Λ^2T_X)⊕H^1(X,T_X)⊕H^2(X,O_X); this makes HH_cont^2(A(X)) a single algebraic object containing both deformations of the complex structure and the extra directions that lead to generalized complex geometry.
The L∞-quasi-isomorphism turns the continuous Hochschild complex into a dg-Lie model for the deformation theory of the Dolbeault algebra, so formal deformations of the algebra correspond to Maurer–Cartan elements in ΛT^{1,0}X⊗A(X)[-1].
For smooth functions on a real manifold, the sheaf-theoretic proof replaces analytic globalization by a short descent argument, recovering the result that continuous Hochschild cohomology is the space of polyvector fields and upgrading it to an L∞-quasi-isomorphism.
For the de Rham algebra, the inclusion of the shifted de Rham complex into the continuous Hochschild complex is a quasi-isomorphism of dg-Lie algebras, identifying the Hochschild deformation complex of the dg-category of ∞-local systems.
For matrix factorisations attached to a function f, continuous Hochschild cohomology is the Koszul-type complex with differential ι_df, with explicit answers for smooth functions, formal power series, polynomial algebras, and holomorphic functions on a complex manifold.

Reading between the lines

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

A closer look at the paper's Lemma A.2 suggests that the product–tensor commutation may fail for arbitrary complete locally convex spaces; if so, the contraderived-category model should be restricted to settings where the identity does hold—Fréchet spaces with countable products, for instance—which still covers the manifolds treated here.
The B∞-structure on the continuous Hochschild complex is announced, but only the dg-Lie part is shown in detail; checking that the higher operations satisfy the homotopy Gerstenhaber relations is a concrete test of the deformation-theoretic conclusions.
The same sheaf-theoretic globalization is likely to prove formality for other fine sheaves of complete locally convex dg-algebras admitting local Koszul resolutions, including twisted Dolbeault algebras with a closed B-field, possibly producing L∞-quasi-isomorphisms where the paper only proves a quasi-isomorphism.
If the twisted case can be upgraded from compact Kähler to arbitrary complex manifolds, the continuous Hochschild complex becomes a deformation complex for generalized complex structures with B-fields, linking the algebraic formality theorem to integrability of those structures.

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,

The paper is not ready for publication. The central framework rests on Lemma A.2, which says products commute with the completed projective tensor product. The lemma is false in general, and the contraderived category machinery in Section 1 uses it at many critical points. The second serious problem is Theorem 2.3.1. On Ω(R), the derivation ∂/∂t is a continuous degree-0 derivation that commutes with d, so it is a 1-cocycle in the continuous Hochschild complex. Since Ω(R) is graded commutative, no 0-cochain has non-zero Hochschild boundary, so this cocycle is not a coboundary. Thus HH^1_cont(Ω(R)) contains a class beyond the constants, contradicting the claimed quasi-isomorphism Ω(R)[-1] → HC_cont(Ω(R)). The theorem is genuinely false, not merely unproved.

That said, the paper is not without merit. The idea of defining continuous Hochschild cohomology via contraderived categories of complete locally convex dg-algebras is new and addresses a real need: a homological interpretation of continuous Hochschild complexes. The sheaf-theoretic globalization for smooth functions in §2.1 is clean, and the Dolbeault HKR statement in §2.2 is a plausible and interesting target. If the Dolbeault theorem could be established on a correct foundation, it would be a useful result.

The soft spots are in proportion. A large number of central statements are deferred to a follow-up paper: the full B∞ structure, the derived-category enhancement underlying Theorem 2.5.1, and the L∞ part of the twisted Dolbeault theorem. The proof of Theorem 2.2.5 relies on gauge transformations that are only sketched. The citation pattern is fine — Pflaum, Kontsevich, Positselski, CDH07 are the right inputs — but the author leans on them without providing the missing details.

For a reader in deformation theory, the framework is worth knowing about, but in its current form it cannot be used as reference. I would not cite it, and I would not send it to peer review until the foundation and the de Rham claim are sorted out.

Referee Report

3 major / 3 minor

Summary. The paper develops a contraderived category for complete locally convex dg-algebras, defines continuous Hochschild cohomology inside this framework, and claims formality theorems for three Fréchet dg-algebras: smooth functions, the Dolbeault algebra of a complex manifold, and the de Rham algebra of a smooth manifold. It also states computations for matrix factorisations and announces applications to generalized complex geometry and derived categories. The central novelty is the claim that continuous Hochschild cochains admit a natural homological interpretation and that HKR-type maps are quasi-isomorphisms in the continuous setting.

Significance. If the results were correct, the paper would give a useful framework for deformation theory of complete locally convex dg-algebras and would connect continuous Hochschild cohomology to generalized complex geometry. The paper has some genuinely useful features: it attempts to build on Positselski's contraderived categories, it gives an explicit local Koszul-type resolution in the Dolbeault case, and it identifies the continuous Hochschild complex as a B∞-algebra. However, the de Rham formality theorem is false by an explicit internal counterexample, and a number of the headline claims are only announced, not proved. The falsity of a main theorem is not a local presentation issue, so the paper cannot be accepted in its present form.

major comments (3)

[§2.3, Theorem 2.3.1] The de Rham formality statement is false. Take A=Ω(R), the de Rham algebra of the real line, and let D=∂/∂t. Then D is a continuous degree-0 derivation, so δD=0 in the Hochschild complex, and D commutes with the de Rham differential, so d_*D=[d,D]=0. Thus D is a closed Hochschild 1-cocycle. Since Ω(R) is graded commutative, for every 0-cochain c the Hochschild differential δc vanishes; d_*c is still a 0-cochain, so D is not a coboundary. Hence [D] is a nonzero class in HH^1_cont(Ω(R)). On the other hand, the cohomology of Ω(R)[-1] in degree 1 is trivial under the standard shift (H^2_dR(R)=0), and under the alternate convention it is only H^0_dR(R)=R. The class [D] cannot correspond to either. The one-sentence proof 'similar to Theorem 2.2.2 but easier' gives no argument that would exclude this cocycle.
[§2.5 and §1.5] Several load-bearing results are explicitly deferred to a follow-up paper. Theorem 1.5.5 proves only the dg-Lie structure and says the full B∞-structure 'will be dealt with in more general cases in the follow-up paper.' Section 2.5 states an invariance result for the continuous Hochschild complex of enhanced dg-categories and Theorem 2.5.1, but the proof is entirely in the follow-up paper. Since the abstract and introduction present these as results, the manuscript as written is an extended announcement rather than a proof of its main claims.
[§2.2, Lemma 2.2.4] The globalisation argument for the Dolbeault formality theorem relies on Lemma 2.2.4, whose proof is only a sketch. The lemma concerns a possibly unbounded complex of soft sheaves and uses two spectral sequences for the Cech totalization. No argument is given for convergence of these spectral sequences in the unbounded case, and the statement 'one can show' in the proof is doing essential work. Since Theorem 2.2.2 is the main geometric application, this needs a complete proof.

minor comments (3)

[§1.5, Definition 1.5.4] The definition of HC_cont(A,M) as Hom_Ae(Bar_red(A),M)[-1] is stated to be neither the product nor the sum totalization of the associated double complex. The intended subspace condition should be spelled out explicitly; as written, the reader must infer the topology from Corollary A.7.
[§1.2, Definition 1.2.1] The symbol Acabs is used before being defined through 'totalizations of short exact sequences'; the minimality and closure conditions are clear, but the notation is easy to confuse with Acctr.
[Throughout] There are several typos and missing accents: 'straightfoward', 'colagebra', 'comology' in the Pflaum reference, and inconsistent use of 'Fréchet' vs 'Frechet'. None of these affect the mathematics.

Circularity Check

0 steps flagged · score 2.0 of 10

No load-bearing circularity: the formality results rest on external input (Kontsevich, CDH07, Pflaum, Positselski) and on explicit local resolutions. The paper's main weaknesses are unsupported assumptions and deferred proofs, not self-referential derivation.

full rationale

I walked the derivation chain from the definition of Dctr(A) through the bar-resolution identification of HCcont(A,A), the local HKR/Koszul resolutions, the sheaf-theoretic globalization, and the L∞-extensions. The continuous Hochschild complex is not assumed to compute HH_cont by definition: Lemma 1.5.3 proves the identification with Hom_{Dctr(A^e)}(A[·],A) via the two-sided bar resolution, and the bar resolution itself is shown to be a graded-projective replacement. The local quasi-isomorphisms in Propositions 2.1.2–2.2.1 are obtained from explicit resolutions, not from the desired conclusion. Globalization uses Theorem 1.4.7 and Ogneva's projectivity result, again external and not circular. The L∞-formality statements are imported from Kontsevich [Kon03] and CDH07 [CDH07, Cor. 5.3]; these are independent external theorems and do not restate the paper's continuous-Hochschild claims. The comparison with Gualtieri's generalized-complex deformation complex is a standard identification of the Hodge pieces, not a fitted prediction. The only self-citation is [Ant24] in Remark 2.2.3, used for context about gerbe generalizations; it is not load-bearing. I also flag, as non-circular caveats, that Lemma A.2 (product–tensor commutation) is asserted with a reference to [Jar81] but is false in general and underpins Corollaries 1.1.2, 1.2.7 and hence Theorem 1.2.8; that Theorem 2.3.1 is dismissed with “the proof is similar to that of Theorem 2.2.2”; that the full B∞-structure in Theorem 1.5.5 is deferred; and that the invariance result used in Section 2.5 is deferred to a follow-up paper. These are mathematical gaps, not reductions of the claims to their own inputs, so they do not raise the circularity score beyond the minor self-citation level.

Assumptions & free parameters 0 free parameters · 4 assumptions · 1 invented entities

The central framework is not self-contained and is built on a false tensor-product commutation result. The remaining axioms are either imported external theorems or assumptions about the exact structure and descent that are not fully proven in the continuous setting.

assumptions (4)

ad hoc to paper Products commute with the completed projective tensor product (Lemma A.2).
False in general; supports Corollary 1.1.2 and most of Section 1.
domain assumption The split-exact structure and contraderived category behave as in Positselski's discrete theory when imported to clct modules.
Used throughout Section 1; Theorem 1.2.8 is obtained by citing [Pos25] rather than a self-contained proof.
standard math Kontsevich's L∞ formality for polydifferential operators and Pflaum's continuous Hochschild quasi-isomorphism for C∞(M) are correct and applicable.
Used in Theorem 2.1.3 for the L∞ extension.
domain assumption Strong descent holds for sheaves of clct algebras with fine 0-component (Lemma 1.4.4).
Used for sheaf-theoretic globalization in Section 2.

invented entities (1)

Contraderived category of complete locally convex dg-algebras D_ctr(A)
purpose: Homological setting for deformation theory of curved topological dg-algebras.
New categorical framework; no empirical handle; validity depends on the false Lemma A.2.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Continuous Hochschild Cohomology and Formality." pith.science (2026). https://pith.science/paper/SFOTWZMS

@misc{pith2026251221090,
  author       = {Pith},
  title        = {Pith review of: Continuous Hochschild Cohomology and Formality},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SFOTWZMS}},
  note         = {Machine review of arXiv:2512.21090}
}

abstract

We define the appropriate homological setting to study deformation theory of complete locally convex (curved) dg-algebras based on Positselski's contraderived categories. We define the corresponding Hochschild complex controlling deformations and prove formality theorems for the Fr\'echet algebras of smooth functions on a manifold, the de Rham algebra and for the Dolbeault algebra of a complex manifold. In the latter case, the Hochschild cohomology is equivalent to Kontsevich's extended deformation complex, the Hochschild cohomology of the derived category in case $X$ is a smooth projective variety and to Gualtieri's deformation complex of $X$ viewed as generalized complex manifold. We also compute the continuous Hochschild cohomology for various categories of matrix factorisations.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

11 extracted references · 5 canonical work pages

[10]

[PP12] Alexander Polishchuk and Leonid Positselski

URL:https://arxiv.org/abs/2110.08237,arXiv: 2110.08237. [PP12] Alexander Polishchuk and Leonid Positselski. Hochschild (co)homology of the second kind i.Transactions of the American Mathematical Soci- ety, 364(10):5311–5368, October

arXiv
[85]

Maurer–cartan mod- uli and theorems of riemann–hilbert type.Applied Categorical Structures, 29(4):685–728, 2021.doi:10.1007/s10485-021-09631-3

[CHL21] Joseph Chuang, Julian Holstein, and Andrey Lazarev. Maurer–cartan mod- uli and theorems of riemann–hilbert type.Applied Categorical Structures, 29(4):685–728, 2021.doi:10.1007/s10485-021-09631-3. [GHL25] Ai Guan, Julian Holstein, and Andrey Lazarev. Hochschild cohomology of the second kind: Koszul duality and morita invariance,

work page doi:10.1007/s10485-021-09631-3 2021
[1993]

org/doc/90218

URL:http://eudml. org/doc/90218. [BR23] Alexey Bondal and Alexei Rosly. Coherent sheaves, chern classes, and super- connections on compact complex-analytic manifolds.Izvestiya: Mathematics, 87:439–468, 09 2023.doi:10.4213/im9386e. [CDH07] Damien Calaque, Vasiliy Dolgushev, and Gilles Halbout. Formality theorems for hochschild chains in the lie algebroid s...

work page doi:10.4213/im9386e 2023
[2004]

org/abs/math/0401221,arXiv:math/0401221

URL:https://arxiv. org/abs/math/0401221,arXiv:math/0401221. [Gua21] Ai Guan.Koszul Duality and Deformation Theory. PhD thesis, Lancaster University, 2021.doi:10.17635/lancaster/thesis/1362. [Jar81] HansJarchow.The Projective Tensor Product, pages323–342. Vieweg+Teubner Verlag, Wiesbaden, 1981.doi:10.1007/978-3-322-90559-8_15. [Kod05] Kunihiko Kodaira.Theo...

arXiv 2021
[2007]

1515/CRELLE.2007.085[cited 2025-09-30],doi:doi:10.1515/CRELLE.2007

URL:https://doi.org/10. 1515/CRELLE.2007.085[cited 2025-09-30],doi:doi:10.1515/CRELLE.2007

work page doi:10.1515/crelle.2007 2007
[2009]

[BN93] Marcel Bökstedt and Amnon Neeman

URL:https://arxiv.org/abs/math/0509284, arXiv:math/0509284. [BN93] Marcel Bökstedt and Amnon Neeman. Homotopy limits in triangulated cate- gories.Compositio Mathematica, 86(2):209–234,

arXiv
[2012]

URL:http://dx.doi.org/10.1090/ S0002-9947-2012-05667-4,doi:10.1090/s0002-9947-2012-05667-4. 30

work page doi:10.1090/s0002-9947-2012-05667-4 2012
[2014]

[Pfl98] Markus J

URL:https://arxiv.org/abs/1405.4094, arXiv:1405.4094. [Pfl98] Markus J. Pflaum. On continuous hochschild comology and cohomology groups.Letters in Mathematical Physics, 44(1):43–51, 1998.doi:10.1023/A: 1007409019084. [Pos09] Leonid Positselski. Two kinds of derived categories, koszul duality, and comodule-contramodule correspondence.Memoirs of the America...

arXiv 1998

Show all 11 references

[2021]

[Gua04] Marco Gualtieri

URL:http://dx.doi.org/10.4171/jncg/ 438,doi:10.4171/jncg/438. [Gua04] Marco Gualtieri. Generalized complex geometry,

doi
[2024]

[Blo09] Jonathan Block

URL:https://arxiv.org/abs/2411.19595, arXiv:2411.19595. [Blo09] Jonathan Block. Duality and equivalence of module categories in noncom- mutative geometry i,

arXiv
[2025]

[GL21] Ai Guan and Andrey Lazarev

URL:https: //arxiv.org/abs/2312.16645,arXiv:2312.16645. [GL21] Ai Guan and Andrey Lazarev. Koszul duality for compactly generated derived categories of second kind.Journal of Noncommutative Geometry, 15(4):1355–1371, December

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.

[1] [10]

[PP12] Alexander Polishchuk and Leonid Positselski

URL:https://arxiv.org/abs/2110.08237,arXiv: 2110.08237. [PP12] Alexander Polishchuk and Leonid Positselski. Hochschild (co)homology of the second kind i.Transactions of the American Mathematical Soci- ety, 364(10):5311–5368, October

arXiv

[2] [85]

Maurer–cartan mod- uli and theorems of riemann–hilbert type.Applied Categorical Structures, 29(4):685–728, 2021.doi:10.1007/s10485-021-09631-3

[CHL21] Joseph Chuang, Julian Holstein, and Andrey Lazarev. Maurer–cartan mod- uli and theorems of riemann–hilbert type.Applied Categorical Structures, 29(4):685–728, 2021.doi:10.1007/s10485-021-09631-3. [GHL25] Ai Guan, Julian Holstein, and Andrey Lazarev. Hochschild cohomology of the second kind: Koszul duality and morita invariance,

work page doi:10.1007/s10485-021-09631-3 2021

[3] [1993]

org/doc/90218

URL:http://eudml. org/doc/90218. [BR23] Alexey Bondal and Alexei Rosly. Coherent sheaves, chern classes, and super- connections on compact complex-analytic manifolds.Izvestiya: Mathematics, 87:439–468, 09 2023.doi:10.4213/im9386e. [CDH07] Damien Calaque, Vasiliy Dolgushev, and Gilles Halbout. Formality theorems for hochschild chains in the lie algebroid s...

work page doi:10.4213/im9386e 2023

[4] [2004]

org/abs/math/0401221,arXiv:math/0401221

URL:https://arxiv. org/abs/math/0401221,arXiv:math/0401221. [Gua21] Ai Guan.Koszul Duality and Deformation Theory. PhD thesis, Lancaster University, 2021.doi:10.17635/lancaster/thesis/1362. [Jar81] HansJarchow.The Projective Tensor Product, pages323–342. Vieweg+Teubner Verlag, Wiesbaden, 1981.doi:10.1007/978-3-322-90559-8_15. [Kod05] Kunihiko Kodaira.Theo...

arXiv 2021

[5] [2007]

1515/CRELLE.2007.085[cited 2025-09-30],doi:doi:10.1515/CRELLE.2007

URL:https://doi.org/10. 1515/CRELLE.2007.085[cited 2025-09-30],doi:doi:10.1515/CRELLE.2007

work page doi:10.1515/crelle.2007 2007

[6] [2009]

[BN93] Marcel Bökstedt and Amnon Neeman

URL:https://arxiv.org/abs/math/0509284, arXiv:math/0509284. [BN93] Marcel Bökstedt and Amnon Neeman. Homotopy limits in triangulated cate- gories.Compositio Mathematica, 86(2):209–234,

arXiv

[7] [2012]

URL:http://dx.doi.org/10.1090/ S0002-9947-2012-05667-4,doi:10.1090/s0002-9947-2012-05667-4. 30

work page doi:10.1090/s0002-9947-2012-05667-4 2012

[8] [2014]

[Pfl98] Markus J

URL:https://arxiv.org/abs/1405.4094, arXiv:1405.4094. [Pfl98] Markus J. Pflaum. On continuous hochschild comology and cohomology groups.Letters in Mathematical Physics, 44(1):43–51, 1998.doi:10.1023/A: 1007409019084. [Pos09] Leonid Positselski. Two kinds of derived categories, koszul duality, and comodule-contramodule correspondence.Memoirs of the America...

arXiv 1998