REVIEW 3 major objections 2 minor 2 cited by
Every quadratic monomial algebra is absolutely Koszul
Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →
T0 review · glm-5.2
2026-07-05 04:46 UTC pith:5YOYIHWX
load-bearing objection Useful synthesis for quadratic monomial algebras; the absolute Koszulness claim is the load-bearing piece that needs proof verification the 3 major comments →
Koszul Duality for Quadratic Monomial Algebras
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
Core claim
The load-bearing discovery is that the Koszul dual of a finite-dimensional quadratic monomial algebra is both graded coherent and graded co-coherent. Coherence alone says that kernels of maps between finite modules stay finite; co-coherence is the dual condition, saying that cokernels of maps between finite comodules (or dually, syzygies in the contravariant direction) also stay finite. This pairing is what makes the tails and cotails categories simultaneously well-behaved, and it is the property from which the refined duality correspondences and finitistic dimension bounds flow.
What carries the argument
Koszul dual algebra (quadratic dual), graded coherence and co-coherence, tails and cotails categories, linear and colinear modules, absolute Koszulness, linearity defect, nonstandard t-structures, finitistic dimension
Load-bearing premise
The claim that the Koszul dual Λ! is graded co-coherent — the dual-direction finiteness property — is the structurally critical premise. Coherence follows relatively standardly for monomial algebras, but co-coherence requires that syzygy-type operations in the contravariant direction preserve finiteness, which does not follow automatically from the monomial structure. If this verification is incomplete, the cotails category results and parts of the singular duality refinement
What would settle it
Construct a finite-dimensional quadratic monomial algebra Λ whose Koszul dual Λ! admits a finitely copresented module that is not coperfect (i.e., does not admit a finite resolution by finite-dimensional projectives in the dual direction). This would directly contradict the co-coherence claim and collapse the cotails hereditary abelian result.
If this is right
- The hereditary abelian structure of tails and cotails categories for monomial algebras provides a concrete setting for studying noncommutative projective geometry of these algebras.
- The coincidence of finitely presented with perfect modules (and copresented with coperfect) simplifies the homological algebra of Λ! substantially, potentially making computational approaches to resolutions tractable.
- The refined singular Koszul duality for monomial algebras may extend to broader classes of algebras if the co-coherence property can be verified beyond the monomial setting.
- The finitistic dimension bounds in terms of finite paths in Λ! give a combinatorial route to computing or estimating finitistic dimensions, connecting algebraic homological invariants to graph-theoretic data.
Where Pith is reading between the lines
- If the co-coherence of Λ! can be characterized combinatorially (e.g., via the structure of the quiver and monomial relations defining Λ), this would give a graph-theoretic criterion for when the cotails category is hereditary, extending the result's reach.
- The rationality of Poincaré and Hilbert series for finitely presented modules, combined with the path-based finitistic dimension bounds, suggests that the homological invariants of quadratic monomial algebras are governed by finite combinatorial data in the Koszul dual — a pattern that, if it extends to non-monomial quadratic algebras, would bridge combinatorics and representation theory more broa
- The explicit nonstandard t-structures on the derived categories may provide new bridging functors analogous to the classical BGG correspondence, potentially yielding new equivalences between derived categories of modules over Λ and coherent sheaves on noncommutative spaces associated to Λ!.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript studies the Koszul dual $Λ^!$ of a finite-dimensional quadratic monomial algebra $Λ$. The main structural results are: (1) $Λ^!$ is graded coherent and graded co-coherent; (2) finitely presented (resp. copresented) graded $Λ^!$-modules coincide with perfect (resp. coperfect) modules; (3) the tails and cotails categories are hereditary abelian with explicit descriptions; (4) every finite-dimensional quadratic monomial algebra is absolutely Koszul with global linearity defect at most one; (5) rationality of Poincaré and Hilbert series for finitely presented graded modules. These results are applied to refine derived and singular Koszul dualities, the BGG correspondence, and to obtain finitistic dimension bounds. This review is based on the abstract and the structural framework described therein; the full text was not available for detailed proof verification.
Significance. The paper addresses a natural and well-motivated problem in the structure theory of Koszul algebras. Quadratic monomial algebras form a class where combinatorial methods (Anick/Bardzell resolutions) give strong control, and the systematic extraction of coherence, co-coherence, absolute Koszulness, and the resulting categorical consequences in a single framework is a valuable contribution if the proofs hold. The falsifiable predictions (rationality of Poincaré series, explicit hereditary abelian structure on tails/cotails, finitistic dimension bounds in terms of finite paths in $Λ^!$) are concrete and checkable. The BGG correspondence refinement and the nonstandard $t$-structure descriptions are potentially useful for researchers in representation theory and homological algebra.
major comments (3)
- The claim that every finite-dimensional quadratic monomial algebra is absolutely Koszul is the most load-bearing assertion in the paper. Absolute Koszulness requires that every finitely generated graded module has a linear free resolution after some shift, which is strictly stronger than Koszulness (linear resolution of simples). While the Anick/Bardzell resolution provides combinatorial control over syzygies, the verification that syzygies remain linear for all finitely generated graded modules — not just simples or modules generated in a single degree — requires explicit argument. The abstract states this as a theorem; the proof must be carefully checked to confirm that the linearity of syzygies is established for arbitrary finitely generated graded modules, not just for a restricted class. If this verification has gaps, the downstream claims about rational Poincaré series, refined Kos
- The graded co-coherence of $Λ^!$ and the claim that finitely copresented modules coincide with coperfect modules are structurally critical for the cotails category results and the singular Koszul duality refinement. Co-coherence is a duality-sensitive property. The argument that $Λ^!$ is also a quadratic monomial algebra (so co-coherence reduces to coherence of $(Λ^!)^{op}$) is plausible, but the abstract should clarify whether this reduction is explicit in the text and whether the coperfect/finitely copresented coincidence is proved directly or derived from the coherence of the opposite algebra. The cotails hereditary abelian claim depends on this verification.
- The relationship between absolute Koszulness (which implies linearity defect 0 for graded modules) and the stated global linearity defect at most one needs clarification. If absolute Koszulness holds for all finitely generated graded modules, the linearity defect should be 0 in the graded setting. The 'at most one' claim may refer to the ungraded setting, which would require separate justification. The abstract should distinguish these cases explicitly.
minor comments (2)
- The abstract is dense and packs many results; a brief indication of the logical dependency graph (which results depend on which) would help readers navigate the paper.
- The finitistic dimension bounds are mentioned as an application but the abstract does not specify their sharpness or how they compare to existing bounds in the literature. A sentence of context would strengthen the presentation.
Simulated Author's Rebuttal
We thank the referee for a careful reading of our abstract and for identifying several points where clarification is needed. We address each major comment below. We note at the outset that this review is based on the abstract alone; the full text contains detailed proofs of all claims, which we are confident will resolve the referee's concerns upon examination.
read point-by-point responses
-
Referee: The claim that every finite-dimensional quadratic monomial algebra is absolutely Koszul is the most load-bearing assertion... verification that syzygies remain linear for all finitely generated graded modules — not just simples or modules generated in a single degree — requires explicit argument.
Authors: The referee is correct that absolute Koszulness requires linear free resolutions for all finitely generated graded modules, not just simples. The full text establishes this in detail. The key argument proceeds as follows. First, we prove that the Koszul dual of any quadratic monomial algebra is graded coherent, and that finitely presented graded modules coincide with perfect modules. This coherence result, combined with the explicit combinatorial structure of the Anick/Bardzell resolution, allows us to verify linearity of syzygies for arbitrary finitely generated graded modules. The crucial point is that any finitely generated graded module admits a presentation by finitely generated free modules, and coherence ensures this presentation theory is well-behaved. The linearity of the resolution then follows from the monomial structure: the syzygy modules at each stage are themselves finitely presented and admit descriptions in terms of monomial ideals, whose resolutions are linear by the combinatorial structure. We are happy to add a summary of this argument to the introduction to make the logical flow clearer. revision: partial
-
Referee: The graded co-coherence of Λ! and the claim that finitely copresented modules coincide with coperfect modules are structurally critical... The abstract should clarify whether this reduction is explicit in the text and whether the coperfect/finitely copresented coincidence is proved directly or derived from the coherence of the opposite algebra.
Authors: The referee's suggestion is well-taken. The reduction is indeed explicit in the text. Since the Koszul dual of a quadratic monomial algebra is again a quadratic monomial algebra, and the opposite algebra of a quadratic monomial algebra is again quadratic monomial, the graded coherence of (Λ!)^op follows from the same argument as for Λ!. The coperfect/finitely copresented coincidence is then derived from this by duality. We agree that the abstract should make this logical dependency clearer and will revise it to indicate that co-coherence is established via the opposite algebra. revision: yes
-
Referee: The relationship between absolute Koszulness (which implies linearity defect 0 for graded modules) and the stated global linearity defect at most one needs clarification.
Authors: The referee raises a valid point about the relationship between these two properties. We clarify: absolute Koszulness in the graded setting indeed implies that the graded linearity defect is 0. The statement about global linearity defect at most one refers to the ungraded setting, where one considers all (not necessarily graded) finitely generated modules. The passage from graded to ungraded requires separate argument, which is given in the full text. The key point is that an ungraded finitely generated module can be filtered by graded pieces, and the graded linearity defect bound of 0 lifts to a bound of at most 1 in the ungraded setting due to the filtration. We agree that the abstract should distinguish these two cases explicitly and will revise accordingly. revision: yes
- The referee's concerns are substantive and well-motivated, but they arise from the fact that only the abstract was available for review. The full text contains complete proofs addressing all three major comments. We cannot fully resolve the referee's verification requests through the rebuttal alone; the referee would need to examine the full proofs. We are confident the arguments hold, but we acknowledge that proof verification has not yet occurred.
Circularity Check
No circularity detected; derivation chain is self-contained from standard constructions
full rationale
This is an abstract-only review, so only the high-level derivation chain is visible. The claims proceed in a standard mathematical order: the Koszul dual Λ! is a standard construction (not defined in terms of the target results), and the properties claimed — coherence, co-coherence, absolute Koszulness, hereditary abelianness of tails/cotails — are external mathematical properties requiring independent proof, not fitted parameters or definitions smuggled through self-citation. The finitistic dimension bounds are stated in terms of finite paths in Λ!, an independent combinatorial invariant. No self-citation chain, no fitted-input-renamed-as-prediction, and no self-definitional reduction is visible in the abstract. The derivation appears to be a genuine forward implication chain from the monomial structure to downstream consequences. Without the full text, one cannot rule out a hidden self-citation in the proofs, but nothing in the abstract exhibits circularity.
Axiom & Free-Parameter Ledger
axioms (4)
- standard math Koszul duality framework for quadratic algebras (standard background)
- standard math Definitions and basic properties of graded coherence and co-coherence
- standard math Bernstein-Gelfand-Gelfand correspondence and its standard formulation
- domain assumption Singular Koszul duality framework
Cite this review
Pith. "Pith review of Koszul Duality for Quadratic Monomial Algebras." pith.science (2026). https://pith.science/paper/5YOYIHWX
@misc{pith2026260420177,
author = {Pith},
title = {Pith review of: Koszul Duality for Quadratic Monomial Algebras},
year = {2026},
howpublished = {\url{https://pith.science/paper/5YOYIHWX}},
note = {Machine review of arXiv:2604.20177}
}
read the original abstract
Let \(\Lambda\) be a finite-dimensional quadratic monomial algebra and let \(\Lambda^{!}\) be its Koszul dual. We investigate the structure of graded modules over \(\Lambda^{!}\) and derive several consequences for Koszul duality. We prove that \(\Lambda^{!}\) is both graded coherent and graded co-coherent. Moreover, finitely presented and finitely copresented graded \(\Lambda^{!}\)-modules coincide with perfect and coperfect modules, respectively. As a consequence, the associated tails and cotails categories are hereditary abelian categories admitting explicit descriptions in terms of linear and colinear modules. We further show that every finite-dimensional quadratic monomial algebra is absolutely Koszul and has global linearity defect at most one. In particular, finitely presented graded modules have rational Poincar\'e and Hilbert series. Using these structural results, we refine graded and ungraded derived Koszul dualities, singular Koszul dualities, and the Bernstein--Gelfand--Gelfand correspondence. We obtain explicit descriptions of the associated triangulated categories and of the induced nonstandard \(t\)-structures. As an application, we derive new bounds on the finitistic dimension of quadratic monomial algebras in terms of finite paths in the Koszul dual algebra.
Forward citations
Cited by 2 Pith papers
-
Koszul Duality for Coherent Sheaves
Claims a bounded Koszul duality for infinite-dimensional Koszul algebras and a BGG-type description of D^b(coh(X)), but key identifications and hypotheses are unproved.
-
Koszul Duality for Coherent Sheaves
For a Koszul quotient Λ of a polynomial ring, the derived category of coherent sheaves on Proj(Λ) is claimed to be equivalent to a quotient of derived categories over the Koszul dual Λ!.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.