Pith. sign in

REVIEW 3 major objections 5 minor 26 references

Factoriality of normal projective varieties

T0 review · 3 major / 5 minor · reviewed 2026-08-03 · deepseek-v4-flash

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

Pith's one-line read This paper proves that on a normal projective variety with 2-semi-rational singularities, the Q-factoriality defect equals h^{2n-2}(X) − h^2(X), and that Q-factoriality implies factoriality for local complete intersections whose singular lo

desk verdict Genuine improvement of Park–Popa's Q-factoriality formula, but the key diagram commutativity in Section 4 is asserted rather than fully proved — deserves expert referee scrutiny. read the letter →

arxiv 2601.13151 v6 pith:CAIMPAB3 submitted 2026-01-19 math.AG

Seung-Jo Jung , Morihiko Saito This is my paper · ORCID
classification math.AG MSC 14C2014B0514E1514F4314M06
keywords Q-factorialityfactorialityWeildivisorsCartiersemi-rationalsingularitiesintersectioncohomologylocalcompleteintersectionsBettinumbers
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 targets a classical question: on a singular variety, when is every Weil divisor numerically indistinguishable from a Cartier divisor? The main theorem states that for a normal projective variety X of dimension n≥2 whose singularities are 2-semi-rational, the Q-factoriality defect σ(X)—the rank of Div(X)/CDiv(X)—equals h^{2n-2}(X) − h^2(X). This upgrades a previously known inequality to an equality under a milder hypothesis than rationality. Along the way the authors prove that for local complete intersections with singular locus of codimension at least three, Q-factoriality already forces factoriality, and they use this to recover and slightly extend a classical projective factoriality criterion. The payoff: for a large class of singular varieties, factoriality becomes a computable topological condition.

What carries the argument

The load-bearing identity is the isomorphism (4), identifying Div(X)/CDiv(X) with the image of the Chern-class restriction map H^{1,1}(Ẽ,Z) → Γ(Σ, R^2π'_*Z_E / Σ_i Z[E_i]_π'). The proof then runs on the long exact sequence (3.5) connecting ordinary cohomology, intersection cohomology, and the sheaves L^k = H^k(IC_XQ[-n]). Proposition 3—stalkwise purity of type (1,1) for L^2 under 2-semi-rationality—is what lets the source H^{1,1}(Ẽ,Z) be replaced by H^2(Ẽ,Z), making the rank topological. For the factoriality half, the key mechanism is a dual-perversity argument: for local complete intersections Z_X[n] lies in the perverse t-structure, and a cyclic-covering argument shows an effective Q-Carti

What would settle it

A 2-semi-rational normal projective variety X with h^{2n-2}(X)=h^2(X) but σ(X)>0 would refute Theorem 1; the paper's own cone examples show the hypothesis is sharp, so any such counterexample must genuinely satisfy 2-semi-rationality. A more surgical test is to compute the maps γ_2 and β for a concrete such variety and check whether their images in H^2(π^{-1}(x),Z) coincide.

Watch

Extended reading notes

Core claim

The central claim is Theorem 1: for a normal projective variety X of dimension n≥2 satisfying R^k π_* O_Ẽ = 0 for k=1,2 (2-semi-rationality), σ(X)=h^{2n-2}(X)-h^2(X). The proof realizes Div(X)/CDiv(X) as the image of a restriction map from H^{1,1}(Ẽ,Z) to a constructible sheaf supported on the singular locus. The 2-semi-rationality hypothesis implies the relevant Hodge structures have type (1,1), so the image is unchanged when H^{1,1} is replaced by H^2. A second thread shows that under the perversity conditions satisfied by local complete intersections, Q-factoriality is equivalent to factoriality when the singular locus has codimension at least three; combining these threads yields factori

Load-bearing premise

The proof that the topological rank equals σ(X) requires that the morphism γ_2 in the long exact sequence (3.5) is induced by the restriction map β in (4); the paper flags this as 'one technical problem' and resolves it by the commutativity of diagram (4.1). If that diagram does not commute, the equality would not be established.

Editorial extensions

If this is right

  • For any 2-semi-rational normal projective variety, Q-factoriality is equivalent to the topological equality h^{2n-2}(X)=h^2(X).
  • The difference h^{2n-2}(X)-h^2(X) is always a lower bound for the defect, so σ(X)=0 forces the two Betti numbers to match.
  • For local complete intersections with codim SingX ≥3, factoriality is the same as Q-factoriality, so the Betti-number equality characterizes factoriality there.
  • The projective case of the classical factoriality criterion is recovered in a slightly stronger form: it suffices that the non-rational and Q-homology singular loci have codimension at least four, without requiring the full singular locus to have codimension four.
  • For projective cones over complete intersections, the defect of the cone equals the defect of the base, so factoriality of the cone reduces to a Betti-number equality on the base.

Reading between the lines

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

  • Because 2-semi-rationality is substantially weaker than rationality, the topological formula likely applies to many singularities of Calabi-Yau type, where h^2 can be nonzero; one could test it on explicit examples such as cones over surfaces of general type.
  • The proof's dependence on diagram (4.1) suggests a potential hidden obstruction: if the non-canonical decomposition of the decomposition theorem can be twisted so that γ_2 and β differ, the equality might fail for some 2-semi-rational variety; checking this commutativity in a concrete example would either certify or break the result.
  • The factoriality/Q-factoriality equivalence for lci singularities of codimension three indicates that the only way a threefold lci with isolated singularities fails to be factorial is through the integer h^4-h^2; one could compute this defect for known examples of non-factorial threefolds.
  • The cone reduction might extend to arbitrary projective cones or to quotient singularities, where the defect could be expressed in terms of the base's cohomology and the group action.
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 / 5 minor

Summary. This paper studies the Q-factoriality defect σ(X) of a normal projective variety X. The main result (Theorem 1, §4) states that if X has 2-semi-rational singularities, i.e. R^kπ_*O_Ẽ=0 for k=1,2 for a desingularization π, then σ(X)=h^{2n-2}(X)-h^2(X). This is presented as a strengthening of a theorem of Park–Popa. The proof combines a description of Div(X)/CDiv(X) as the image of a restriction map β (Proposition 1), mixed-Hodge-module computations of the intersection-cohomology pieces L^1,L^2 (Propositions 2 and 3), and a long exact sequence (3.5) involving Γ(Σ,L^2); the equality is obtained by identifying γ_2 with β. Section 6 proves a criterion (Theorem 6.1) under which effective Q-Cartier divisors are Cartier, and the paper derives several corollaries for local complete intersections, including a projective analogue of Grothendieck's factoriality theorem.

Significance. If correct, Theorem 1 gives a purely topological formula for the Q-factoriality defect under a substantially weaker hypothesis than rational singularities, improving the earlier Park–Popa result. The paper also provides a useful generalization of Grothendieck's factoriality criterion for local complete intersections. The arguments rely on deep external machinery — mixed Hodge modules, the decomposition theorem, Mumford's intersection-matrix criterion — and not on machine-checked proofs, so the assessment hinges on the completeness and correctness of the written Hodge-module arguments. The main strengths are the explicit weakening of hypotheses and the concrete applications to hypersurfaces and complete intersections.

major comments (3)
  1. [§4, diagram (4.1), eqs. (3.5), (4)] The proof of Theorem 1 requires that the morphism γ_2 in the long exact sequence (3.5) coincide with the restriction map β in (4). The paper explicitly flags this as 'one technical problem' and then asserts that the needed commutativity is 'assured by the octahedral axiom.' However, the vertical natural transformations in diagram (4.1) are not defined, and no proof of commutativity is supplied. Since the decomposition H^2(Ẽ,Q) ≅ IH^2 ⊕ H^2_2 is non-canonical, if (4.1) commutes only after factoring through the Σ-supported summand, the rank computed from (3.5) would be the rank of a different subgroup of H^2(Ẽ,Z), and equality (6) would not follow. This is a load-bearing gap; a complete proof of the commutativity of (4.1) is needed.
  2. [§3, Proposition 3] Proposition 3 is essential: it provides the Hodge-theoretic input used in §4 to conclude H^k(Σ,C^•)=0 for k≥2n−4, which is needed to pass from the exact sequence (3.5) to the Betti-number equality. The proof for L^1 is only sketched via a general hyperplane section and a Zariski-locally closed subset argument, and for L^2 the text simply says 'The argument is similar for L^2'. Since the paper's main theorem depends on the assertion that each stalk L^2_x has type (1,1), this part needs to be written out in enough detail to allow verification, or the reduction must be justified explicitly.
  3. [§4, first paragraph after eq. (3.5)] The sentence 'since the latter implies that H^k(Σ,C^•)=0 for k≥2n−4' is stated without proof. Proposition 3 concerns the stalks of L^1 and L^2, but the passage from stalk-wise type (1,1) and vanishing to the vanishing of H^k(Σ,C^•) in the stated range is not demonstrated. Since this vanishing is used directly for the rank computation, a precise derivation should be included.
minor comments (5)
  1. [Abstract / Introduction] There is a minor inconsistency in the dimension assumption: the abstract says 'n:=dim X > 2' while the introduction states n≥2. Please harmonize.
  2. [§3, diagram (3.1)] The 'functoriality' of cubic hyperresolutions for the morphism Ẽ→X is invoked without a precise reference or construction. Please specify the source or give the construction.
  3. [§4, eq. (6)] The notation h^{2,0}(Y) in Remark 4.4 is not defined in the paper, which uses h^k for Betti numbers. Please define Hodge numbers where used.
  4. [§5, proof of Corollary 4] In the proof of the implication (a)⇒(c), the surjectivity of Pic(P^N)→Pic(X) used in diagram (5.5) is not fully justified; a short explanation would improve readability.
  5. [§6, Theorem 6.1] The proof of Theorem 6.1 is very compressed, especially the passage from (6.5)–(6.7) to (6.8) and the conclusion that the normalized cyclic cover is étale. Even if the steps are standard for specialists, expanding them or giving precise references would make the argument verifiable.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the Q-factoriality defect is defined independently and the topological formula is derived, not assumed; the acknowledged diagram-commutativity issue is a proof gap, not a circular reduction.

full rationale

The central claim, Theorem 1, derives the equality σ(X) = h^{2n-2}(X) − h^2(X) from a genuinely independent chain: σ(X) is defined as the rank of Div(X)/CDiv(X) (a divisor-group invariant), while the right-hand side is a Betti-number expression. The proof uses Proposition 1 to identify Div(X)/CDiv(X) with an image of a restriction map β, and then uses external Hodge-module machinery (mixed Hodge modules, the decomposition theorem, weight spectral sequences) to compute the rank of that image. The 2-semi-rationality hypothesis R^kπ_*O_X~ = 0 for k = 1,2 is an input condition, not a restatement of the conclusion. The paper explicitly flags one technical step: 'one technical problem is to show that the morphism γ_2 in (3.5) for k=2 is induced by β in (4) (since the decomposition in the decomposition theorem is non-canonical)', and it asserts this via diagram (4.1) and the octahedral axiom. This is a possible gap or correctness concern, but it is not circular: the asserted commutativity is not assumed as a definition of γ_2 or β, and no fitted parameter is renamed as a prediction. The self-citations, mainly to Saito's foundational papers on Hodge modules and mixed Hodge modules, are used as external theorems with stated assumptions that do not include the target equality; they are standard, externally validated tools rather than self-referential support for σ(X) = h^{2n-2} − h^2. No step reduces a predicted quantity to an input by construction. Therefore the paper exhibits no significant circularity; at most it has an acknowledged technical gap that should be assessed as a correctness risk, not as circularity.

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

The paper relies on standard foundational results in Hodge theory, intersection cohomology, and desingularization. No numerical fitting is involved; the only numbers are Betti numbers h^k(X), which are inputs, not fitted parameters. No new physical or mathematical entities are postulated.

assumptions (5)
  • standard math Existence of a desingularization π: Ẽ→X whose exceptional divisor E is simple normal crossings.
    Used at the start to define E_i and set up Propositions 1–3.
  • standard math Decomposition theorem for polarizable mixed Hodge modules (Saito [Sa 90]).
    Used throughout to split π_*Q_h[n] and to show (3.3)/(3.4) and Corollary 1.
  • standard math Mumford's negative definiteness of the intersection matrix of exceptional divisors.
    Used in Section 2 to prove Proposition 2 and Theorem 1 for n=2; Remark 4.1 notes [Mu 61] gives only an outline.
  • standard math Stability of middle and dual middle perversities under nearby/vanishing cycles (Schürmann).
    Used in Proposition 6.1 to prove the lci case of the perversity condition (6.2).
  • standard math Cohomological purity / universal coefficient theorem for Z-complexes (BBD 3.3.2).
    Used in Section 6 to derive (6.5)-(6.8).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Factoriality of normal projective varieties." pith.science (2026). https://pith.science/paper/CAIMPAB3

@misc{pith2026260113151,
  author       = {Pith},
  title        = {Pith review of: Factoriality of normal projective varieties},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CAIMPAB3}},
  note         = {Machine review of arXiv:2601.13151}
}
abstract

For a normal projective variety $X$, the $\mathbb Q$-factoriality defect $\sigma(X)$ is defined to be the rank of the quotient of the group of Weil divisors by the subgroup of Cartier ones. We prove an improvement of a topological formula of S.G. Park and M. Popa asserting that $\sigma(X)\le h^{2n-2}(X)-h^2(X)$ by assuming only 1-semi-rational singularities instead of rational singularities, and the equality holds in the 2-semi-rational case. Here the singularities are called $j$-semi-rational if $R^k\pi_*{\mathcal O}_{\widetilde{X}}=0$ for any $k\in[1,j]$ with $\pi:\widetilde{X}\to X$ a desingularization, $h^k(X):=\dim H^k(X,{\mathbb Q})$, and $n:=\dim X\ge2$. We also show (a slight generalization of) the assertion that $\mathbb Q$-factoriality implies factoriality if $X$ is a local complete intersection whose singular locus has at least codimension three. We then get a new proof for the projective case of Grothendieck's theorem asserting that $X$ is factorial if it is a local complete intersection whose singular locus has at least codimension four. We also show that a local complete intersection $X$ of dimension 3 having only isolated singularities is factorial if the (topological) defect ${\rm def}(X):=h^4(X)-h^2(X)$ vanishes, without any assumption on rational singularities.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

26 extracted references · 2 linked inside Pith

  1. [1]

    Beilinson, A., Bernstein, J., Deligne, P., Faisceaux pervers, Ast\'erisque 100 (1982)

    1982

  2. [2]

    2014, 21--50

    Brevik, J., Nollet, S., Developments in Noether-Lefschetz theory, Contemp.\ Math., 608, A.M.S. 2014, 21--50

    2014

  3. [3]

    osung der rationalen Singularit\

    Brieskorn, E., Die Aufl\"osung der rationalen Singularit\"aten holomorpher Abbildungen, Math.\ Ann.\ 178 (1968), 255--270

    1968

  4. [4]

    Alg.\ Geom.\ 19 (2010), 781--791

    Cheltsov, I., Factorial threefold hypersurfaces, J. Alg.\ Geom.\ 19 (2010), 781--791

    2010

  5. [5]

    40 (1971), 5--57

    Deligne, P., Th\'eorie de Hodge II, Publ.\ Math.\ I.H.E.S. 40 (1971), 5--57

    1971

  6. [6]

    Fulton, W., Intersection Theory, Springer, New York, 1984

    1984

  7. [7]

    Goresky, M., MacPherson, R., Intersection homology II, Inv. Math. 72 (1983), 77--129

    1983

  8. [8]

    Grauert, H., \"Uber Modifikationen und exzeptionelle analytische Mengen, Math.\ Ann.\ 146 (1962), 331--368

    1962

Show all 26 references
  1. [9]

    atze f\"ur analytische Kohomologiegruppen auf komplexen R\

    Grauert, H., Riemenschneider., Verschwindungss\"atze f\"ur analytische Kohomologiegruppen auf komplexen R\"aumen, Inv.\ Math.\ 11 (1970), 263--292

    1970

  2. [10]

    Grothendieck, A., Cohomologie locale des faisceaux coh\'erents et th\'eor\`emes de Lefschetz locaux et globaux (SGA 2), North Holland, Amsterdam, 1968

    1968

  3. [11]

    Hartshorne, R., Algebraic Geometry, Springer, New York, 1977

    1977

  4. [12]

    Kac, V., Watanabe, K.-i., Finite linear groups whose ring of invariants is a complete intersection, Bull.\ Amer.\ Math.\ Soc.\ (N.S.) 6 (1982), 221--223

    1982

  5. [13]

    Kov\'acs, S.J., Rational, log canonical, Du Bois singularities: on the conjectures of Koll\'ar and Steenbrink, Comp.\ Math.\ 118 (1999), 123--133

    1999

  6. [14]

    Lamotke, K., The topology of complex projective varieties after S.\ Lefschetz, Topology 20 (1981), 15--51

    1981

  7. [15]

    Milnor, J., Singular Points of Complex Hypersurfaces, Princeton Univ.\ Press, 1968

    1968

  8. [16]

    9 (1961), 5--22

    Mumford, D., The topology of normal singularities of an algebraic surface and a criterion for simplicity, Publ.\ Math.\ I.H.E.S. 9 (1961), 5--22

    1961

  9. [17]

    Namikawa, Y, Steenbrink, J.H.M., Global smoothing of Calabi-Yau threefolds, Inv.\ Math.\ 122 (1995), 403--419

    1995

  10. [18]

    Park, S.G., Popa, M., -factoriality and Hodge-Du Bois theory (arxiv:2508.17748)

    arXiv

  11. [19]

    Reichelt, T., Saito, M., Walther, U., Dependence of Lyubeznik numbers of cones of projective schemes on projective embeddings, Selecta Math.\ (N.S.) 27 (2021), Paper No.\ 6, 22 pp

    2021

  12. [20]

    Saito, M., On the exponents and the geometric genus of an isolated hypersurface singularity, Proc. Symp. Pure Math. 40, Part 2, A.M.S. (1983), 465--472

    1983

  13. [21]

    RIMS, Kyoto Univ.\ 26 (1990), 221--333

    Saito, M., Mixed Hodge modules, Publ. RIMS, Kyoto Univ.\ 26 (1990), 221--333

    1990

  14. [22]

    Providence, RI, 1991, pp 509--517

    Saito, M., On Koll\'ar's conjecture, Proc.\ Symp.\ Pure Math., 52, Part 2, A.M.S. Providence, RI, 1991, pp 509--517

    1991

  15. [23]

    Saito, M., Mixed Hodge complexes on algebraic varieties, Math.\ Ann.\ 316 (2000), 283–331

    2000

  16. [24]

    Saito, M., Lowest non-zero vanishing cohomology of holomorphic functions (arxiv:2008.10529)

    2008 arXiv

  17. [25]

    Saito, M., Bernstein-Sato polynomials for projective hypersurfaces with weighted homogeneous isolated singularities (arxiv:1609.04801v11)

  18. [26]

    urmann, J., Topology of Singular Spaces and Constructible Sheaves, Monografie Matematyczne, vol. 63, Birkh\

    Sch\"urmann, J., Topology of Singular Spaces and Constructible Sheaves, Monografie Matematyczne, vol. 63, Birkh\"auser, Basel, 2001

    2001

Pith tools

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