Single-author papers · review archive
Refine reviews, audits, and errata
An organized record of the automated Refine reports for Daniel Litt's single-author published papers, the subsequent mathematical audits, and the resulting errata. Automated comments are reproduced as received; the audit and erratum panels record the independent evaluation and adopted correction.
Coauthored papers are omitted from this public draft pending permission from the coauthors.
Publication note. Before making the archive public, confirm that redistribution of the Refine-generated text is permitted under the agreement governing the reviews.
Collection-wide findings
Overall audit
Single-author published papers only.
Of the 105 detailed comments, 97 were correct, 7 were partially correct, and 1 were incorrect. Thus 104 comments (99.0%) identified a real issue, although the proposed explanation or repair sometimes needed revision.
Most findings were local: 79 were correctable errors confined to a statement, proof step, formula, citation, or hypothesis. The audit found 1 substantial theorem-preserving defect and 4 cases requiring a technical correction to a main result. No published-paper finding was classified as a fundamental failure.
100 assessments were high confidence and 5 were medium confidence. The 4 technical corrections to main results are summarized below.
Validity of Refine comments
| Code | Meaning | Comments | Share |
|---|---|---|---|
V0 | Incorrect | 1 | 1.0% |
V1 | Not applicable or already addressed | 0 | 0.0% |
V2 | Uncertain | 0 | 0.0% |
V3 | Partially correct | 7 | 6.7% |
V4 | Correct | 97 | 92.4% |
Impact after audit
| Code | Meaning | Comments | Share |
|---|---|---|---|
I0 | No defect | 3 | 2.9% |
I1 | Expository or stylistic | 18 | 17.1% |
I2 | Local correctable error | 79 | 75.2% |
I3 | Substantial but theorem-preserving defect | 1 | 1.0% |
I4 | Technical correction to a main result | 4 | 3.8% |
I5 | Fundamental failure | 0 | 0.0% |
IP | Impact pending | 0 | 0.0% |
Technical corrections to main results
| Paper | Location | Necessary modification |
|---|---|---|
| P20 | PDF pp. 598-599, Theorem 1.10 | Add the omitted hypothesis $p>\dim(X)$. |
| P20 | PDF pp. 602-603 and 625, Theorems 1.20 and 4.29 | Add the omitted hypothesis that $X$ is projective. |
| P20 | PDF pp. 624-626, Lemma 4.28 and Theorem 4.29 | Add the omitted hypothesis that $D$ is reduced. |
| P20 | PDF p. 625, Theorem 4.29(3); compare Theorem 1.20(2) | Add the omitted hypothesis that $Y$ is proper in the no-rational-curves case. |
How to read the audit codes
The audit separates validity from impact. V0 means the comment is incorrect; V3 means it locates a real issue but misstates some material aspect; and V4 means it is correct. Impact runs from I0 (no defect) through I4 (technical correction to a main result). A difficult or initially uncertain comment is not assigned a higher impact merely because it required investigation.
An omitted argument that is standard and safely reconstructible is treated as exposition, not as a mathematical error. Every apparent I3 or I4 finding received a separate mathematical challenge before the final classification and repair were adopted.
Paper by paper
Single-author published papers
8 papers
P01 Motives, mapping class groups, and monodromy32 detailed comments · 24 numbered corrections 8 I124 I2
These errata refer to the version published in Current Developments in Mathematics 2023--2024 (2024), no. 1, pp. 165--239. Page references are to the printed pages of that version. The numbering below follows the order of the paper.
Page 166, Introduction. The sentence beginning “given (say) a smooth proper morphism” does not include the connectedness hypotheses needed for the displayed homotopy exact sequence. Replace it by:
Given a smooth proper morphism
\[ f:\mathcal X\longrightarrow S \]of connected complex algebraic varieties with geometrically connected fibers, and points $x\in\mathcal X$ and $s=f(x)\in S$, how is the geometry of $f$ reflected in the exact sequence
\[ \pi_1(X_s,x)\longrightarrow \pi_1(\mathcal X,x) \longrightarrow \pi_1(S,s)\longrightarrow 1, \]in the induced outer action of $\pi_1(S,s)$ on $\pi_1(X_s)$, and in the induced action on conjugacy classes of representations of $\pi_1(X_s)$?
The families used later in the paper have connected fibers, so no subsequent statement is changed.
Page 169, question (2) in the Introduction to §2. The sentence “the question of classifying tuples $\underline C$ such that $Y(\underline C)$ is a singleton” inadvertently enlarges the locus from the irreducible locus used in the question and in the subsequent definition. Replace that phrase by
the question of classifying tuples $\underline C$ such that $Y(\underline C)^{\mathrm{irr}}$ is a singleton.
Pages 170 and 181, middle convolution. The assertions that middle convolution preserves rigid irreducible objects and that $\MC_\lambda$ and $\MC_{\lambda^{-1}}$ are quasi-inverse are not statements about the whole category $\operatorname{Rep}(\pi_1(X))$: exceptional rank-one objects can be killed by middle convolution. Replace the first two bullets on p. 170 by:
Interpret middle convolution in Katz's middle-convolution category, namely the Serre quotient of middle-extension perverse sheaves by the exceptional rank-one constant/Kummer objects. In this category, convolution by the nontrivial Kummer character of monodromy $\lambda$ is an equivalence with inverse convolution by the character of monodromy $\lambda^{-1}$. For the nonexceptional irreducible local systems occurring in the rank-reduction argument, $\MC_\lambda$ preserves irreducibility and rigidity, and $\MC_{\lambda^{-1}}\MC_\lambda$ is naturally isomorphic to the identity.
After Definition 2.3.9 on p. 181, insert:
The formula above can vanish on exceptional rank-one local systems. All preservation and inverse statements about middle convolution in §2.1 are understood in the quotient category, or equivalently on the nonexceptional irreducible objects used in Katz's rank-reduction procedure.
The rank-reduction argument only applies these statements to those nonexceptional objects; see [Kat96, §2.8 and Chapter 6] for the middle-convolution formalism used here.
Page 171, final paragraph of §2.1. The output of middle convolution need not have determinant one at each puncture. In the sentence beginning “Thus given a tuple of conjugacy classes,” replace
\[ C'_1,\ldots,C'_n\subset \SL_{r'}(\mathbb C) \]by
\[ C'_1,\ldots,C'_n\subset \GL_{r'}(\mathbb C). \]Thus the last display in the paragraph remains
\[ Y(\underline C)\longrightarrow Y(\underline C'), \]with $\underline C'$ regarded as a tuple of $\GL_{r'}$-conjugacy classes. No later use requires the individual output classes to lie in $\SL_{r'}$.
Pages 171--172, beginning of §2.2. The displayed Artin presentation is the braid group $B_n$, but $B_n/Z(B_n)$ is not the full mapping class group $\Mod_{0,n}$: it is the subgroup of $\Mod_{0,n+1}$ fixing one distinguished puncture. Replace the sentence identifying the Artin presentation with $\Mod_{0,n}$ by:
The displayed presentation is the usual Artin presentation of $B_n$. Its quotient by the center is the mapping class group of an $(n+1)$-punctured sphere fixing one puncture. A presentation of $\Mod_{0,n}$ using the same half-twists also imposes the sphere relations
\[ (\sigma_1\cdots\sigma_{n-1})^n=1, \qquad \sigma_1\cdots\sigma_{n-2}\sigma_{n-1}^2 \sigma_{n-2}\cdots\sigma_1=1. \]The Hurwitz action on simultaneous-conjugacy classes of product-one tuples satisfies these relations and therefore induces the asserted action of $\Mod_{0,n}$ (and of $\PMod_{0,n}$) on $Y(0,n,r)$.
With this replacement, the later mapping-class-group actions are unchanged.
Page 172, §2.2.1. The sentence saying that all irreducible two-dimensional representations on the three-punctured sphere are “precisely” hypergeometric omits the standard rank-one normalization. Replace it by:
After tensoring by a rank-one local system and, if necessary, permuting the three punctures and normalizing the local exponents, every irreducible two-dimensional local system on $\mathbb{CP}^1\setminus\{x_1,x_2,x_3\}$ is the monodromy local system of a Gauss hypergeometric equation ${}_2F_1(a,b;c\mid z)$.
The rigidity assertion preceding this sentence is unaffected.
Page 173, Markoff equation discussion. The sentence “the integral solutions to (2.2) form a single orbit” is false without a positivity restriction; for example, $(0,0,0)$ is a fixed integral solution. Replace it by:
Markoff showed that the positive integral solutions to (2.2) form a single orbit under the Vieta involutions (equivalently, under the Vieta involutions and permutations), with representative $(1,1,1)$.
No subsequent result uses the unrestricted statement.
Page 174, Example 2.2.4. The displayed matrix $A_1$ contains $1/x_1$, but the stated parameter range allows $x_1=0$ (for example, $\alpha=\beta=\tfrac12$). Replace the opening sentence of the example by:
A countably infinite subfamily of the orbits mentioned above has representatives given by the following matrices, for $\alpha,\beta\in\mathbb Q$ satisfying
\[ x_1=2\cos\!\left(\frac{\pi(\alpha+\beta)}2\right)\ne0. \]The displayed matrices then define a representative on the stated parameter chart; the example is not used later.
Page 177, paragraph after Theorem 2.2.8. The sentence “So we have classified finite $\Mod_{0,n}$-orbits” drops both the “interesting” hypothesis and the infinite-local-order hypothesis of Theorem 2.2.8. Replace it by:
Thus Theorem 2.2.8 classifies the interesting finite $\Mod_{0,n}$-orbits on $Y(0,n,2)$ for which at least one local monodromy matrix $A_i$ has infinite order, in terms of certain finite subgroups of $\GL_{n-2}(\mathbb C)$.
The caveat on p. 178 and Corollary 2.2.12 already use this restricted range.
Page 177, Theorem 2.2.10. The natural three-dimensional image of $\PSL_2(\mathbb F_7)$ lies in $\SL_3(\mathbb C)$ and is not itself generated by pseudoreflections. In the list of exceptional complex reflection groups, replace
the group $\PSL_2(\mathbb F_7)$ with its natural 3-dimensional representation
by
the Shephard--Todd group $G_{24}$, the scalar extension of the natural three-dimensional representation of $\PSL_2(\mathbb F_7)$; its projective quotient is $\PSL_2(\mathbb F_7)$.
Only this example in the list is changed.
Pages 178--180, §2.3. The sentence preceding Proposition 2.3.3 says immediately that Question 2.3.2 is the same as classifying finite mapping-class-group orbits. Proposition 2.3.3 only produces an extension over a dominant family; the Corlette--Simpson dichotomy then has a separate pullback branch. Replace the paragraph ending “as we now explain” by:
Finite $\Mod_{0,n}$-orbits first give local systems on dominant families, as in Proposition 2.3.3. For a Zariski-dense rank-two local system on the total space, the Corlette--Simpson and Loray--Pereira--Touzet dichotomy gives either a rigid local system of geometric origin or a projective local system pulled back from a Deligne--Mumford curve. Section 2.3.4 treats the pullback branch. After that branch and the degenerate cases have been separated, classifying the remaining finite orbits is equivalent to Question 2.3.2.
The organization and the conclusions of §§2.3.4--2.3.6 are unchanged.
Page 182, Question 2.4.1. The word “finite” is missing before the orbit condition. Replace the question by:
Can one classify conjugacy classes of tuples of matrices $(A_1,\ldots,A_n)\in Y(0,n,2)$ with finite $\Mod_{0,n}$-orbit, without the condition that some $A_i$ have infinite order?
Page 183, first paragraph of §3.1. The parenthetical assertion that $\Mod_{g,0}$ is the subgroup of $\operatorname{Out}(\pi_1(\Sigma_g))$ acting on $H_1(\Sigma_g,\mathbb Z)$ with determinant one is false when $g$ is even: an anti-symplectic map then also has determinant one. Replace the parenthesis by:
in particular, it is the subgroup whose action on $H_1(\Sigma_g,\mathbb Z)$ preserves the algebraic intersection form; the other coset acts anti-symplectically.
The stated index-two identification remains correct.
Page 186, Definition 3.2.2 and the residue paragraph. For $\dim X>1$, the fiber of $\Omega_X^1(\log D)$ at a point of $D$ also contains tangential cotangent directions and is not canonically one-dimensional. Replace the sentence beginning “The fiber of the sheaf” by:
The residue exact sequence identifies
\[ \Omega_X^1(\log D)/\Omega_X^1\simeq\mathcal O_D. \]If $z$ is a local equation for $D$, the residue class of $dz/z$ maps to $1\in\mathcal O_D$; this description is independent of the choice of $z$.
The composite defining $\operatorname{Res}_x(\nabla)$ in the following sentence already uses this quotient and is unchanged.
Pages 190--192, proof sketch of Theorem 3.1.5. At the boundary value $g=r^2$, Theorem 3.3.1 cannot be applied to the full endomorphism local system, whose rank is $r^2$; scalar determinant deformations also remain in that tangent space. Replace the sentence “Now by Theorem 3.3.1, applied to $\operatorname{ad}(\mathbb V')$, $\mathbb V'$ is cohomologically rigid” and the ensuing rigidity step by the following fixed-determinant argument:
Use the extension supplied in the proof of [LL24b, Corollary 2.3.5], whose determinant on the total space has finite order, and perform Mochizuki's deformation with this determinant fixed. After the dominant \'{e}tale base change used in [LL24b, Lemma 2.4.2], write
\[ \mathbb V'=\mathbb U\otimes\pi^*\mathbb L, \]where $\mathbb U$ is unitary on the total space. Then
\[ \End^0(\mathbb V')=\End^0(\mathbb U) \]is unitary on the total space and has rank $r^2-1<g$. Fiberwise irreducibility gives
\[ \pi_*\End^0(\mathbb V')=0, \]while Theorem 3.3.1 gives
\[ H^0\!\left(M,R^1\pi_*\End^0(\mathbb V')\right)=0. \]The low-degree Leray sequence therefore yields
\[ H^1\!\left(\mathcal X,\End^0(\mathbb V')\right)=0. \]This is the tangent space to fixed-determinant deformations of $\mathbb V'$. Hence $\mathbb V'$ is isolated in the fixed-determinant moduli space. Scalar infinitesimal deformations have been removed, and the remaining scalar twists preserving the determinant are $r$-torsion and therefore discrete. The deformation from $\mathbb V$ to $\mathbb V'$ and the subsequent integrality argument may thus be carried out with determinant fixed.
This supplies the strict rank inequality needed in the equality case and leaves Theorem 3.1.5 and its later uses unchanged.
Pages 194--195, Definition 4.1.2 and the Atiyah sequence. An $\mathcal O_X$-linear splitting of the Atiyah sequence is a connection, but it is flat only when its curvature vanishes. Replace the sentence “The data of a flat connection $\nabla$ on $E$ is the same as the data of an $\mathcal O$-linear splitting” by:
The data of a connection $\nabla$ on $E$ is the same as the data of an $\mathcal O_X$-linear splitting $q_\nabla$ of the Atiyah sequence. The connection is flat precisely when the splitting preserves Lie brackets:
\[ [q_\nabla(v),q_\nabla(w)]=q_\nabla([v,w]) \]for local vector fields $v,w$; equivalently, the curvature of $\nabla$ vanishes.
With this condition, the splitting on p. 195 is a map of complexes exactly as claimed.
Page 200, proof sketch of Proposition 4.3.5. The proof invokes Conjecture 4.3.4, whose statement explicitly omits the finite-order $p$-integrality condition needed in the argument, and it places the original connection rather than its pullback on the covering curve in the genus-$g$ moduli space. Replace the three sentences beginning “By a direct computation with Taylor series” by:
By a direct computation with Taylor series, the isomonodromy leaf through
\[ [(Y,(E,\nabla)|_Y)] \in M_{\mathrm{dR}}(\mathcal C_g/\mathcal M_g,r) \]is $p$-integral to order $\omega(p)$ for almost all $p$. Conjecture 4.3.1, including its finite-order $p$-integrality assertion, then implies that this leaf is algebraic. Equivalently, the monodromy of $(E,\nabla)|_Y$ has finite orbit under $\Mod_g=\pi_1(\mathcal M_g)$. Since $g\ge r^2$, Theorem 3.1.5 shows that $(E,\nabla)|_Y$ has finite monodromy. The subgroup $\pi_1(Y)\subset\pi_1(X)$ has finite index, so $(E,\nabla)$ itself has finite monodromy.
Pages 200--201, paragraph preceding Theorem 4.3.9. The sentence suggesting that the Picard--Fuchs hypothesis of Theorem 4.3.7 is mild conflates a Picard--Fuchs equation with a direct summand of one. Definition 2.3.1 only gives the latter notion of geometric origin, while Remark 4.2.9 records that the relevant result is not known for arbitrary direct summands. Replace the paragraph by:
Examples naturally produce local systems of geometric origin, hence direct summands of Picard--Fuchs local systems. Theorem 4.3.7 applies when the flat bundle is the full Picard--Fuchs equation of Definition 4.2.7. Theorem 4.3.9 below supplies geometric origin, and therefore a direct-summand realization, but does not by itself verify the stronger hypothesis of Theorem 4.3.7.
Theorem 4.3.9 itself is unchanged.
Pages 201--202, proof of Corollary 4.4.3. The inclusion of a fiber $X$ into $\mathcal X_{\widetilde S}:=\mathcal X\times_S\widetilde S$ need not induce an isomorphism on fundamental groups: the homotopy sequence contains a boundary map
\[ \pi_2(\widetilde S)\longrightarrow\pi_1(X). \]Replace the proof by the following argument, which works on relative character varieties and does not require a local system on $\mathcal X_{\widetilde S}$:
Over the universal cover $\widetilde S$, the local system of relative Betti character varieties is trivial. The monodromy class of $(E,\nabla)$ therefore defines a horizontal holomorphic section
\[ \sigma:\widetilde S\longrightarrow M_B(\mathcal X/S,r)^{\mathrm{an}}\times_{S^{\mathrm{an}}}\widetilde S. \]The possible $\pi_2(\widetilde S)$-ambiguity in based transport acts by inner automorphisms and is invisible on character-variety points.
Let $\mathcal H$ be Simpson's closed relative nonabelian Hodge locus of points underlying polarizable $\mathbb Z$-variations of Hodge structure, and set
\[ N=\sigma^{-1}(\mathcal H)\subset\widetilde S. \]By [Sim97, §12], $N$ is closed analytic. The assumed formal Griffiths-transverse extension of the Hodge filtration says that the formal germ of $\sigma$ at the chosen lift $\widetilde s$ lies in $\mathcal H$. Consequently, the pullback of the defining ideal of $\mathcal H$ vanishes in the completed analytic local ring at $\widetilde s$. The analytic local ring injects into its completion, so this ideal already vanishes on a neighborhood of $\widetilde s$. Thus $N$ contains a nonempty open subset; because $N$ is a closed analytic subset of the connected manifold $\widetilde S$, analytic continuation gives $N=\widetilde S$.
For a deck transformation $\delta\in\pi_1(S,s)$, the value $\sigma(\delta\widetilde s)$ is the character-variety point obtained from $\sigma(\widetilde s)$ by the corresponding outer monodromy action. Hence every point in the $\pi_1(S,s)$-orbit of $(E,\nabla)$ underlies a polarizable $\mathbb Z$-variation of Hodge structure. Deligne's finiteness theorem (Theorem 4.4.1) now shows that this orbit is finite.
Thus Corollary 4.4.3 retains its stated conclusion; only the total-space fundamental-group argument is replaced.
Page 204, Example 5.1.1. The description of $M_{\mathrm{Dol}}(X,r)$ as the coarse moduli space of semistable Higgs bundles of degree zero omits the conditions required by the nonabelian Hodge correspondence in higher dimension. Replace that sentence by:
After fixing a polarization on $X$, we let $M_{\mathrm{Dol}}(X,r)$ be the coarse moduli space of polystable rank-$r$ Higgs bundles $(E,\theta)$ on $X$ with vanishing rational Chern classes
\[ c_i(E)=0\in H^{2i}(X,\mathbb Q)\qquad(i>0). \]With this definition, the real-analytic homeomorphism with $M_B(X,r)$ on p. 205 has the stated meaning.
Page 208, first paragraph of §5.2. The normalization says $x_3=\infty$ and $x_4=\lambda$, whereas the next sentence assigns $C_3$ to $\lambda$ and $C_4$ to $\infty$. Replace the normalization by
\[ x_1=0,\qquad x_2=1,\qquad x_3=\lambda,\qquad x_4=\infty. \]The subsequent assignment of $C_i$ to the four punctures and all later formulas in §5.2 are then consistent.
Page 218, first paragraph of §6.2. Under the convention used in the paper, the full group $\Mod_{g,n+1}$ may move the distinguished point $x_0$ and therefore does not act canonically on the based group $\pi_1(\Sigma_{g,n},x_0)$. Replace the first paragraph after “Fixing a base-point $x_0$” by:
Let $\PMod_{g,n+1}$ be the pure mapping class group of the surface with the $n$ punctures and the additional marked point $x_0$. Since it fixes $x_0$, it acts naturally on $\pi_1(\Sigma_{g,n},x_0)$. If $\Sigma_{g'}\to\Sigma_g$ is a cover branched at the $n$ punctures, let $\Gamma\subset\PMod_{g,n+1}$ be the stabilizer of $\pi_1(\Sigma_{g',n'})\subset\pi_1(\Sigma_{g,n},x_0)$. Then $\Gamma$ has finite index and acts on $H_1(\Sigma_{g'},\mathbb Z)$ as described below.
Replacing the acting group by this finite-index pure subgroup does not change the subsequent finite-orbit or finite-index formulations.
Page 230, sentence following Conjecture 6.4.6. Conjecture 6.4.6 asserts only that an integral formal isomonodromic deformation implies invariance under a finite-index subgroup. When $Z$ is a point, this is only one implication in Conjecture 4.3.4. Replace “it specializes to that statement if $Z$ is a point” by:
When $Z$ is a point, Conjecture 6.4.6 gives the implication from an integral formal isomonodromic deformation to a finite monodromy orbit in Conjecture 4.3.4; it does not assert the converse implication.
Page 230, §6.4.7. The claim that nontrivial geometric subgroups cannot lie in the Torelli group must exclude families whose underlying unpointed curves are isotrivial. For example, marked points may move on a fixed curve while the homological monodromy remains trivial. Replace the paragraph beginning “There are some evident restrictions” through the Torelli assertion by:
There are some evident restrictions on geometric subgroups arising from non-isotrivial families of underlying curves. Such a subgroup cannot be contained in the Torelli group. Indeed, after passing to a finite cover, finite homological monodromy becomes trivial. The theorem of the fixed part then makes the weight-one variation $R^1q_*\mathbb Q$ constant, so the period map to $\mathcal A_g$ is constant; Torelli's theorem implies that the underlying family of curves is isotrivial. This argument does not apply to families obtained by moving marked points on a fixed curve, and such families must be treated separately.
This qualification concerns the concluding expectation only and is not used in any proof.
References
N. M. Katz, Rigid Local Systems, Annals of Mathematics Studies, vol. 139, Princeton University Press, Princeton, NJ, 1996.
A. Landesman and D. Litt, Canonical representations of surface groups, Ann. of Math. (2) 199 (2024), no. 2, 823--897.
C. T. Simpson, The Hodge filtration on nonabelian cohomology, in Algebraic Geometry---Santa Cruz 1995, Proc. Sympos. Pure Math., vol. 62, part 2, Amer. Math. Soc., Providence, RI, 1997, pp. 217--281.
Report metadata
| Field | Value |
|---|---|
| Category | Published |
| Processing status | completed |
| Detailed comments | 32 |
| Domain | stem/mathematics |
| Completed | 2026-07-29T14:34:39.605787+00:00 |
| Refine document ID | b887c6da-bd5e-4bde-8f18-9c9b6528b92d |
Refine summary
This survey paper explores the intersection of algebraic geometry, surface topology, and ordinary differential equations. It focuses on the actions of mapping class groups on character varieties and their algebraic counterparts, such as isomonodromy differential equations.
Overall feedback
Character-variety definitions
Readers working through the text will likely notice that the underlying character-variety object changes across the article. Sections 2 and 3 define $Y(g,n,r)$ as the orbit set of all representations under conjugation. However, the subsequent discussion relies on algebraic notions—such as Zariski density, tangent spaces, fixed loci, and character-variety geometry—that naturally belong to the GIT quotient, where points represent semisimplifications.
While Section 4 distinguishes between the representation stack, its isomorphism classes, and the coarse GIT space, this distinction does not propagate back to earlier statements, including Conjecture 3.1.2, Theorem 3.1.5, Proposition 3.1.7, or the general finite-orbit discussion. For instance, the Markoff cubic in Section 2.2.2 explicitly parametrizes only semisimple representations. Because a finite orbit of a GIT point need not imply a finite orbit of a non-semisimple extension class, it is important that the major results specify consistently whether they govern actual conjugacy classes, semisimple representations, irreducible representations, stacks, or coarse spaces.
The rigidity mechanism in Section 3.4
When presenting the proof sketch for the flagship higher-genus theorem in Section 3.4, a load-bearing rigidity mechanism is currently skipped. After Theorem 3.3.1 yields the vanishing of $H^0(\mathscr{M}, R^1\pi_* \mathrm{ad}(\mathbb{V}'))$, the sketch immediately declares $\mathbb{V}'$ cohomologically rigid and applies Esnault–Groechenig integrality. Passing from that vanishing to $H^1(\mathscr{X}, \mathrm{ad}(\mathbb{V}'))=0$ requires a Leray argument and control over $R^0\pi_* \mathrm{ad}(\mathbb{V}')$, typically handled via a trace-free adjoint or a fixed-determinant deformation problem. Yet Section 3.3 uses $\mathrm{ad}(\rho)$ for the tangent space to the $GL_r$ character variety, which seemingly includes scalar endomorphisms.
Furthermore, the sketch relies on Mochizuki’s deformation without explaining why it remains in the relevant extension locus, why the rigidity of its limit forces the original local system to coincide with that limit, or how the reducible case is recovered. Although the text acknowledges omitting complications, these transitions are the structural pillars of Theorem 3.1.5. Exposing them explicitly would greatly serve a survey built around that result.
Formulation of Conjecture 4.3.4
Looking closely at Conjecture 4.3.4, the precisification requires a more targeted invariant formal-moduli formulation. Condition (1) stipulates that the isomonodromic deformation descends to the completed base $\widehat{S}_R[1/N]$. Given that the relevant object is a flat bundle on the formal completion of $\mathscr{X}$ along the fiber (or equivalently a formal horizontal section of a moduli stack), the statement leaves several parameters unspecified: descent up to gauge equivalence, the treatment of automorphisms or singular points, and the dependence on the chosen spreading.
Additionally, the conjecture expressly omits the $\omega(p)$-integrality condition, yet Proposition 4.3.5 derives $\omega(p)$-integrality and subsequently invokes Conjecture 4.3.4. As written, the implication does not follow from the stated conjecture. Formulating the central arithmetic classification to clearly separate and logically relate the characteristic-zero descent and the modulo-$p$ refinement will secure this argument.
Geometric origin versus Picard-Fuchs
The narrative currently treats the concepts of geometric origin and Picard-Fuchs systems as closer than the stated results justify. Definition 2.3.1 defines geometric origin using a direct summand of a Gauss–Manin local system, whereas Definition 4.2.7 reserves the "Picard-Fuchs" terminology for the entire Gauss–Manin system.
Remark 4.2.9 explicitly points out that the relevant arithmetic theorem is not known generally for direct summands. Consequently, Theorem 4.3.9—which shows certain finite-orbit systems are of geometric origin—does not wholly support the subsequent claim that the Picard-Fuchs hypothesis in Theorem 4.3.7 is not unduly restrictive, nor does it establish that the theorem approaches a characterization of those finite orbits. Bridging this gap in the evidentiary narrative, or providing a suitable direct-summand theorem with its associated hypotheses, is necessary here.
Superrigidity and projective representations
The superrigidity consequences explored in Section 6 require a stronger conjecture than the one explicitly stated. Conjecture 6.1.2 defines a rigidity property for irreducible linear representations of the full group $\mathrm{Mod}_{g,n}$. However, Proposition 6.1.5 applies this conjecture to a projective representation of a finite-index subgroup $\Gamma \subset \pi_1(\mathscr{C}_g) \cong \mathrm{Mod}_{g,1}$.
Rigidity for linear representations of a full group does not automatically transfer to projective representations of finite-index subgroups, and no central-extension or induction argument is supplied to cover the difference. Since the identical unspoken strengthening underpins the proposed route to Conjectures 6.1.6 and 6.1.7, the article must explicitly formulate the required virtual/projective version or demonstrate how Conjecture 6.1.2 safely transfers to this broader setting.
Detailed comments
1. Connected-fiber hypothesis missing in the introduction
- ID:
1393706f-f6b5-4ebe-a153-018c2f752338 - Refine score:
0.3 - Original types: general
- Refine status: open
Comment
The displayed homotopy sequence requires the fiber $X_s$ to be connected; smoothness and properness alone do not ensure the asserted surjectivity of $\pi_1(X,x)\to\pi_1(S,s)$. The paper explicitly imposes connected fibers in its later general setup, but that hypothesis is absent here.
Quoted passage
As is traditional in algebraic geometry, we view it as a special case of a much more general question about families of algebraic varieties: given (say) a smooth proper morphism
$$ f: X \rightarrow S, $$and points $x \in X, s=f(x) \in S$, how is the geometry of $f$ reflected in the exact sequence
$$ \pi_{1}\left(X_{s}, x\right) \rightarrow \pi_{1}(X, x) \rightarrow \pi_{1}(S, s) \rightarrow 1, $$
2. Uniqueness criterion shifts in Section 2
- ID:
d8e9baf8-e3f6-46e9-ab6c-e06a5df3f81e - Refine score:
0.23 - Original types: general
- Refine status: open
Comment
The uniqueness criterion shifts from $Y(\underline{C})^{\mathrm{irr}}$ being a singleton to $Y(\underline{C})$ being a singleton. Since the latter also excludes reducible representations and the subsequent definition of rigidity uses $Y(\underline{C})^{\mathrm{irr}}$, the sentence states a formally stronger and internally inconsistent condition.
Quoted passage
(2) (Uniqueness) For which $\underline{C}$ is $Y(\underline{C})^{\mathrm{irr}}$ a singleton? That is, when is a solution to (2.1) determined uniquely (up to simultaneous conjugation) by the conjugacy classes $C_{i}$ of the matrices $A_{i}$ ? For reasons that will soon become clear, the question of classifying tuples $\underline{C}$ such that $Y(\underline{C})$ is a singleton is typically referred to as the classification of rigid local systems, and was studied by Katz in his book of the same name, [Kat96].
3. Middle convolution claims need category restrictions
- ID:
a16f3718-5361-49d1-b323-91f6a6a01b8a - Refine score:
0.45 - Original types: general
- Refine status: open
Comment
The middle-convolution properties are overbroad as stated. For $\lambda\ne 1$, applying the later definition to the trivial rank-one local system gives zero, so middle convolution neither preserves irreducibility in every stated case nor defines an autoequivalence of all $\operatorname{Rep}(\pi_1(X))$. The preservation and quasi-inverse claims require the appropriate restrictions or quotient category excluding exceptional objects.
Quoted passage
For each $\lambda \in \mathbb{C}^{\times} \backslash\{1\}$, Katz produces a functor
$$ \mathrm{MC}_{\lambda}: \operatorname{Rep}\left(\pi_{1}(X)\right) \rightarrow \operatorname{Rep}\left(\pi_{1}(X)\right) $$with the following properties:
- If $\rho$ is a rigid irreducible $\pi_{1}(X)$-representation, $\mathrm{MC}_{\lambda}(\rho)$ is rigid and irreducible.
- The functors $\mathrm{MC}_{\lambda}, \mathrm{MC}_{\lambda^{-1}}$ are quasi-inverse.
- If $\rho$ is a rigid irreducible $\pi_{1}(X)$-representation of rank at least 2, there exists a rank one representation
4. Middle convolution does not visibly preserve SL
- ID:
22519809-6841-4822-921c-fc4d00afb297 - Refine score:
0.3 - Original types: general
- Refine status: open
Comment
The asserted special-linear target is not generally preserved by middle convolution: $\mathrm{MC}_\lambda$ naturally produces $\mathrm{GL}_{r'}$ local monodromies, and the product-one relation constrains only the product of their determinants, not each determinant separately. Thus the classes $C_i'$ need not lie in $\mathrm{SL}_{r'}(\mathbb{C})$ unless an additional determinant-preservation result or rank-one normalization is included.
Quoted passage
It turns out that the conjugacy class of $\mathrm{MC}_{\lambda}(\rho)\left(\gamma_{i}\right)$ depends only on $\lambda$ and the conjugacy class of $\rho\left(\gamma_{i}\right)$. Thus given a tuple of conjugacy classes $C_{1}, \ldots, C_{n} \subset \mathrm{SL}_{r}(\mathbb{C})$, there exists another (explicit) tuple $C_{1}^{\prime}, \ldots, C_{n}^{\prime} \subset$ $\mathrm{SL}_{r^{\prime}}(\mathbb{C})$ such that $\mathrm{MC}_{\lambda}$ induces a map
$$ Y(\underline{C}) \rightarrow Y\left(\underline{C^{\prime}}\right) . $$
5. Mapping class group identification in §2.2
- ID:
69e1dbf7-a9b8-4ef8-b448-2f7c7c73c2b3 - Refine score:
0.37 - Original types: general
- Refine status: open
Comment
The displayed Artin presentation is that of $B_n$, but $B_n/Z(B_n)$ is the mapping class group of an $(n+1)$-punctured sphere fixing one distinguished puncture, not the full group $\operatorname{Mod}_{0,n}$ defined here. The latter is obtained from the spherical braid group after the appropriate central quotient. The distinction is visible for $n=3$: $B_3/Z(B_3)\cong\operatorname{PSL}_2(\mathbb{Z})$, whereas $\operatorname{Mod}_{0,3}\cong S_3$. Although the Hurwitz action on product-one tuples does factor through the sphere mapping class group, the stated group identification is incorrect.
Quoted passage
That is, the action of $\left\langle\sigma_{1}, \ldots, \sigma_{n-1}\right\rangle$ on $Y(0, n, r)$ factors through the quotient
$$ \left.\left\langle\sigma_{1}, \ldots, \sigma_{n-1}\right| \sigma_{i} \sigma_{i+1} \sigma_{i}=\sigma_{i+1} \sigma_{i} \sigma_{i+1} \text { and } \sigma_{i} \sigma_{j}=\sigma_{j} \sigma_{i} \text { for }|i-j| \geq 2\right\rangle, $$which is the usual Artin presentation of the braid group, which is (up to quotienting by the center) the mapping class group
$$ \operatorname{Mod}_{0, n}:=\pi_{0}\left(\operatorname{Homeo}^{+}\left(\mathbb{C P}^{1} \backslash\left\{x_{1}, \ldots, x_{n}\right\}\right)\right) $$
6. Hypergeometric classification omits rank-one twists
- ID:
af4c43a3-b5dd-4043-ba49-4fd186bea473 - Refine score:
0.26 - Original types: general
- Refine status: open
Comment
The hypergeometric classification is correct only up to a rank-one twist and the associated normalization of local exponents. A raw Gauss ${}_2F_1(a,b;c;z)$ equation has local monodromy eigenvalue $1$ at both $0$ and $1$, whereas a general irreducible $\mathrm{GL}_2$ representation of the three-punctured sphere need not have eigenvalue $1$ at two punctures. Thus “precisely” is too strong without the normalization qualification.
Quoted passage
In fact, in this last case all irreducible 2-dimensional representations are rigid in the sense of § 2.1; while the dynamics are not interesting, this is the source of the (extremely rich) theory of hypergeometric functions, and the corresponding representations of $\pi_{1}\left(\mathbb{C} \mathbb{P}^{1} \backslash\left\{x_{1}, x_{2}, x_{3}\right\}\right)$ are precisely given by the monodromy of the hypergeometric functions ${ }_{2} F_{1}(a, b, c \mid z)$ (see e.g. [Beu07]).
7. Markoff orbit claim needs a positivity restriction
- ID:
f69db39e-1213-458b-b31c-47eb0f049b30 - Refine score:
0.27 - Original types: general
- Refine status: open
Comment
The claim that all integral solutions of the Markoff equation form a single orbit is false without a restriction: $(0,0,0)$ is an integral solution fixed by all three Vieta involutions and therefore cannot lie in the orbit of $(1,1,1)$. The classical transitivity statement concerns an appropriately restricted set, such as positive Markoff triples.
Quoted passage
Markoff was not interested in finite orbits-rather, he showed that the integral solutions to (2.2) form a single orbit under these dynamics. We will return to questions about integral points later in these notes, in § 5.4.10; instead we now turn to the origin of our question about finite orbits. 2.2.3. $n=4$ and the Painlevé VI equation.
8. Example 2.2.4 is undefined for some rational parameters
- ID:
29719079-adae-4483-93d3-8662de09780b - Refine score:
0.28 - Original types: general
- Refine status: open
Comment
The displayed parametrization is not defined for every stated pair $\alpha,\beta\in\mathbb{Q}$ because $A_1$ contains divisions by $x_1$. For example, $\alpha=\beta=\tfrac12$ gives $x_1=0$ and $x_2=x_3=\sqrt{2}$, so the matrix entries are undefined. The parameter domain must exclude such values unless they are handled by a different representative or limiting construction.
Quoted passage
where
$$ x_{1}=2 \cos \left(\frac{\pi(\alpha+\beta)}{2}\right), x_{2}=2 \sin \left(\frac{\pi \alpha}{2}\right), x_{3}=2 \sin \left(\frac{\pi \beta}{2}\right) $$for $\alpha, \beta \in \mathbb{Q}$. See [LL23a, Example 1.1.7] for a discussion of this example, and the rest of that paper for an involved analysis of some related arithmetic questions. See also § 5.2 for a brief further discussion of this example.
9. Scope of the classification is overstated in §2.2
- ID:
baabcd09-dda4-4ba5-927d-33312c5df85c - Refine score:
0.29 - Original types: general
- Refine status: open
Comment
The claim that finite $\operatorname{Mod}_{0,n}$-orbits on $Y(0,n,2)$ have been classified is too broad. Theorem 2.2.8 treats interesting orbits for which some $A_i$ has infinite order; the later discussion expressly leaves the all-finite-order interesting case open.
Quoted passage
Here $\mathrm{MC}_{\lambda}$ is the middle convolution operation introduced by Katz in [Kat96] and discussed earlier in § 2.1.
Note that if $n>4$, the $B_{i}$ are not $2 \times 2$ matrices-they are $(n-2) \times(n-2)$ matrices. So we have classified finite $\operatorname{Mod}_{0, n}$-orbits on $Y(0, n, 2)$ in terms of certain finite subgroups of $\mathrm{GL}_{n-2}(\mathbb{C})$. In what sense is this actually a classification? The point is that finite complex reflection groups were classified by Shephard and Todd [ST54] in 1954.
10. The §2.2 Shephard–Todd list misidentifies $PSL_2(7)$
- ID:
e474222f-8d0b-4765-8e16-1ad0399b304b - Refine score:
0.26 - Original types: general
- Refine status: open
Comment
The natural irreducible three-dimensional representation of $PSL_2(\mathbb{F}_7)$ is not itself a complex reflection group under Definition 2.2.9: its image lies in $SL_3(\mathbb{C})$, while every nonidentity pseudoreflection in dimension three has nontrivial determinant. The corresponding exceptional Shephard–Todd group is a scalar extension whose projective quotient is $PSL_2(\mathbb{F}_7)$.
Quoted passage
Theorem 2.2.10 (Shephard-Todd [ST54]). There is one infinite class of finite complex reflection groups, denoted $G(m, p, n) \subset G L_{n}(\mathbb{C})$, where $p$ divides $m$. The group $G(m, 1, n)$ consists of all $n \times n$ matrices with exactly one non-zero entry in each row and column, where that non-zero entry is an $m$-th root of unity. The group $G(m, p, n) \subset G(m, 1, n)$ is the subgroup consisting of matrices whose non-zero entries multiply to an $m / p$-th root of unity.
There are 34 exceptional irreducible finite complex reflection groups not conjugate to one of the $G(m, p, n)$, including the Weyl groups $W\left(E_{6}\right), W\left(E_{7}\right)$, $W\left(E_{8}\right)$, the Valentiner group, the group $P S L_{2}\left(\mathbb{F}_{7}\right)$ with its natural 3-dimensional representation, the automorphism group of the icosahedron, and so on.
11. Equivalence overstated at the start of §2.3
- ID:
310ebc20-1fa3-4c9e-98b9-46c70d2211aa - Refine score:
0.32 - Original types: general
- Refine status: open
Comment
The statement that Question 2.3.2 is “the same as” classifying all finite $\operatorname{Mod}_{0,n}$-orbits is too broad. Proposition 2.3.3 identifies finite orbit with extension over a dominant family, not with geometric origin. Under the subsequent rank-two, Zariski-dense dichotomy, the finite-orbit problem has a pullback-type branch and a rigid/geometric-origin branch; only after the pullback branch and the earlier degenerate cases are handled does the remaining problem coincide with Question 2.3.2.
Quoted passage
Question 2.3.2. Let $x_{1}, \ldots, x_{n} \subset \mathbb{C P}^{1}$ be $n$ generic points. Can one classify local systems of rank 2 on $\mathbb{C P}^{1} \backslash\left\{x_{1}, \ldots, x_{n}\right\}$ that are of geometric origin?
It turns out that this question is the same as classifying finite $\operatorname{Mod}_{0, n^{-}}$ orbits on $Y(0, n, 2)$, as we now explain.
The following is immediate from the proof of [LL24b, Corollary 2.3.5] (note that the condition that $g \geq 1$ is unnecessary in our setting, as we are working with $\mathrm{SL}_{2}(\mathbb{C})$-representations):
12. Missing finiteness condition in Question 2.4.1
- ID:
167946db-6357-440b-80e3-2d2bee8e60a7 - Refine score:
0.3 - Original types: general
- Refine status: open
Comment
Question 2.4.1 omits the requirement that the $\operatorname{Mod}_{0,n}$-orbit be finite. As written, “finite tuples” only describes an $n$-tuple, and every point of $Y(0,n,2)$ has an orbit, so the question does not formally state the finite-orbit classification problem developed in §2.
Quoted passage
2.4. Some questions. These results leave a number of questions unresolved; we briefly record two such questions here for the reader who will depart prematurely - there are many, many more such questions later in these notes.
Question 2.4.1. Can one classify conjugacy classes of finite tuples of matrices $\left(A_{1}, \ldots, A_{n}\right) \in Y(0, n, 2)$ with $\operatorname{Mod}_{0, n}$-orbit, without the condition that some $A_{i}$ have infinite order?
Question 2.4.2. Can one say anything about finite $\operatorname{Mod}_{0, n}$-orbits in $Y(0, n, r)$ with $r>2$ ?
13. Determinant does not characterize orientation in §3.1
- ID:
30c73e27-b93f-4526-be88-163066de2848 - Refine score:
0.29 - Original types: general
- Refine status: open
Comment
The parenthetical characterization by determinant is false when $g$ is even. Orientation-preserving mapping classes preserve the algebraic intersection form, whereas orientation-reversing classes act anti-symplectically. An anti-symplectic automorphism of a rank-$2g$ symplectic lattice has determinant $(-1)^g$, which is also $1$ for even $g$.
Quoted passage
The group $\operatorname{Mod}_{g, 0}=\pi_{1}\left(\mathscr{M}_{g}\right)$ has a simple group-theoretic interpretation: it is of index 2 in $\operatorname{Out}\left(\pi_{1}\left(\Sigma_{g}\right)\right)$ (in particular, it is the subgroup acting on $H_{1}\left(\Sigma_{g}, \mathbb{Z}\right)$ with determinant 1).
Generalizing our previous definitions, we set
14. Residue discussion conflates a sheaf fiber with its quotient
- ID:
5c0514b5-2c2e-4a68-9361-cff6fbaeaca9 - Refine score:
0.25 - Original types: general
- Refine status: open
Comment
The claim that the fiber of $\Omega_X^1(\log D)$ is canonically trivialized by $dz/z$ is false when $\dim X>1$: that fiber also has tangential cotangent directions, and the class of $dz/z$ depends on the local equation. The canonical object used in the following display is instead the residue quotient $\Omega_X^1(\log D)/\Omega_X^1\simeq\mathscr{O}_D$, where the class of $dz/z$ is well defined. The subsequent residue construction is therefore correct, but the preceding description of the full fiber is not.
Quoted passage
The fiber of the sheaf $\Omega_{X}^{1}(\log D)$ at a point $x \in D$ is canonically trivialized by the 1-form $d z / z$, where $z$ is any local equation for $D$ at $x$. A local computation shows that the composite map
$$ \left.\mathscr{E} \xrightarrow{\nabla} \mathscr{E} \otimes \Omega_{X}^{1}(\log D) \rightarrow \mathscr{E} \otimes\left(\Omega_{X}^{1}(\log D) / \Omega_{X}^{1}\right) \simeq \mathscr{E}\right|_{D} $$is $\mathscr{O}_{X}$-linear and hence factors through $\left.\mathscr{E}\right|_{D}$.
15. Schlesinger derivation leaves the $C_i$ terms unresolved
- ID:
35f587a7-81ba-442c-b89d-5abe8a77797f - Refine score:
0.34 - Original types: general
- Refine status: open
Comment
In Example 3.2.6, the displayed curvature also contains terms involving $dC_i$, $[A_i,C_j]$, and $[C_i,C_j]$. Thus flatness gives the stated Schlesinger equations for the residues only after a base-dependent gauge normalization eliminating the $C_i$, or under equivalent conditions that make their contributions vanish; without that qualification, the computation as written is incomplete.
Quoted passage
$$ \widetilde{\nabla}=d-\sum_{i=1}^{n} \frac{A_{i}(\underline{x})}{z-x_{i}} d\left(z-x_{i}\right)+\sum C_{i} d x_{i}, $$i.e. a connection with regular singularities along the evident divisors where $z=x_{i}$. Note that $\sum A_{i}(\underline{x})=0$ by our assumption that there is no pole at $\infty$.
The condition that $\widetilde{\nabla}$ be isomonodromic is simply the condition that $\widetilde{\nabla} \circ \widetilde{\nabla}=0$. What condition does this impose on the $A_{i}$ ? For $i \neq j$, considering the coefficient of $d\left(z-x_{i}\right) \wedge d\left(z-x_{j}\right)$ gives precisely the first line of (2.4); differentiating the identity $\sum A_{i}(\underline{x})=0$ gives the second.
16. Rigidity step needs the traceless adjoint
- ID:
7e5c68a7-f8b7-4289-a63e-acdbd8c8b5f6 - Refine score:
0.42 - Original types: general
- Refine status: open
Comment
The rigidity deduction requires the fixed-determinant deformation problem and the traceless adjoint $\operatorname{End}^0(\mathbb V')$ to be explicit. For the full endomorphism local system, the rank is $r^2$, so Theorem 3.3.1 does not apply when $g=r^2$, and scalar determinant-changing deformations remain. With $\operatorname{End}^0(\mathbb V')$, the rank is $r^2-1<g$; fiberwise irreducibility also gives $\pi_*\operatorname{End}^0(\mathbb V')=0$, allowing the Leray sequence to convert Theorem 3.3.1’s vanishing into the needed fixed-determinant cohomological rigidity.
Quoted passage
By work of Mochizuki [Moc06, Th. 10.5], $\mathbb{V}$ can be deformed to a polarizable complex variation of Hodge structure $\mathbb{V}^{\prime}$ (see Theorem 5.4.7 for a variant of this result and a sketch of the proof). We assume for simplicity that $\mathbb{V}^{\prime}$ is irreducible when restricted to a fiber of $\pi .{ }^{12}$ By Corollary 3.2.10, $\mathbb{V}^{\prime}$ is in fact unitary. Now by Theorem 3.3.1, applied to $\operatorname{ad}\left(\mathbb{V}^{\prime}\right), \mathbb{V}^{\prime}$ is cohomologically rigid, i.e. it admits no non-trivial infinitesimal deformations.
This observation has two consequences:
(1) as we have deformed $\mathbb{V}$ to a representation with no non-trivial deformations, we must have that $\mathbb{V}$ and $\mathbb{V}^{\prime}$ are isomorphic to one another, and (2) $\mathbb{V}^{\prime}$ (and hence $\mathbb{V}$ ) are defined over $\mathscr{O}_{K}$, the ring of integers of some number field $K$, by work of Esnault-Groechenig [EG18, Theorem
17. Flatness is missing from the Atiyah splitting criterion
- ID:
e0d7ddb3-ba3d-4fa5-8c08-1b73505001d1 - Refine score:
0.3 - Original types: general
- Refine status: open
Comment
An $\mathscr{O}_X$-linear splitting of the Atiyah sequence corresponds to a connection, not necessarily a flat one. Flatness additionally requires the splitting to preserve Lie brackets, equivalently to have zero curvature. This distinction matters here because the later claim that $q^\nabla$ is a map of complexes uses precisely that extra condition.
Quoted passage
By construction there is a short exact sequence
$$ 0 \rightarrow \operatorname{End}(\mathscr{E}) \rightarrow \operatorname{At}(\mathscr{E}) \stackrel{\tau}{\rightarrow} T_{X} \rightarrow 0, $$called the Atiyah exact sequence. The data of a flat connection ∇ on $\mathscr{E}$ is the same as the data of an $\mathscr{O}$-linear splitting $q^{\nabla}$ of this sequence, where $q^{\nabla}(v)(s)$ is given by contracting $v$ with ∇ (s).
Now suppose we are given a flat connection ∇ on $\mathscr{E}$, and consider the complex
$$ \operatorname{At}(\mathscr{E})_{d R}^{\bullet}: \operatorname{At}(\mathscr{E}) \rightarrow \operatorname{End}(\mathscr{E}) \otimes \Omega_{X}^{1} \xrightarrow{\nabla} \operatorname{End}(\mathscr{E}) \otimes \Omega_{X}^{2} \xrightarrow{\nabla} \cdots $$
18. Ambiguous reduction in the rank-one example
- ID:
62aae96b-5229-4800-a97e-0d97634c89e9 - Refine score:
0.28 - Original types: general
- Refine status: open
Comment
The phrase “identically zero $\bmod p$” is ambiguous for coefficients in $K=\mathbb{Q}(a)$. Vanishing after reduction at one prime $\mathfrak p\mid p$ only implies that the residue of $a$ lies in $\mathbb{F}_p$, whereas complete splitting follows when vanishing is imposed in the full ring $\mathcal{O}_K/p\mathcal{O}_K$, equivalently at every prime above $p$. Without that global quantifier, the stated equivalence with complete splitting is not valid.
Quoted passage
On the other hand,
$$ \left(\frac{d}{d z}-\frac{a}{z}\right)^{p} z^{n}=(n-a)(n-a-1) \cdots(n-a-p+1) z^{n-p} $$is identically zero $\bmod p$ for almost all $p$ iff $p$ splits completely in $\mathbb{Q}(a)$ for almost all $p$; this happens if and only if $a \in \mathbb{Q}$, by the Chebotarev density theorem.
There is a more intrinsic formulation of Conjecture 4.2.3, which makes sense on general smooth bases.
19. Proposition 4.3.5 switches conjectures prematurely
- ID:
2a514e2f-fd0f-4122-a911-cf71fb84ae63 - Refine score:
0.35 - Original types: general
- Refine status: open
Comment
The proof of Proposition 4.3.5 establishes only $p$-integrality through order $\omega(p)$ but then invokes Conjecture 4.3.4, whose stated formulation explicitly omits that condition and instead assumes full integral descent. The argument therefore needs the finite-order $p$-integrality equivalence from Conjecture 4.3.1. In addition, the relevant point of the genus-$g$ de Rham moduli space is the pullback of $(\mathscr{E},\nabla)$ to $Y$, rather than the original bundle on $X$.
Quoted passage
By a direct computation with Taylor series, the leaf of the isomonodromy foliation through $[(\mathscr{E}, \nabla)] \in \mathscr{M}_{d R}(\mathscr{X} / \mathscr{S})$ is $p$-integral to order $\omega(p)$. Then Conjecture 4.3.4 implies that the monodromy of $\left.(\mathscr{E}, \nabla)\right|_{Y}$ has finite orbit under
$$ \operatorname{Mod}_{g}=\pi_{1}\left(\mathscr{M}_{g}\right) ; $$hence by Theorem 3.1.5, $(\mathscr{E}, \nabla)$ has finite monodromy. $\square$
20. Geometric origin does not meet the stated Picard–Fuchs hypothesis
- ID:
ea753aa7-89f5-418b-aa96-9474a8bede14 - Refine score:
0.41 - Original types: general
- Refine status: open
Comment
Theorem 4.3.9 does not by itself establish the Picard–Fuchs hypothesis used in Theorem 4.3.7. Under Definition 2.3.1, a local system of geometric origin may be only a direct summand of $R^i\pi_*\mathbb{C}$, while Definition 4.2.7 requires the flat bundle itself to equal the full Gauss–Manin system. The connection drawn here therefore requires an additional result extending Theorem 4.3.7 to such summands or realizing these local systems as full Picard–Fuchs systems.
Quoted passage
In examples, the hypothesis that the flat bundle in question be a PicardFuchs equation does not seem unduly restrictive. For example, the following is a consequence of our discussion in § 2.3:
Theorem 4.3.9. Let $\left(C_{1}, \ldots, C_{n}\right)$ be an $n$-tuple of quasi-unipotent conjugacy classes in $\mathrm{SL}_{2}(\mathbb{C})$. Any finite orbit of the $\operatorname{PMod}_{0, n}$-action on $Y(\underline{C})^{\text {irr }}$ corresponds to a local system of geometric origin.
21. Universal cover does not ensure the claimed π₁ isomorphism
- ID:
5556f7b3-4be4-4c8e-b5a2-cf1eafb3bae4 - Refine score:
0.5 - Original types: general
- Refine status: open
Comment
In the proof of Corollary 4.4.3, passing to the universal cover of $S(\mathbb{C})^{\mathrm{an}}$ does not by itself make $\pi_1(X)\to\pi_1(\mathscr{X}_{\widetilde S})$ an isomorphism: the kernel may contain the image of the boundary map $\pi_2(\widetilde S)\to\pi_1(X)$. Thus the claimed global extension of the local system requires an additional argument. The proof could instead be formulated using the horizontal section of the relative Betti moduli space and locally defined fiberwise parallel transport, which is all the Noether–Lefschetz argument appears to require.
Quoted passage
Proof. Let $\widetilde{S}$ be the universal cover of $S(\mathbb{C})^{\text {an }}$, and let $\mathscr{X}_{\widetilde{S}}$ be the basechange of $\mathscr{X}(\mathbb{C})^{\text {an }}$ to $\widetilde{S}$. Choosing a lift $\widetilde{s} \in \widetilde{S}$ of $s$, the inclusion $X \rightarrow \mathscr{X}_{\widetilde{S}}$ as the fiber over $\widetilde{s}$ induces an isomorphism of fundamental groups. Hence we have a local system $\mathbb{V}$ on $\mathscr{X}_{\widetilde{S}}$ with monodromy the same as that of $(\mathscr{E}, \nabla)$.
22. Dolbeault moduli conditions in Example 5.1.1
- ID:
9cff27a4-2e16-40b7-b772-03fc98806500 - Refine score:
0.37 - Original types: general
- Refine status: open
Comment
The stated definition of $M_{\mathrm{Dol}}(X,r)$ omits the relevant vanishing conditions on rational Chern classes. For a higher-dimensional smooth projective $X$, semistability and degree zero alone do not define the Dolbeault moduli space corresponding under non-abelian Hodge theory to $M_B(X,r)$.
Quoted passage
We write this in a form analogous to that of the usual Hodge conjecture, above. Namely, we let $M_{\text {Dol }}(X, r)$ be the (coarse) moduli space of semistable Higgs bundles of degree zero. Here a Higgs bundle is a pair $(\mathscr{E}, \theta: \mathscr{E} \rightarrow$ $\mathscr{E} \otimes \Omega_{X}^{1}$ ) with $\theta$ an $\mathscr{O}_{X}$-linear map such that the natural composition
$$ \theta \circ \theta: \mathscr{E} \rightarrow \mathscr{E} \otimes \Omega_{X}^{2} $$is identically zero (see [Sim95, §6] for details). The map $\theta$ is referred to as a Higgs field.
[^13]There is a natural $\mathbb{C}^{\times}$-action on $M_{\text {Dol }}(X, r)$, given by scaling the Higgs field:
$$ \lambda \cdot(\mathscr{E}, \theta)=(\mathscr{E}, \lambda \theta) . $$Moreover there is a natural (real-analytic!) homeomorphism $M_{\text {Dol }}(X, r) \simeq$ $M_{B}(X, r)$ [Sim95, Theorem 7.18].
23. Integral points versus integral local systems
- ID:
36c277a5-a930-443c-8ded-618d648440b9 - Refine score:
0.29 - Original types: general
- Refine status: open
Comment
The notation $M_B(X,r)(\mathbb Z)$ is ambiguous here: ordinary $\mathbb Z$-points of a coarse GIT quotient need not coincide with complex local systems that admit a $\mathbb Z$-lattice. The displayed reformulation is valid only if this notation is understood as the image of genuine $\mathbb Z$-local systems in the complex Betti moduli space.
Quoted passage
Moreover there is a natural (real-analytic!) homeomorphism $M_{\text {Dol }}(X, r) \simeq$ $M_{B}(X, r)$ [Sim95, Theorem 7.18]. Now Simpson's [Sim97, Conjecture 12.4] may be rephrased as saying that the points of
$$ M_{B}(X, r)(\mathbb{Z}) \cap M_{\mathrm{Dol}}(X, r)^{\mathbb{C}^{\times}} $$correspond to local systems of geometric origin, where we make sense of the intersection here using the homeomorphism of the previous sentence.
24. Undefined Betti moduli in the Tate analogue
- ID:
64ee5786-ebb8-4c98-b840-52e11571b433 - Refine score:
0.35 - Original types: general
- Refine status: open
Comment
The notation $M_B(X,r)(\overline{\mathbb Q}_\ell)$ is not defined in this arithmetic setting. When $X$ may be over a positive-characteristic field, the relevant objects are continuous representations of the geometric étale fundamental group, not points of the previously defined Betti moduli space of a complex topological fundamental group. The displayed union is meaningful only after introducing an appropriate $\ell$-adic representation space or treating it explicitly as set-theoretic shorthand for isomorphism classes of continuous $\ell$-adic local systems.
Quoted passage
Conjecture 5.1.4 (Fontaine-Mazur-Petrov [Pet23, Conjecture 1 bis]). Let $\mathbb{V}$ be an irreducible $\ell$-adic local system on $X_{K^{s}}$. Then $\mathbb{V}$ is of geometric origin if and only if it its isomorphism class has finite orbit under $\operatorname{Gal}\left(K^{s} / K\right)$.
That is, the $\ell$-adic local systems on $X$ of geometric origin and rank $r$ are precisely
$$ \underset{K^{\prime} / K}{\lim } \mathrm{M}_{B}(X, r)\left(\overline{\mathbb{Q}_{\ell}}\right)^{\operatorname{Gal}\left(K^{s} / K^{\prime}\right)} . $$Our primary evidence for this conjecture comes from the case where $X$ is a curve and $K$ is finite, where the conjecture follows from work of Lafforgue [Laf02].
25. Puncture labels conflict in the case study
- ID:
aae66f8a-07aa-4f07-ad09-b8806f7cf0db - Refine score:
0.3 - Original types: general
- Refine status: open
Comment
In §5.2, the assignments of $C_3$ and $C_4$ conflict with $x_3=\infty$ and $x_4=\lambda$. Under the paper’s componentwise definition of $Y(\underline{C})$, the displayed indexing places $C_3$ at $\infty$ and $C_4$ at $\lambda$, whereas the prose places them at $\lambda$ and $\infty$, respectively.
Quoted passage
this time with $n=4$. By applying a fractional linear transformation we may assume
$$ x_{1}=0, x_{2}=1, x_{3}=\infty $$and set $x_{4}=\lambda$. Our goal will be to classify rank 2 local systems of geometric origin on $X$ with local monodromy in the conjugacy class
$$ C_{1}=C_{2}=C_{3}=\left[\left(\begin{array}{ll} 1 & 1 \\ 0 & 1 \end{array}\right)\right] $$at 0, 1, $\lambda$ and in the conjugacy class
$$ C_{4}=\left[\left(\begin{array}{cc} -1 & 1 \\ 0 & -1 \end{array}\right)\right] $$at $\infty$.
26. Missing dense-orbit hypothesis in Theorem 5.4.13
- ID:
6f3c4e0b-4348-4ad9-a111-ca9662b3f87e - Refine score:
0.35 - Original types: general
- Refine status: open
Comment
The proof sketch does not state the condition placing the unitary Galois conjugate in the dense-orbit regime, such as density of its image in $\mathrm{SU}(2)$. Unitarity alone is insufficient: integral finite-image representations are unitary but have finite mapping class group orbits. The intended Shimura-curve representation may satisfy the stronger condition, but that essential property is not established in the sketch.
Quoted passage
From the main result of [PX02b, Theorem 1.4], taking the boundary to be empty, and the Zariski-density of the unitary locus, it is enough to produce one $\pi_{1}(X)$-representation defined over $\overline{\mathbb{Z}}$ with a unitary Galois conjugate-the orbit of this representation under the mapping class group of $X$ will be Zariski dense. But now the tautological local system on any compact Shimura curve (or étale cover thereof) of genus $g$ suffices (and such exist for all $g \geq 2$, as one may take étale covers of a compact Shimura curve of genus 2). $\square$
27. Basepoint stabilizer needed in the Putman–Wieland setup
- ID:
5ca204f1-ec07-4b41-ac41-5ec12dcd55df - Refine score:
0.31 - Original types: general
- Refine status: open
Comment
The asserted action on $\pi_1(\Sigma_{g,n},x_0)$ requires restricting $\operatorname{Mod}_{g,n+1}$ to the finite-index subgroup preserving the distinguished extra point $x_0$. Under the paper’s convention, the full mapping class group may permute all $n+1$ punctures and therefore does not act on this fixed based fundamental group. The stabilizer $\Gamma$ should consequently be understood inside the point-preserving subgroup; this is a local setup issue and does not affect the ensuing Putman–Wieland statements once that restriction is made.
Quoted passage
Let $\Sigma_{g, n}$ be an orientable surface of genus $g$ with $n$ punctures. Fixing a base-point $x_{0}$ in $\Sigma_{g, n}$, there is a natural action of $\operatorname{Mod}_{g, n+1}$ on $\pi_{1}\left(\Sigma_{g, n}, x_{0}\right)$; hence if $\Sigma_{g^{\prime}} \rightarrow \Sigma_{g}$ is a cover branched at $n$ points, a finite index subgroup of $\operatorname{Mod}_{g, n+1}$ naturally acts on $H_{1}\left(\Sigma_{g^{\prime}}, \mathbb{Z}\right)$. Indeed, let $\Sigma_{g^{\prime}, n^{\prime}}$ be the complement of the ramification points in $\Sigma_{g^{\prime}}$; then $\pi_{1}\left(\Sigma_{g^{\prime}, n^{\prime}}\right)$ is a subgroup of $\pi_{1}\left(\Sigma_{g, n}\right)$, and hence admits a natural action by its stabilizer $\Gamma$ in $\operatorname{Mod}_{g, n+1}$, a finite index subgroup.
28. Branched Putman–Wieland implication needs an extension
- ID:
8bc489cf-6937-46d4-9b28-f3a2411f80d7 - Refine score:
0.37 - Original types: general
- Refine status: open
Comment
The deduction of the branched Putman–Wieland conjecture is immediate from Proposition 6.3.2 only when $n=0$, given the standing assumption there that $q$ is proper. For $n>0$, the argument requires the corresponding logarithmic/parabolic statement for $W^1R^1q_*\mathbb{U}$, with generic generation formulated using $\widehat{\mathscr{E}}_0\otimes\omega_{\bar X}(D)$. The text appears to rely on this unstated extension, so its availability and applicability should be made explicit.
Quoted passage
Taking $\rho$ to have finite monodromy, these conjectures imply the PutmanWieland conjecture (Conjecture 6.2.1), by Proposition 6.3.2. More generally, they would give some evidence for Conjecture 6.1.2:
Proposition 6.3.6. Suppose $g \geq 3$ and assume Conjecture 6.3.4. Let
$$ \rho: \operatorname{PMod}_{g, n+1} \rightarrow U(r) $$be a representation whose restriction to the point-pushing subgroup $\pi_{1}\left(\Sigma_{g, n}\right) \subset \operatorname{PMod}_{g, n+1}$ is irreducible. Then $\rho$ is cohomologically rigid.
29. Rank dependence in Conjecture 6.3.17 is unclear
- ID:
5134841b-6ea2-436e-a93e-63f3590fc976 - Refine score:
0.32 - Original types: general
- Refine status: open
Comment
The quantifiers in Conjecture 6.3.17 do not determine whether the non-decreasing bound is universal over ranks and monodromy representations or may depend on the previously fixed bundle $\mathscr E$. This distinction is substantive: a genus-only $f(g)$ gives rank-uniform bounds, whereas the stated evidence produces a rank-sensitive $f(g,r)$ and establishes degree-one vanishing only when $g\geq 2+2\operatorname{rk}(\mathscr E)$.
Quoted passage
Conjecture 6.3.17 (Horizontal generic vanishing). Let $g \geq 3$, and let $(X, D)$ be a marked smooth projective curve of genus $g \geq 3$. Let $(\mathscr{E}, \nabla)$ be a flat bundle on $X$ with irreducible, non-trivial unitary monodromy, and regular singularities along $D$, whose residue matrices have eigenvalues with real parts in $[0,1) .{ }^{20}$ There exists some non-decreasing function $f$, with $f(3)=1$, such that after isomonodromic deformation to a general nearby curve $X^{\prime}$, we have:
(1) (weak form) $H^{0}\left(X^{\prime}, \mathscr{E}(Z)\right)=0$ for a general effective divisor $Z$ on $X^{\prime}$ of degree $d \leq f(g)$. (2) (strong form) $H^{0}\left(X^{\prime}, \mathscr{E}(Z)\right)=0$ for all effective divisors $Z$ on $X^{\prime}$ of degree $d \leq f(g)$.
30. Maximality makes Question 6.4.3 potentially vacuous
- ID:
9e4dc3c3-9589-4814-b5d5-0950fa63de6b - Refine score:
0.34 - Original types: general
- Refine status: open
Comment
The maximality condition in Question 6.4.3 does not presently distinguish the special invariant loci under discussion. Under ordinary inclusion-theoretic maximality, an invariant irreducible subvariety is contained in an invariant irreducible component of $Y(g,n,r)$, so the maximal candidates are generally just ambient components. The intended class of proper or otherwise distinguished invariant subvarieties therefore remains unspecified.
Quoted passage
6.4.2. Invariant subvarieties. In § 2, § 3, and §4, we studied the finite orbits of $\pi_{1}(\mathscr{M})$ on $Y(g, n, r)$. What about higher-dimensional invariant subvarieties? A natural (imprecise) expectation in the case $\mathscr{M}=\mathscr{M}_{g, n}$, analogous to Conjecture 6.1.6, is that for $g \geq 3$, any such subvariety should all be motivic, in the sense of §5.5.
Question 6.4.3 (Imprecise). Let $Z \subset Y(g, n, r)$ be a maximal irreducible subvariety stable under the action of a finite index subgroup of $\operatorname{Mod}_{g, n}$. Then is $Z$ "of geometric origin" for any complex structure on $\Sigma_{g, n}$ ?
31. Conjecture 6.4.6 gives only half the claimed specialization
- ID:
214cc504-8506-43ae-885e-59a7d1a1b7bf - Refine score:
0.24 - Original types: general
- Refine status: open
Comment
The stated specialization to Conjecture 4.3.4 accounts explicitly only for the implication from an integral formal isomonodromic deformation to a finite orbit. Since Conjecture 4.3.4 is formulated as an equivalence, the claim that Conjecture 6.4.6 specializes to the full statement requires the converse—presumably the comparatively formal algebraic-leaf-implies-integrality direction—to be made explicit.
Quoted passage
Conjecture 6.4.6. Let $\mathscr{X} \rightarrow S$ be a smooth proper morphism over a finitely-generated integral $\mathbb{Z}$-algebra $R, s \in S$ an $R$-point, and $Z \subset$ $\mathscr{M}_{d R}(\mathscr{X} / S, r)_{s}$ a closed substack. Then $Z(\mathbb{C})$ is invariant under a finite index subgroup of $\pi_{1}(S, s)$ if its formal isomonodromic deformation has an integral model.
This is meant to be the higher-dimensional analogue of Conjecture 4.3.4; it specializes to that statement if $Z$ is a point. It is arguably the non-abelian analogue of [Kat82, Conjecture 9.2], which aims to characterize the identity component of the Zariski-closure of the monodromy group of an ODE in terms of its $p$-curvatures.
32. Torelli containment does not follow from Torelli alone
- ID:
793481c1-ed16-4b11-bc3b-02747a2249c6 - Refine score:
0.34 - Original types: general
- Refine status: open
Comment
The restriction on Torelli containment requires more than the Torelli theorem alone: finite homological monodromy must first be converted into constancy of the pulled-back variation using a fixed-part argument. In addition, the assertion that every geometric subgroup has infinite monodromy needs an infinite or non-isotrivial qualification, since isotrivial families can produce nontrivial finite geometric subgroups.
Quoted passage
Question 6.4.8. What are the geometric subgroups of $\operatorname{Mod}_{g, n}$ ? There are some evident restrictions on such subgroups. For example, by the Torelli theorem, non-trivial geometric subgroups of $\operatorname{Mod}_{g, n}$ cannot be contained in the Torelli group. Indeed, if $\mathbb{V}$ is any variation of Hodge structure on $\mathscr{M}_{g, n}$ with quasi-finite period map, then $\mathbb{V}$ yields an analogous restriction: the restriction of $\mathbb{V}$ to any geometric subgroup of $\operatorname{Mod}_{g, n}$ must have infinite monodromy. Moreover any variation of Hodge structure whatsoever on $\mathscr{M}_{g, n}$ must have semisimple monodromy when restricted to a geometric subgroup (as variations of Hodge structure are always semisimple).
Scope
- Paper:
03 Published and Submitted Work/Published/P01_Litt_Motives_Mapping_Class_Groups_and_Monodromy.pdf - Refine report:
.refine/results/Published/P01_Litt_Motives_Mapping_Class_Groups_and_Monodromy.review.json - Rubric:
rubric/comment_triage_rubric.md - Version assessed: the local published PDF
- Detailed Refine comments assessed: 32
- Assessment date: 2026-07-30
The unanchored feedback.overall material was used only as context. Suspicious notation in Examples 2.2.4 and 4.2.4, Theorem 2.2.10, and Section 5.2 was checked against rendered PDF pages.
Summary
| # | Short title | Validity | Category | Standardness | Impact | Challenge | Repair | Disposition | Priority | Confidence |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Connected fibers | V4 | C5 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 2 | Rigid locus shifts | V4 | C5 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 3 | Middle-convolution category | V4 | C5 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 4 | Special-linear target | V4 | C5 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 5 | Braid/mapping-class group | V4 | C6 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 6 | Hypergeometric twists | V4 | C9 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 7 | Positive Markoff triples | V4 | C5 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 8 | Undefined parameters | V4 | C5 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 9 | Classification scope | V4 | C9 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 10 | \(PSL_2(7)\) scalar extension | V4 | C7 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 11 | Geometric-origin equivalence | V4 | C9 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 12 | Missing “finite orbit” | V4 | C1 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 13 | Orientation and determinant | V4 | C6 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 14 | Logarithmic residue quotient | V4 | C4 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 15 | Schlesinger gauge | V4 | C3 | E2 | I1 | Q1 | R1 | D1 | P3 | HIGH |
| 16 | Traceless adjoint | V4 | C6 | E3 | I2 | Q2 | R2 | D3 | P1 | HIGH |
| 17 | Flat Atiyah splitting | V4 | C5 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 18 | Meaning of reduction mod \(p\) | V3 | C3 | E2 | I1 | Q0 | R1 | D1 | P3 | HIGH |
| 19 | Wrong conjecture invoked | V4 | C6 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 20 | Summand versus Picard-Fuchs | V4 | C9 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 21 | Universal-cover \(\pi_1\) | V4 | C6 | E3 | I2 | Q2 | R2 | D3 | P2 | HIGH |
| 22 | Dolbeault Chern classes | V4 | C5 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 23 | Integral-moduli notation | V4 | C3 | E2 | I1 | Q0 | R1 | D1 | P3 | HIGH |
| 24 | Arithmetic Betti notation | V4 | C4 | E-NA | I1 | Q0 | R1 | D1 | P3 | HIGH |
| 25 | Puncture labels | V4 | C4 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 26 | Dense Shimura conjugate | V4 | C3 | E2 | I1 | Q1 | R1 | D1 | P3 | HIGH |
| 27 | Basepoint stabilizer | V4 | C5 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 28 | Logarithmic PW extension | V4 | C3 | E3 | I1 | Q1 | R1 | D1 | P3 | HIGH |
| 29 | Dependence of \(f\) | V4 | C3 | E2 | I1 | Q0 | R1 | D1 | P3 | HIGH |
| 30 | Meaning of maximality | V4 | C3 | E2 | I1 | Q0 | R1 | D1 | P3 | HIGH |
| 31 | Only one implication | V4 | C9 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 32 | Torelli qualification | V4 | C9 | E2 | I2 | Q0 | R2 | D3 | P2 | HIGH |
No issue is classified at I3 or above. Items 16 and 21 initially appeared serious; separate independent challenges verified both repairs.
Detailed assessments
1. Connected fibers in the introductory exact sequence
Comment ID: 1393706f-f6b5-4ebe-a153-018c2f752338 Location: PDF p. 2.
Smoothness and properness do not alone make \(\pi_1(X)\to\pi_1(S)\) surjective. Add geometrically connected fibers (and the usual connectedness/basepoint assumptions). The later family setup does impose connected fibers, so no result depends on the broader introductory wording. Classification: V4/C5/I2/R2/D3.
2. The uniqueness locus changes from \(Y(\underline C)^{\rm irr}\)
Comment ID: d8e9baf8-e3f6-46e9-ab6c-e06a5df3f81e Location: PDF p. 5.
The question and the definition on the next page concern the irreducible locus, whereas the explanatory sentence says \(Y(\underline C)\) is a singleton. Restore the superscript irr. This is a meaning-bearing local scope error: V4/C5/I2/R2/D3.
3. Middle convolution is not an autoequivalence of all representations
Comment ID: a16f3718-5361-49d1-b323-91f6a6a01b8a Location: PDF pp. 6 and 17.
With the later cohomological definition, middle convolution by a nontrivial Kummer character kills exceptional objects such as the trivial rank-one system. The preservation and quasi-inverse statements require Katz's nonexceptional subcategory (or the corresponding perverse-sheaf quotient); the rank-reduction use concerns the permitted irreducible objects. Classification: V4/C5/I2/R2/D3.
4. Middle convolution does not automatically preserve determinant one
Comment ID: 22519809-6841-4822-921c-fc4d00afb297 Location: PDF pp. 6-7.
The output local classes naturally lie in \(\mathrm{GL}_{r'}\); product one does not force every local determinant to be one. Replace the displayed \(\mathrm{SL}_{r'}\) target by \(\mathrm{GL}_{r'}\), or include the rank-one normalization that restores determinant one. Classification: V4/C5/I2/R2/D3.
5. The Artin presentation is not \(\operatorname{Mod}_{0,n}\)
Comment ID: 69e1dbf7-a9b8-4ef8-b448-2f7c7c73c2b3 Location: PDF pp. 7-8.
The displayed presentation is \(B_n\), and \(B_n/Z(B_n)\) is the mapping class group of an \((n+1)\)-punctured sphere fixing one puncture, not the full \(\operatorname{Mod}_{0,n}\). The latter requires the additional sphere relations. The Hurwitz outer action on product-one tuples does descend to the sphere mapping class group, so the later dynamics survive. Classification: V4/C6/I2/R2/D3.
6. Hypergeometric monodromy requires a rank-one normalization
Comment ID: af4c43a3-b5dd-4043-ba49-4fd186bea473 Location: PDF p. 8.
A raw Gauss \({}_2F_1\) system has a normalized eigenvalue \(1\) at two punctures. General irreducible rank-two three-point systems are hypergeometric only up to a rank-one twist and permutation/normalization of local exponents. Replace “precisely” by that qualified statement. Classification: V4/C9/I2/R2/D3.
7. Markoff transitivity needs positivity
Comment ID: f69db39e-1213-458b-b31c-47eb0f049b30 Location: PDF p. 9.
\((0,0,0)\) is a fixed integral solution and is not in the orbit of \((1,1,1)\). State the classical result for positive Markoff triples (with the conventional permutation/sign qualifications). Classification: V4/C5/I2/R2/D3.
8. Example 2.2.4 divides by a possibly zero parameter
Comment ID: 29719079-adae-4483-93d3-8662de09780b Location: PDF p. 10.
The rendered matrix \(A_1\) contains \(1/x_1\). For \(\alpha=\beta=\tfrac12\), \(x_1=0\), so the displayed representative is undefined. Add \(x_1\ne0\) and treat excluded parameters by a different representative or limit. Classification: V4/C5/I2/R2/D3.
9. The classification sentence omits Theorem 2.2.8's range
Comment ID: baabcd09-dda4-4ba5-927d-33312c5df85c Location: PDF pp. 12-13.
The theorem classifies interesting orbits only when some \(A_i\) has infinite order; the all-finite-order case is explicitly left open on the following page. Carry that qualifier into “we have classified.” Classification: V4/C9/I2/R2/D3.
10. The Shephard-Todd group is a scalar extension of \(PSL_2(7)\)
Comment ID: e474222f-8d0b-4765-8e16-1ad0399b304b Location: PDF p. 13.
The rendered page identifies the natural three-dimensional \(PSL_2(\mathbb F_7)\)-image itself as a reflection group. Its image lies in \(\mathrm{SL}_3\), while a nonidentity pseudoreflection has nontrivial determinant. The exceptional reflection group is the corresponding scalar extension (Shephard-Todd \(G_{24}\)), whose projective quotient is \(PSL_2(7)\). Classification: V4/C7/I2/R2/D3.
11. Finite orbit is not immediately the same as geometric origin
Comment ID: 310ebc20-1fa3-4c9e-98b9-46c70d2211aa Location: PDF pp. 14-16.
Proposition 2.3.3 gives extension over a dominant family. The subsequent Corlette-Simpson dichotomy has a pullback branch and a rigid/geometric-origin branch; only after the pullback and degenerate cases are separated does the remaining classification become Question 2.3.2. Qualify the advance claim. Classification: V4/C9/I2/R2/D3.
12. Question 2.4.1 is missing the word “finite”
Comment ID: 167946db-6357-440b-80e3-2d2bee8e60a7 Location: PDF p. 18.
The intended object is a tuple “with finite \(\operatorname{Mod}_{0,n}\)-orbit.” As printed, every tuple merely “has an orbit.” Classification: V4/C1/I2/R2/D3; tag meaning_changing_typo.
13. Determinant does not detect orientation for even genus
Comment ID: 30c73e27-b93f-4526-be88-163066de2848 Location: PDF p. 19.
Orientation-preserving mapping classes act symplectically and reversing ones anti-symplectically. An anti-symplectic operator on rank \(2g\) has determinant \((-1)^g\), which is \(1\) when \(g\) is even. Replace the determinant parenthesis by preservation of the intersection form. Classification: V4/C6/I2/R2/D3.
14. The residue trivializes a quotient, not the full logarithmic fiber
Comment ID: 5c0514b5-2c2e-4a68-9361-cff6fbaeaca9 Location: PDF p. 22.
In dimension greater than one, the fiber of \(\Omega_X^1(\log D)\) also contains tangential cotangent directions. The canonical object is \(\Omega_X^1(\log D)/\Omega_X^1\simeq\mathscr O_D\), whose generator is the residue class of \(dz/z\). The subsequent composite uses this correct quotient. Classification: V4/C4/I2/R2/D3.
15. Schlesinger equations require a standard gauge normalization
Comment ID: 35f587a7-81ba-442c-b89d-5abe8a77797f Location: PDF p. 24.
The \(C_i\,dx_i\) terms contribute to curvature. Restricting the flat connection to the section \(z=\infty\) gives a flat base connection; on the local parameter polydisk it can be gauge-trivialized, setting the \(C_i\) to zero. The usual coefficient calculation then gives Schlesinger's equations. This is a verified standard normalization, not a proof error: V4/C3/E2/I1/Q1/R1/D1.
16. The rigidity argument must use the traceless adjoint
Comment ID: 7e5c68a7-f8b7-4289-a63e-acdbd8c8b5f6 Location: PDF pp. 26-28.
For the full endomorphism system the rank is \(r^2\), so the strict rank-\(<g\) hypothesis of Theorem 3.3.1 fails when \(g=r^2\), and scalar determinant deformations remain.
Severity challenge (Q2, independently verified): Use the cited LL24/Corollary 2.3.5 extension, which has finite determinant on the total space, and carry out Mochizuki's deformation with this determinant fixed. The relevant tangent system is \(\operatorname{End}^0(\mathbb V')\), of rank \(r^2-1<g\). Fiberwise irreducibility gives \(\pi_*\operatorname{End}^0(\mathbb V')=0\). After the dominant étale base change, Lemma 2.4.2 writes \(\mathbb V'=\mathbb U\otimes\pi^*\mathbb L\), so \(\operatorname{End}^0(\mathbb V')=\operatorname{End}^0(\mathbb U)\) is unitary on the total space and Theorem 3.3.1 kills the fixed-determinant invariant tangent space. Leray then gives \(H^1(\operatorname{End}^0(\mathbb V'))=0\).
Scalar twists are exactly the failure mode of the uncorrected full-adjoint argument; fixing determinant removes their infinitesimal directions, while residual \(r\)-torsion scalar twists are discrete. Nonzero obstruction spaces cannot create a deformation when the first nontrivial small-extension stage would require an \(H^1\)-class. The isolation and integrality consequences therefore survive.
Final classification: V4/C6/E3/I2/Q2/R2/D3, HIGH.
17. An Atiyah splitting is not automatically flat
Comment ID: e0d7ddb3-ba3d-4fa5-8c08-1b73505001d1 Location: PDF pp. 30-31.
An \(\mathscr O_X\)-linear splitting of the Atiyah sequence is a connection. It is flat precisely when the splitting preserves Lie brackets, equivalently has zero curvature. The next page correctly uses flatness to obtain a map of complexes. Add the bracket/curvature condition. Classification: V4/C5/I2/R2/D3.
18. “Mod \(p\)” is clarified by the following equivalence
Comment ID: 62aae96b-5229-4800-a97e-0d97634c89e9 Location: PDF p. 32.
Visual inspection confirms \(a\in\overline{\mathbb Q}\). Vanishing in a single residue field only says the residue of \(a\) lies in \(\mathbb F_p\), whereas vanishing in \(\mathscr O_K/p\mathscr O_K\) gives complete splitting. The immediately following “if and only if \(p\) splits completely” makes the intended global interpretation recoverable, so this is an ambiguity rather than a false argument. Classification: V3/C3/E2/I1/R1/D1.
19. Proposition 4.3.5 invokes the wrong formulation
Comment ID: 2a514e2f-fd0f-4122-a911-cf71fb84ae63 Location: PDF p. 36.
The hypothesis is Conjecture 4.3.1, including finite-order \(p\)-integrality, while Conjecture 4.3.4 explicitly elides that condition. Invoke Conjecture 4.3.1 (or its unelided precise version) and place the pullback \((\mathscr E,\nabla)|_Y\), not the original object on \(X\), in the genus-\(g\) moduli space. Classification: V4/C6/I2/R2/D3.
20. Geometric origin does not imply “full Picard-Fuchs”
Comment ID: ea753aa7-89f5-418b-aa96-9474a8bede14 Location: PDF pp. 36-37.
Definition 2.3.1 allows a direct summand of \(R^i\pi_*\mathbb C\); Definition 4.2.7 requires equality with the full Gauss-Manin system. Remark 4.2.9 itself notes that the summand case is not known in general. Therefore Theorem 4.3.9 does not by itself show that Theorem 4.3.7's hypothesis is mild. Qualify the example or cite a special realization theorem. Classification: V4/C9/I2/R2/D3.
21. The universal-cover base change need not preserve \(\pi_1\)
Comment ID: 5556f7b3-4be4-4c8e-b5a2-cf1eafb3bae4 Location: PDF pp. 37-38.
For the fibration over \(\widetilde S\), the homotopy sequence contains
so simple connectedness of the base does not make the middle map an isomorphism.
Severity challenge (Q2, independently verified): The proof does not need a local system on the total space. The local system of relative Betti character varieties trivializes over \(\widetilde S\), so the isomonodromic point defines a horizontal section \(\sigma\). The \(\pi_2(\widetilde S)\)-boundary changes based transport only by inner automorphism, which is invisible on character variety points even when it obstructs a total-space representation.
Pull back Simpson's closed relative nonabelian Hodge locus along \(\sigma\). Formal Griffiths-transverse extension and Artin approximation make the pulled-back locus open near the starting point; it is also closed analytic, hence equals connected \(\widetilde S\). Deck transformations carry its values to the mapping-class-group orbit of the initial representation. Deligne finiteness then gives the required finite orbit; \(\sigma\) need not descend or be deck-invariant.
Final classification: V4/C6/E3/I2/Q2/R2/D3, P2, HIGH.
22. The Dolbeault moduli space needs vanishing Chern classes
Comment ID: 9cff27a4-2e16-40b7-b772-03fc98806500 Location: PDF p. 40.
For higher-dimensional smooth projective \(X\), nonabelian Hodge theory matches the Betti moduli space with polystable Higgs bundles having vanishing rational Chern classes, not merely semistable bundles of degree zero. Add polystability and the Chern-class conditions. Classification: V4/C5/I2/R2/D3.
23. \(M_B(X,r)(\mathbb Z)\) needs an explicit convention
Comment ID: 36c277a5-a930-443c-8ded-618d648440b9 Location: PDF p. 41.
Ordinary integral points of a coarse GIT quotient need not equal complex local systems admitting a \(\mathbb Z\)-lattice. The preceding Conjecture 5.1.2 makes the intended meaning clear. Define the notation as the image of genuine \(\mathbb Z\)-local systems. Classification: V4/C3/E2/I1/R1/D1.
24. The arithmetic \(M_B\) notation is only shorthand
Comment ID: 64ee5786-ebb8-4c98-b840-52e11571b433 Location: PDF p. 41.
For an arithmetic variety there is no previously defined complex Betti moduli space whose \(\overline{\mathbb Q}_\ell\)-points are the continuous étale representations in question. The preceding sentence supplies the intended set. Define an \(\ell\)-adic representation space or call the display set-theoretic shorthand. Classification: V4/C4/I1/R1/D1.
25. The puncture numbering conflicts with the conjugacy classes
Comment ID: aae66f8a-07aa-4f07-ad09-b8806f7cf0db Location: PDF p. 44.
The rendered page sets \(x_3=\infty\), \(x_4=\lambda\), but places \(C_3\) at \(\lambda\) and \(C_4\) at infinity. Swap the labels \(x_3,x_4\), or the corresponding \(C_3,C_4\) indexing, consistently. Classification: V4/C4/I2/R2/D3.
26. The Shimura conjugate must have dense compact image
Comment ID: 6f3c4e0b-4348-4ad9-a111-ca9662b3f87e Location: PDF p. 50.
Unitarity alone does not imply a dense mapping-class orbit. For the tautological system on a compact Shimura curve, the non-split real Galois conjugate has dense image in the relevant compact \(\mathrm{SU}(2)\) factor (by the standard irreducibility/strong-approximation property of the arithmetic lattice). This supplies the hypothesis needed from the cited dynamics theorem. Add that sentence. Classification: V4/C3/E2/I1/Q1/R1/D1.
27. The based action requires the point stabilizer
Comment ID: 5ca204f1-ec07-4b41-ac41-5ec12dcd55df Location: PDF p. 54.
Under the paper's convention, the full \(\operatorname{Mod}_{g,n+1}\) may permute the distinguished \(x_0\). Restrict to its finite-index point stabilizer (or to the pure mapping class group) before acting on \(\pi_1(\Sigma_{g,n},x_0)\), and define \(\Gamma\) inside it. The subsequent finite-index statements are unchanged. Classification: V4/C5/I2/R2/D3.
28. The branched Putman-Wieland deduction uses the logarithmic version
Comment ID: 8bc489cf-6937-46d4-9b28-f3a2411f80d7 Location: PDF pp. 57-58.
Proposition 6.3.2 is written after assuming \(q\) proper. For \(n>0\), apply the same period-map argument to \(W^1R^1q_*\mathbb U\); the Hodge bundle is the relevant Deligne/parabolic extension and the adjoint multiplication map uses \(\widehat{\mathscr E}_0\otimes\omega_{\bar X}(D)\), exactly the bundle in Conjecture 6.3.4. This standard logarithmic extension proves the claimed implication. Classification: V4/C3/E3/I1/Q1/R1/D1.
29. Conjecture 6.3.17 should state what \(f\) may depend on
Comment ID: 5134841b-6ea2-436e-a93e-63f3590fc976 Location: PDF pp. 62-63.
Because \((\mathscr E,\nabla)\) is fixed before \(f\) is quantified, formal logic permits dependence on it, while the notation \(f(g)\) suggests a genus-only universal function. The evidence immediately uses \(f(g,r)\). State the intended dependence explicitly. Classification: V4/C3/E2/I1/R1/D1.
30. “Maximal” is vacuous without a properness convention
Comment ID: 9e4dc3c3-9589-4814-b5d5-0950fa63de6b Location: PDF p. 65.
Under ordinary inclusion-maximality, an invariant irreducible subvariety lies in an invariant ambient component, so the maximal objects are generally those components. The question is already labelled “Imprecise”; specify maximal proper invariant subvariety, maximal in a chosen class, or another intended notion. Classification: V4/C3/E2/I1/R1/D1.
31. Conjecture 6.4.6 gives only one direction for a point
Comment ID: 214cc504-8506-43ae-885e-59a7d1a1b7bf Location: PDF p. 66.
Conjecture 6.4.6 states integral formal deformation \(\Rightarrow\) finite orbit, while Conjecture 4.3.4 is an equivalence. Say that it specializes to one direction, or add the converse (a finite-orbit algebraic leaf spreads out and its formal completion is integral after inverting finitely many elements). Classification: V4/C9/I2/R2/D3.
32. The Torelli restriction needs fixed part and an isotrivial qualifier
Comment ID: 793481c1-ed16-4b11-bc3b-02747a2249c6 Location: PDF p. 66.
Finite homological monodromy becomes trivial after a finite cover. The theorem of the fixed part then makes the full \(H^1\)-variation constant, and Torelli forces the underlying family of curves to be isotrivial. Thus the conclusion applies to non-isotrivial underlying curve families, not to every nontrivial geometric subgroup; motion of marked points and other isotrivial families must be excluded. Classification: V4/C9/E2/I2/R2/D3.
Proposed correction queue
Subject to author confirmation and the independent challenges for items 16 and 21, this report yields 24 likely living-errata entries and 8 optional clarifications. The optional items are the Schlesinger gauge, the convention for reduction modulo \(p\), the two moduli-space notations, the dense compact Shimura conjugate, the logarithmic Putman-Wieland extension, the dependence of \(f\), and the intended maximality convention.
P16 ARITHMETIC REPRESENTATIONS OF FUNDAMENTAL GROUPS, II: FINITENESS14 detailed comments · 10 numbered corrections 2 I02 I110 I2
These errata refer to Daniel Litt, “Arithmetic representations of fundamental groups, II: finiteness,” Duke Mathematical Journal 170, no. 8 (2021), pp. 1851--1897, doi:10.1215/00127094-2020-0086. Page references below are to that published version.
Page 1859, Section 1.3. The finiteness theorem cited as [16, Theorem 2.1] concerns irreducible lisse sheaves, not arbitrary semisimple representations. In the sentence beginning “Work of Deligne, Drinfel'd, and Lafforgue,” replace “semisimple” by “irreducible.” Thus the sentence should read:
Work of Deligne, Drinfel'd, and Lafforgue implies (via automorphic methods) that if $X$ is a variety over a finite field $\mathbb F_q$, then the set of irreducible $\overline{\mathbb Q}_\ell$-representations of its Weil group $W(X)$, with fixed rank and bounded wild ramification at infinity, is finite up to twist by characters of $W(\mathbb F_q)$ (see [16, Theorem 2.1]).
This occurs only in the comparison with previous work; the paper's proof of Theorem 1.1.3 is unchanged.
Page 1865, Remark 3.1.2. The citation [15, Proposition 4.6] is not a valid pinpoint: item 4.6 of [15] is a definition and does not contain the asserted argument. Replace the remark by:
A similar argument appears in [14, Proposition 3.1].
Proposition 3.1.1 has its own proof, so this bibliographic correction has no effect on Corollary 3.1.3 or any later result.
Page 1875, proof of Lemma 4.1.3. The proof asserts a uniform finite order for the full reduction of an arbitrary formal automorphism and later substitutes a scalar function into an $N$-tuple-valued interpolation map. Replace the proof by the following scaled coordinate argument:
Choose a rational number $0<c'<\min\{c,\tfrac12\}$, and choose a totally ramified finite extension $\Lambda'/\Lambda$ containing an element $\varpi$ with $|\varpi|_\ell=\ell^{-c'}$. Let $F$ be the coordinate map on points induced by $\varphi$; explicitly,
\[ F(\mathbf a)= \bigl(\varphi(x_1)(\mathbf a),\ldots, \varphi(x_N)(\mathbf a)\bigr). \]On the closed unit ball define
\[ \widetilde F(\mathbf y)=\varpi^{-1}F(\varpi\mathbf y). \]The constant and nonlinear terms of $\widetilde F$ vanish modulo $\varpi$, while its linear term is invertible. Hence the reduction of $\widetilde F$ is an element of $\operatorname{GL}_N(\mathbb F_{\ell^r})$. Choose a uniform exponent $M_1$ for this finite group. A further integer $M_2$, depending only on $c'$, $\ell^r$, and $N$, makes
\[ H:=\widetilde F^{M_1M_2} \]satisfy the hypotheses of Lemma 4.1.1. Put $M=M_1M_2$, and let
\[ \vartheta(\mathbf y,m)\in \Lambda'\langle y_1,\ldots,y_N,m\rangle^N \]be the resulting analytic interpolation, so that $\vartheta(\mathbf y,m)=H^m(\mathbf y)$ for every $m\in\mathbb Z_{\geq0}$.
Let $\mathbf a=(z(x_1),\ldots,z(x_N))$. For $f\in\mathcal I$ and $0\leq j<M$, define the scalar analytic function
\[ h_{f,j}(m)= f\!\left( \varpi\,\vartheta\!\left( \varpi^{-1}F^j(\mathbf a),m \right) \right). \]Since $\varphi$ preserves $U_c(R)$ and $c'<c$, every argument in this formula lies in the closed unit ball. Moreover,
\[ h_{f,j}(m)=0 \quad\Longleftrightarrow\quad \varphi^{j+Mm}(z)\in V(f). \]Each $h_{f,j}$ either has finitely many zeros in $\mathbb Z_{\geq0}$ or vanishes identically. It follows that the return-time set to $V(f)$ is semilinear with period $M$. Finally,
\[ \{m\geq0:\varphi^m(z)\in V(\mathcal I)\} =\bigcap_{f\in\mathcal I} \{m\geq0:\varphi^m(z)\in V(f)\}, \]and an arbitrary intersection of semilinear sets with the same period $M$ is again semilinear with period $M$. This proves the lemma.
The integer $M$ still depends only on $c$, $\ell^r$, and $N$. Thus Corollaries 4.1.5 and 4.1.6, Theorem 1.1.3, and Corollary 1.1.5 retain their stated conclusions.
Page 1876, first paragraph of the proof of Corollary 4.1.5. The printed presentation uses the absolute cotangent dimension, which also counts the class of $\ell$, and unnecessarily asserts that its kernel lies in $\mathfrak m_S^2$. Replace the paragraph through the construction of the lift of $\varphi$ by:
Let $\mathfrak m_R$ be the maximal ideal of $R$, and put
\[ N=\dim_{\mathbb F_{\ell^r}} \frac{\mathfrak m_R}{\mathfrak m_R^2+\ell R}. \]Choose a relative Cohen presentation
\[ S=\Lambda[[x_1,\ldots,x_N]]\twoheadrightarrow R \]with kernel $\mathcal J$. No condition $\mathcal J\subseteq\mathfrak m_S^2$ is needed: the ideal $\mathcal J$ still cuts out the rigid generic fibre of $R$ as a closed analytic subspace of the open unit ball. Lift $\varphi$ to an endomorphism $\widetilde\varphi$ of $S$. Its linear part on the relative cotangent space is invertible, so the formal inverse function theorem shows that $\widetilde\varphi$ is an automorphism.
The remainder of the proof, with the geometric point ideal specified in the next correction, gives the same uniform-period conclusion.
Page 1876, final paragraph of the proof of Corollary 4.1.5. The contracted kernel of $S\to R\xrightarrow{z}L$ need not isolate the chosen geometric point. Replace the paragraph beginning “Now let $z:R\to L$” by:
Let $z:R\to L\subset\mathbb C_\ell$ be a $\varphi$-periodic point of $U$, and regard it as a $\widetilde\varphi$-periodic point of the open unit ball. Apply Lemma 4.1.3 to the ideal
\[ \mathcal I_z= \bigl(x_1-z(x_1),\ldots,x_N-z(x_N)\bigr) \subset\mathcal O_{\mathbb C_\ell}[[x_1,\ldots,x_N]]. \]Its zero locus is exactly $\{z\}$. If $z$ has exact period $q$, its return-time set is $q\mathbb Z_{\geq0}$. Since this set is semilinear with the uniform period $M$ furnished by Lemma 4.1.3, one has $q\mid M$, and therefore $\varphi^M(z)=z$.
This completes Corollary 4.1.5 and leaves Corollary 4.1.6 and the finiteness argument unchanged.
Pages 1880--1881, Definition 5.1.3. The recursive sum includes the terms $(i,j)=(0,m)$ and $(m,0)$ and is therefore circular. Replace its last display by
\[ W^{-m}S_\rho= \sum_{\substack{i+j=m\\ i,j\geq1}} (W^{-i}S_\rho)(W^{-j}S_\rho) \qquad\text{for }m>2. \]This is the multiplicative recursion used in Remark 5.1.4 and throughout the rest of Section 5, so no later statement changes.
Page 1883, Step 2 in the proof of Lemma 5.1.5. The displayed surjection and the monomial relations that follow are statements about character lattices, not cocharacter lattices. Replace the sentence beginning “The inclusion $T\hookrightarrow D$” by:
The inclusion $T\hookrightarrow D$ induces a surjection on character lattices
\[ X^*(D)\twoheadrightarrow X^*(T) \]with kernel $K$; the torus $T$ is precisely the subtorus of $D$ cut out by the characters in $K$.
The subsequent monomial calculation is already the corresponding character-lattice calculation, and the conclusion of Lemma 5.1.5 is unchanged.
Page 1884, Theorem 5.1.8. The theorem uses the positive index $i$, although the nonzero graded pieces of $S_\rho$ are indexed by $-i$. Replace its concluding sentence by:
Then, for $\alpha\in\mathbb Z_\ell^\times$ sufficiently close to $1$, there exists $\sigma_\alpha\in G_k$ such that, for every $i\geq0$, $\sigma_\alpha$ acts on $\operatorname{gr}_W^{-i}S_\rho$ via $\alpha^i\operatorname{Id}$.
The proof on pages 1887--1890 and Lemma 5.2.2 already use this indexing. The construction used in Theorem 1.1.11 is unchanged.
Page 1887, Step 1 in the proof of Theorem 5.1.8. The ordinary symmetric algebra in the displayed isomorphism is not complete, whereas $S_\rho$ is a complete local algebra. Replace the display and the sentence introducing it by:
Such a splitting extends continuously to a $\sigma_\alpha^{-1}$-equivariant isomorphism of complete local algebras
\[ \widehat{\operatorname{Sym}}_{\mathbb Q_\ell} \!\left(\mathfrak m_\rho/\mathfrak m_\rho^2\right) \xrightarrow{\sim}S_\rho, \qquad \widehat{\operatorname{Sym}}(V) :=\prod_{j\geq0}\operatorname{Sym}^j(V). \]The isomorphism respects the weight filtrations. All subsequent eigenvalue and filtration calculations are degreewise and therefore remain unchanged.
Pages 1892--1893, proof of Theorem 1.1.11 and Remark 5.3.2. The $\alpha$- and $\alpha^2$-eigenspaces need not split the integral cotangent lattice, so an integral eigenbasis and an integral linear change of coordinates cannot be assumed. Moreover, controlling a coefficient of ordinary degree $q$ requires a remainder in $\mathfrak m_\rho^{q+1}$, rather than merely in $\mathfrak m_\rho^q$. Replace the argument beginning “Choose a basis of integral $\sigma_\alpha$-eigenvectors” through the common-zero estimate by:
With the chosen coordinates $S_\rho^{\mathrm{int}}\simeq\mathcal O_L[[x_1,\ldots,x_m]]$, put
\[ \mathfrak n=(x_1,\ldots,x_m), \qquad \Lambda=\mathfrak n/\mathfrak n^2. \]Let $T$ be the action of $\sigma_\alpha$ on $\Lambda$, and let $V_1$ and $V_2$ be the $\alpha$- and $\alpha^2$-eigenspaces in $\Lambda\otimes_{\mathcal O_L}L$. Set
\[ \Lambda_i=\Lambda\cap V_i, \qquad \Lambda'=\Lambda_1\oplus\Lambda_2, \qquad d=\alpha-\alpha^2. \]The spectral projectors are
\[ P_1=\frac{T-\alpha^2}{d}, \qquad P_2=\frac{T-\alpha}{\alpha^2-\alpha}. \]For $v\in\Lambda$, one has $dP_i(v)\in\Lambda_i$ and $dv=dP_1(v)+dP_2(v)$. Consequently,
\[ d\Lambda\subseteq\Lambda'\subseteq\Lambda. \]Choose integral bases of $\Lambda_1$ and $\Lambda_2$. For a chosen vector of weight $w\in\{1,2\}$, first choose an integral representative $g\in S_\rho^{\mathrm{int}}\cap W^{-w}S_\rho$. The equivariant splitting in the proof of Theorem 5.1.8 lifts its class in $\operatorname{gr}_W^{-w}S_\rho$ to a $\sigma_\alpha$-eigenfunction $e$ with
\[ e\equiv g\pmod{W^{-w-1}S_\rho}. \]In this way obtain eigenfunctions $e_1,\ldots,e_m\in S_\rho'$, to which Lemma 5.2.2 applies. Write
\[ \mathbf e=B\mathbf x+\text{terms of degree at least two}, \qquad B\in M_m(\mathcal O_L). \]The lattice inclusions above imply that $B$ is invertible over $L$ and that $dB^{-1}\in M_m(\mathcal O_L)$. Define
\[ \mathbf f=B^{-1}\mathbf e; \]then $f_i\equiv x_i\pmod{\mathfrak n^2}$. The normalization from $\mathbf e$ to $\mathbf f$ costs at most
\[ \delta=v_\ell(d)=v_\ell(1-\alpha)\leq C(\alpha). \]Put $\mathfrak m_\rho=\mathfrak nS_\rho$. The corrected weight filtration satisfies
\[ W^{-n}S_\rho\subseteq\mathfrak m_\rho^{\lceil n/2\rceil} \qquad(n\geq1). \]Indeed, this follows by induction from Definition 5.1.3: it holds for $n=1,2$, and every summand $W^{-a}S_\rho\,W^{-b}S_\rho$ with $a+b=n$ lies in $\mathfrak m_\rho^{\lceil a/2\rceil+\lceil b/2\rceil}$. Consequently, in order that the remainder make no contribution in ordinary degree $q$, Lemma 5.2.2 must be applied through $r=2q$, since $W^{-2q-1}S_\rho\subseteq\mathfrak m_\rho^{q+1}$.
If $e_j$ has weight $w\in\{1,2\}$, the resulting denominator is
\[ D_{w,q}=\prod_{a=w+1}^{2q}(\alpha^w-\alpha^a). \]Its valuation is at most
\[ \sum_{k=1}^{2q-w}v_\ell(1-\alpha^k) \leq(2q-w)C(\alpha) \leq(2q-1)C(\alpha). \]After applying $B^{-1}$, every coefficient $b_{I,i}$ of degree $q=|I|$ in
\[ f_i=x_i+\sum_{|I|\geq2}b_{I,i}x^I \]therefore satisfies
\[ v_\ell(b_{I,i}) \geq-(2q-1)C(\alpha)-\delta \geq-2qC(\alpha). \]This is the replacement for estimate (5.3.1) in the chosen integral coordinates. It also shows that each $f_i$ belongs to $\mathcal O_{U_{\ell^{-s}}}$ whenever $s>4C(\alpha)$.
The common zero loci of the $e_j$ and the $f_i$ agree. If an arithmetic point $\widetilde\rho\in U_{\ell^{-s}}$ is fixed by a power of $\sigma_\alpha$, then $z_{\widetilde\rho}(e_j)=0$ for all $j$, hence $z_{\widetilde\rho}(f_i)=0$ for all $i$. Put
\[ t=\min_i v_\ell\!\left(z_{\widetilde\rho}(x_i)\right)\geq s. \]For every $q\geq2$ and $s>4C(\alpha)$,
\[ qt-2qC(\alpha)>t, \]because $(q-1)t>2qC(\alpha)$, with the strongest condition occurring at $q=2$. Thus every nonlinear term in the equation $z_{\widetilde\rho}(f_i)=0$ has valuation strictly greater than the least valuation of the linear terms. The ultrametric inequality forces $z_{\widetilde\rho}(x_i)=0$ for every $i$, and hence $\widetilde\rho\simeq\rho$.
This restores the integral coordinate estimate. The qualitative statement of Theorem 1.1.11 is unchanged: one may choose its constant $N=N(c(\rho),\ell)$ so that $s>4C(\alpha)$. In the proof on page 1893 and in Remark 5.3.2, replace the stated sufficient bound $3C(\alpha)$ by $4C(\alpha)$.
Report metadata
| Field | Value |
|---|---|
| Category | Published |
| Processing status | completed |
| Detailed comments | 14 |
| Domain | stem/mathematics |
| Completed | 2026-07-29T16:30:50.536183+00:00 |
| Refine document ID | e24ed041-37fa-4c72-8366-1338c08721d8 |
Refine summary
This paper analyzes the dynamics of the Galois action on the deformation rings of mod ℓ representations of the geometric fundamental group of smooth curves. Its main contributions include various finiteness results for function fields over algebraically closed fields in arbitrary characteristic and a weak variant of the Frey-Mazur conjecture for function fields in characteristic 0.
Overall feedback
Rigidity over the completed local ring
The proof of the algebraicity result in Theorem 3.2.3 handles the critical task of excluding non-algebraic points from a positive-dimensional Frobenius-fixed locus. In Step 4, the argument applies Lemma 3.2.1 to the completed local ring $\widehat S$. However, Lemma 3.2.1 is stated only for Artinian local algebras.
To make this step rigorous, it appears necessary to pass through the quotients $\widehat S/\mathfrak m^a$. The argument must verify that the specialized irreducible representation is arithmetic and fixed by the relevant Frobenius power, and subsequently use Krull intersection to obtain constancy over $\widehat S$. An explicit explanation of why the injection $S\hookrightarrow\widehat S$ forces the original traces in $\mathbb C_\ell$ to lie in $\overline{\mathbb Q_\ell}$ is necessary to finalize the deduction.
Specialization models in arbitrary characteristic
In Step 2 of the proof of Theorem 1.1.3, an initial congruence with the trivial residual characteristic polynomial is treated as yielding a residually trivial representation, which is then used to factor through $\pi_1^\ell$. Because what is initially controlled is strictly the residual pseudorepresentation, readers must see why the residual image is necessarily a finite $\ell$-group and how the full compact image is proven to be pro-$\ell$. This must be shown either directly or after descent to a finite coefficient extension.
Regarding the arbitrary-characteristic scope of the theorem, the spreading-out construction selects $R\subset k$ finitely generated over $\mathbb Z$. This construction is impossible when $\operatorname{char}k>0$. Establishing a separate characteristic-$p$ model is required to support the theorem's application to arbitrary characteristics.
Integral eigensplitting and effective radius
The effective radius computation in Section 5.3 relies on assuming a basis of integral $\sigma_\alpha$-eigenvectors in $\mathfrak m'_{\rho}/\mathfrak m_{\rho}'^2$, passing to a coordinate change that satisfies $e_i\equiv x_i\pmod{\mathfrak m^2}$ without altering the integral power-series structure or the rigid ball. Theorem 5.1.8 supplies eigensplitting over the coefficient field. Because $\alpha$ and $\alpha^2$ are congruent modulo $\ell$, the integral tangent lattice is not automatically guaranteed to split into eigensublattices, and the current puncturing argument does not establish this integral compatibility.
To secure the asserted bound $N>3C(\alpha)$, which depends only on $c(\rho)$ and $\ell$, it is necessary to apply Lemmas 5.2.1–5.2.2 to explicitly bound the index and denominators of an integral eigensplitting, and then propagate that bound through the coefficient estimates. Without this control, the bound may conceal an unconstrained lattice discriminant.
Recovering the original representation after restriction
In Section 5.4, the proof strategy involves passing to the cover associated with $\ker(\pi_1\to G)$. Proving triviality on this cover demonstrates that $\widetilde\rho$ factors through $G$, but it does not establish that the resulting representation of $G$ is isomorphic to the specific $\rho$ initially identified.
To derive Theorem 1.1.13 and its stated unique-lifting consequences, the proof requires a finite-character separation argument tailored for semisimple representations of $G$. Combining this separation argument with the radius obtained on the cover, and verifying that this additional bound remains compatible with $N_G(\ell)\to0$, will resolve this gap.
Deducing the higher-dimensional and complex conclusions
The abstract emphasizes discreteness and finiteness results for normal connected complex varieties, whereas the substantive main theorem concerns curves over finitely generated fields. The extension to the complex and higher-dimensional settings is delegated to Remark 1.1.9 via a Lefschetz argument and standard specialization.
Because these are principal advertised conclusions of the paper, the document requires a formal corollary producing the appropriate curve. The proof accompanying this corollary needs to show explicitly that restriction detects isomorphism and convergence of semisimple representations. Furthermore, it must verify that representations arising from geometry remain arithmetic through restriction, spreading out, and specialization. This structural addition is vital for normal, potentially singular, or nonproper varieties, where the required fundamental-group reduction cannot be assumed without proof.
Detailed comments
1. Theorem 1.1.13 needs an \(\ell\)-adic embedding
- ID:
bdf27e34-8c0c-42be-a9bb-15e0c1bbc50d - Refine score:
0.31 - Original types: general
- Refine status: open
Comment
Theorem 1.1.13 is under-specified: the congruence involving $\operatorname{ch}(\rho)$ and the scalar extension $\rho\otimes\overline{\mathbb{Q}}_\ell$ require a choice of embedding $\overline{\mathbb{Q}}\hookrightarrow\overline{\mathbb{Q}}_\ell$. Different choices can give different $\ell$-adic realizations and reductions of the finite-image representation.
Quoted passage
THEOREM 1.1.13 Let $X, k, \bar{x}$ be as in Theorem 1.1.11. Let
$$ \rho: \pi_{1}^{\mathrm{et}}\left(X_{\bar{k}}, \bar{x}\right) \rightarrow \mathrm{GL}_{n}(\overline{\mathbb{Q}}) $$be a representation which factors through a finite quotient $G$ of $\pi_{1}^{\text {ét }}\left(X_{\bar{k}}, \bar{x}\right)$. Then there exists a sequence of constants $N_{G}(\ell)>0$ with $N_{G}(\ell) \rightarrow 0$ as $\ell \rightarrow \infty$ such that if
$$ \tilde{\rho}: \pi_{1}^{\mathrm{et}}\left(X_{\bar{k}}, \bar{x}\right) \rightarrow \mathrm{GL}_{n}\left(\overline{\mathbb{Q}_{\ell}}\right) $$is semisimple arithmetic with
$$ \operatorname{ch}(\rho) \equiv \operatorname{ch}(\tilde{\rho}) \bmod \ell^{N_{G}(\ell)}, $$then we have $\rho \otimes \overline{\mathbb{Q}_{\ell}} \simeq \tilde{\rho}$.
2. Esnault–Kerz Theorem 2.1 proves finiteness only for irreducible representations
- ID:
6d008c16-b231-46f9-86d6-15956cb6495a - Refine score:
0.73 - Original types: external_references
- Refine status: open
Comment
The cited theorem does not establish the claim for all semisimple representations. Esnault–Kerz’s Theorem 2.1 says that the set of irreducible rank-r sheaves with the prescribed ramification bound is finite up to a single twist by a character from the base finite field. Replacing “irreducible” with “semisimple” is material: direct sums permit independently varying base-field characters, and an overall twist does not eliminate their relative characters. For example, even representations of the form 1 ⊕ χ give infinitely many semisimple classes modulo an overall twist as χ varies. The sentence should say “irreducible,” not “semisimple,” unless additional determinant, purity, or constituent-wise twisting conditions are imposed. See https://arxiv.org/pdf/1208.0128.
Quoted passage
Work of Deligne, Drinfel'd, and Lafforgue implies (via automorphic methods) that if $X$ is a variety over a finite field $\mathbb{F}_{q}$, then the set of semisimple $\overline{\mathbb{Q}_{\ell}}$ -representations of its Weil group $W(X)$, with fixed rank and bounded wild ramification at infinity, is finite up to twist by characters of $W\left(\mathbb{F}_{q}\right)$ (see [16, Theorem 2.1]).
3. Residue field mismatch in Section 2.1
- ID:
6609c7b1-aa60-4ab7-8ad9-5c803f493d09 - Refine score:
0.21 - Original types: general
- Refine status: open
Comment
The stated residue field of $R_{\bar{\rho}}^{\square}$ should be $\mathbb{F}_{\ell^{r}}$, not $\mathbb{F}_{\ell}$. The coefficient ring is $\Lambda=W(\mathbb{F}_{\ell^{r}})$, and the deformation category consists of local Artinian $\Lambda$-algebras with residue field $\mathbb{F}_{\ell^{r}}$.
Quoted passage
There is an evident map $D_{\bar{\rho}}^{\square} \rightarrow D_{\bar{\rho}}$, given by forgetting the framing. As $G$ satisfies Mazur's finiteness condition $\left(\Phi_{\ell}\right), D_{\bar{\rho}}^{\square}$ is prorepresentable by a local Noetherian $\Lambda$-algebra $R_{\bar{\rho}}^{\square}$ with residue field $\mathbb{F}_{\ell}$ (see proof of [27, Proposition 1]). In general the functor $D_{\bar{\rho}}$ is not prorepresentable, although it is if $\bar{\rho}$ is absolutely irreducible; in this case we call the prorepresenting object $R_{\bar{\rho}}$. The groups
4. Rigid connections and F-isocrystals has no Proposition 4.6
- ID:
0cc84971-2a05-4a28-add2-23ff880d8b71 - Refine score:
0.73 - Original types: external_references
- Refine status: open
Comment
The pinpoint citation to Esnault and Groechenig’s “Rigid connections and F-isocrystals” is incorrect. In the final Acta Mathematica article identified by DOI 10.4310/ACTA.2020.v225.n1.a2, item 4.6 is Definition 4.6, not a proposition; it defines an f-periodic flat connection and does not give the claimed descent argument. A somewhat related projective-extension result occurs later as Proposition 5.7, but that does not validate the stated citation to “Proposition 4.6.” The pinpoint should therefore be corrected or removed. Source: https://archive.intlpress.com/site/pub/files/_fulltext/journals/acta/2020/0225/0001/ACTA-2020-0225-0001-a002.pdf
Quoted passage
A similar argument appears in [15, Proposition 4.6] and [14, Proposition 3.1].
5. Interpolation step in Lemma 4.1.3 needs repair
- ID:
62890e59-5868-467e-b736-3f12e95b0f4f - Refine score:
0.77 - Original types: general
- Refine status: open
Comment
The proof of Lemma 4.1.3 has a substantive gap. The asserted congruence $\varphi^{M_1}(\mathbf{x})\equiv\mathbf{x}\pmod{\varpi}$ need not hold for a general formal automorphism; for example, the reduction of $\varphi(x)=x+x^\ell$ has infinite compositional order in characteristic $\ell$. In addition, $\vartheta$ is an $N$-tuple interpolating coordinate maps, so the later expression $\vartheta(\varpi^{-1}f,m)$ does not define the required pullback of the scalar function $f$. Consequently, the claimed analytic interpolation of $z\circ\varphi^{j+Mm}(f)$, including its uniformity in $\varphi$, is not established by the argument as written.
Quoted passage
There exists $M_{1}$ depending only on $c^{\prime}, N, \ell^{r}$ such that $\varphi^{M_{1}}(\mathbf{x})=\mathbf{x} \bmod \varpi$. Let
$$ \tilde{\varphi}(\mathbf{x})=\frac{1}{\varpi} \varphi^{M_{1}}(\varpi \cdot \mathbf{x}) . $$Note that $\tilde{\varphi}$ lies in $\Lambda^{\prime}\left\langle x_{1}, \ldots, x_{N}\right\rangle^{N}$. Then there exists $M_{2}>0$ depending only on $c^{\prime}, \ell^{r}, N$ such that $\tilde{\varphi}^{M_{2}}$ satisfies the hypotheses of Lemma 4.1.1; let $\vartheta \in$ $\Lambda^{\prime}\left\langle x_{1}, \ldots, x_{N}, n\right\rangle^{N}$ be such that $\vartheta(\mathbf{x}, m)=\tilde{\varphi}^{M_{2} m}(\mathbf{x})$ for each $m \in \mathbb{Z}_{\geq 0}$, and let $M=M_{1} M_{2}$.
6. Wrong embedding dimension in Corollary 4.1.5
- ID:
9161d256-6a01-403a-88d5-9f3badaa5de3 - Refine score:
0.47 - Original types: general
- Refine status: open
Comment
The presentation used in Corollary 4.1.5 has an off-by-one cotangent-space count. For $S=\Lambda[[x_1,\ldots,x_N]]$, the space $\mathfrak m_S/\mathfrak m_S^2$ has dimension $N+1$ over the residue field because it includes the class of $\ell$. If the kernel is contained in $\mathfrak m_S^2$, the quotient has the same cotangent-space dimension, contradicting the stated definition $N=\dim(\mathfrak m_R/\mathfrak m_R^2)$. Consequently, the claimed invertibility of the lifted endomorphism is not justified by the presentation as written; the argument requires a corrected relative embedding-dimension presentation or an equivalent adjustment.
Quoted passage
Let $\mathfrak{m}_{R}$ be the maximal ideal of $R$. Write $R=S / \mathcal{L}$, where $S=\Lambda\left[\left[x_{1}, \ldots, x_{N}\right]\right]$ with $N=\operatorname{dim} \mathfrak{m}_{R} / \mathfrak{m}_{R}^{2}$ and where $\mathscr{J} \subset \mathfrak{m}_{S}^{2}$ is an ideal, so the rigid generic fiber of $R$ is a closed analytic subset of the open unit ball. We may lift $\varphi$ to an automorphism $\tilde{\varphi}$ of $\Lambda\left[\left[x_{1}, \ldots, x_{N}\right]\right]$ (indeed any lift of $\varphi$ to an endomorphism of $\Lambda\left[\left[x_{1}, \ldots, x_{n}\right]\right]$ is an isomorphism, as the induced map on $\mathfrak{m}_{S} / \mathfrak{m}_{S}^{2}$ is invertible by our choice of $S$ ).
7. The kernel used in Corollary 4.1.5 need not cut out the point
- ID:
ddf3c379-7fd8-44ee-a565-229218344ded - Refine score:
0.42 - Original types: general
- Refine status: open
Comment
The ideal $\ker(S\to R\xrightarrow{z}L)$ is generally not maximal in the integral ring $S$ and may define all conjugates of the corresponding closed point rather than the selected geometric point $z$. Semilinearity of visits to that larger zero locus does not by itself imply $\varphi^M(z)=z$. The application of Lemma 4.1.3 therefore requires an ideal over $\mathcal O_{\mathbb C_\ell}$ whose zero locus is the chosen geometric point, rather than the contracted kernel displayed here.
Quoted passage
By quasicompactness of affinoids, $U$ is contained in $U_{c}(S)$ for some $c$ with $1 \geq c>0$. Now let $z: R \rightarrow L$ be a $\varphi$-periodic point of $U$; it is a $\tilde{\varphi}$-periodic point of $U_{c}(S)$. Let $\mathcal{I} \subset S$ be the maximal ideal cutting out $z$ (i.e., $\mathcal{d}=\operatorname{ker}(S \rightarrow R \xrightarrow{z} L)$ ). Now the result follows from Lemma 4.1.3, applied to $z, \ell$; note that the integer $M$ coming from Lemma 4.1.3 depends only on $U$, and not on $\varphi, z$, and so on. $\square$
8. Recursive weight filtration is circular in Definition 5.1.3
- ID:
08e53838-d0b1-4a28-a4ec-e16dc7c9ac12 - Refine score:
0.3 - Original types: general
- Refine status: open
Comment
The recursive formula for $W^{-m}S_\rho$ does not restrict the summation indices. As written, the pairs $(i,j)=(0,m)$ and $(m,0)$ make the right-hand side depend on $W^{-m}S_\rho$ itself, so the formula does not uniquely define the higher filtration pieces.
Quoted passage
$$ \begin{aligned} W^{i} S_{\rho} & =S_{\rho} \quad \text { for } i \geq 0, \\ W^{-1} S_{\rho} & =\mathfrak{m}_{\rho}, \\ W^{-2} S_{\rho} & =\mathfrak{m}_{\rho}^{2}+W^{-2}\left(\mathfrak{m}_{\rho} / \mathfrak{m}_{\rho}^{2}\right), \end{aligned} $$and
$$ W^{-m} S_{\rho}=\sum_{i+j=m}\left(W^{-i} S_{\rho}\right) \cdot\left(W^{-j} S_{\rho}\right) \quad \text { for } m>2 . $$Remark 5.1.4 If $X$ is proper, then $W^{-i}=\mathfrak{m}_{\rho}^{i}$ for $i \geq 0$.
9. Character and cocharacter lattices are conflated in Step 2
- ID:
2f2fe13e-5c5b-4c16-84ed-947f77e5a2ea - Refine score:
0.28 - Original types: general
- Refine status: open
Comment
The passage conflates character and cocharacter lattices. For the inclusion $T\hookrightarrow D$, the displayed surjection, its kernel $K$, the monomial relations $\prod_i\lambda_i^{a_i}=1$, and the equations cutting out $T$ belong to the character lattice $X^*(D)\to X^*(T)$. The induced map on cocharacter lattices instead goes injectively from $X_*(T)$ to $X_*(D)$.
Quoted passage
The inclusion $T \hookrightarrow D$ induces a surjection on cocharacter lattices $X(D) \rightarrow X(T)$ with kernel $K ; T$ is precisely the subtorus of $D$ cut out by the characters in $K$. If we identify $D$ with $\mathbb{Z}^{\operatorname{dim} V}$ via the choice of basis $\left\{e_{i}\right\}$, then $K$ consists of the vectors $\underline{a}=\left(a_{1}, \ldots, a_{\operatorname{dim} V}\right)$ such that
$$ \prod_{i} \lambda_{i}^{a_{i}}=1 . $$
10. Weight sign in Theorem 5.1.8 appears reversed
- ID:
3fa1cf9a-dc20-4f09-8296-0e992697e4ea - Refine score:
0.35 - Original types: general
- Refine status: open
Comment
Theorem 5.1.8 has a sign/index mismatch. The nontrivial filtration pieces of $S_\rho$ are indexed as $\operatorname{gr}_W^{-i}S_\rho$, and the inverse element used in the proof acts on those pieces by $\alpha^i$. Read literally, the stated action on $\operatorname{gr}_W^iS_\rho$ by $\alpha^i$ gives the opposite scalar when $i<0$.
Quoted passage
THEOREM 5.1.8 Let
$$ \rho: \pi_{1}^{\text {et }}\left(X_{\bar{k}}, \bar{c}\right) \rightarrow \mathrm{GL}_{n}\left(\overline{\mathbb{Q}_{\ell}}\right) $$be an irreducible representation which arises from geometry. Then for $\alpha \in \mathbb{Z}_{\ell}^{\times}$sufficiently close to 1, there exists $\sigma_{\alpha} \in G_{k}$ such that $\sigma_{\alpha}$ acts on $\operatorname{gr}_{W}^{i} S_{\rho}$ via $\alpha^{i} \cdot \mathrm{Id}$.
In fact, we will be able to choose the element $\sigma_{\alpha}$ in Theorem 5.1.8 to be inverse to the element $\sigma_{\alpha}$ constructed in Lemma 5.1.5.
11. The symmetric algebra must be completed in Step 1
- ID:
55a3b820-7d52-4ef8-a5aa-c81bd3fb0bba - Refine score:
0.25 - Original types: general
- Refine status: open
Comment
The displayed isomorphism is not literally correct for the ordinary symmetric algebra. Since $S_\rho$ is a complete formal power-series ring, the isomorphism requires the completed symmetric algebra $\widehat{\operatorname{Sym}}(\mathfrak m_\rho/\mathfrak m_\rho^2)$ and the continuous extension of the equivariant splitting.
Quoted passage
Such a splitting induces a $\sigma_{\alpha}^{-1}$-equivariant isomorphism
$$ \operatorname{Sym}^{*}\left(\mathfrak{m}_{\rho} / \mathfrak{m}_{\rho}^{2}\right) \xrightarrow{\sim} S_{\rho}, $$which respects the weight filtrations, where the weight filtration on $\operatorname{Sym}^{*}\left(\mathfrak{m}_{\rho} / \mathfrak{m}_{\rho}^{2}\right)$ is induced from the filtration on $\mathfrak{m}_{\rho} / \mathfrak{m}_{\rho}^{2}$, by the multiplicativity of the weight filtration. The element $\sigma_{\alpha}^{-1} \in G_{k}$ clearly acts on $\operatorname{Sym}^{*}\left(\mathfrak{m}_{\rho} / \mathfrak{m}_{\rho}^{2}\right)$ as desired (again by multiplicativity), so we are done.
12. Module hypotheses are unchecked in Lemma 5.2.2
- ID:
69e9a63a-fac8-452a-b60c-36da6d9b74d6 - Refine score:
0.25 - Original types: general
- Refine status: open
Comment
The application of Lemma 5.2.1 is not formally justified under that lemma’s stated finite-freeness hypotheses: the quotients involving $S_\rho^{\mathrm{int}}\cap W^{-r}$ are not shown to be finite free over the non-Noetherian ring $\overline{\mathbb{Z}}_\ell$. This does not appear to threaten Lemma 5.2.2, because the scalar argument used in Lemma 5.2.1 works here without finite freeness; if $\alpha^r=\alpha^i$, the new product factor is zero and the induction step is immediate.
Quoted passage
Set
$$ \begin{aligned} V & =\left(S_{\rho}^{\mathrm{int}} \cap W^{-r}+\overline{\mathbb{Z}_{\ell}} \cdot \bar{x}\right) /\left(S_{\rho}^{\mathrm{int}} \cap W^{-r-1}\right), \\ W & =\left(S_{\rho}^{\mathrm{int}} \cap W^{-r}\right) /\left(S_{\rho}^{\mathrm{int}} \cap W^{-r-1}\right), \end{aligned} $$and
$$ v=\left(\prod_{j=i+1}^{r-1}\left(\alpha^{i}-\alpha^{j}\right)\right) \cdot x \bmod W^{-r-1} $$Then the hypotheses of Lemma 5.2.1 are satisfied by the induction hypothesis, giving the proof.
13. Integral eigen-coordinates need justification in Section 5.3
- ID:
b8a15f1f-6f16-4890-89ed-9192bcd9bb3a - Refine score:
0.74 - Original types: general
- Refine status: open
Comment
The integral-coordinate step in Section 5.3 is not established. Theorem 5.1.8 gives the $\alpha$- and $\alpha^2$-eigensplitting over $L$, but because $\alpha-\alpha^2$ is generally not a unit, the tangent lattice need not admit an $\mathcal O_L$-basis of eigenvectors. Consequently, the linear change making $e_i\equiv x_i\pmod{(\mathfrak m_\rho')^2}$ may lie only in $\mathrm{GL}_m(L)$, in which case the $x_i$ need not remain integral coordinates and the stated Gauss-norm and coefficient bounds do not follow with the claimed radius $s>3C(\alpha)$. An integral splitting or a uniform bound on the resulting coordinate denominators is required.
Quoted passage
Choose a basis of integral $\sigma_{\alpha}$-eigenvectors of $\mathfrak{m}_{\rho}^{\prime} / \mathfrak{m}_{\rho}^{\prime 2}$, and lift it to a set of $\sigma_{\alpha}$ eigenvectors $\left\{e_{1}, \ldots, e_{m}\right\}$ of $S_{\rho}^{\prime}$ (we may do this by the proof of Theorem 5.1.8). After a linear change of coordinates, we may assume that $e_{i} \equiv x_{i} \bmod \left(\mathfrak{m}_{\rho}^{\prime}\right)^{2}$.
14. Link from \(H_\ell\) to \(c(\rho_\ell)\) is unclear
- ID:
a37008ab-4ec7-46fb-a2d7-b05be66e37c2 - Refine score:
0.35 - Original types: general
- Refine status: open
Comment
The implication in Remark 5.4.1 is not formal from the displayed definition of $H_\ell$. The index $c(\rho_\ell)$ also involves the weight-graded pieces of $H^1(Y_{\bar{k}},\rho_\ell\otimes\rho_\ell^\vee)$ and the boundary term from Lemma 5.1.5. The conclusion is plausible if the Tate-conjectural argument simultaneously realizes and controls these representations—through tensor constructions, geometric projectors, and the relevant boundary contribution—but that comparison is not made explicit here.
Quoted passage
Assuming the Tate conjecture, one may show this index is uniformly bounded via the argument of [33, Section 2.3]. This gives, by the proof of Theorem 1.1.11, a much stronger version of Theorem 1.1.11. It implies (on the Tate conjecture) that if $\rho_{\ell}$ is a compatible system of $\ell$-adic representations arising from geometry (i.e., the monodromy representation underlying $R^{q} f_{*} \mathbb{Q}_{\ell}$ as above), then the constants $N\left(c\left(\rho_{\ell}\right), \ell\right)$ of Theorem 1.1.11 tend to zero as $\ell \rightarrow \infty$.
Scope
- Paper:
03 Published and Submitted Work/Published/P16_Litt_Arithmetic_Representations_II.pdf - Refine report:
.refine/results/Published/P16_Litt_Arithmetic_Representations_II.review.json - Rubric:
rubric/comment_triage_rubric.md - Version assessed: the local published PDF, Duke Mathematical Journal 170 (2021), 1851-1897
- Detailed Refine comments assessed: 14
- Assessment date: 2026-07-30
The local published PDF is the authority for the text under review. Relevant pages were rendered and inspected visually, in particular where overlines, filtration signs, and residue-field subscripts could be lost in extraction. External sources were consulted only for comments 2 and 4, which allege specific citation errors.
Summary
| # | Short title | Validity | Category | Standardness | Impact | Challenge | Repair | Disposition | Priority | Confidence |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Theorem 1.1.13 needs an \(\ell\)-adic embedding | V4 | C4 Notation | E2 | I1 | Q1 | R1 | D1 | P3 | HIGH |
| 2 | Esnault-Kerz finiteness is irreducible, not semisimple | V4 | C7 Citation | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 3 | Alleged residue-field mismatch | V0 | C4 Notation | E-NA | I0 | Q0 | R0 | D0 | P4 | HIGH |
| 4 | Wrong Esnault-Groechenig pinpoint | V4 | C7 Citation | E-NA | I2 | Q0 | R1 | D3 | P2 | HIGH |
| 5 | Interpolation in Lemma 4.1.3 | V4 | C6 Correctness | E4 | I2 | Q2 | R2 | D3 | P1 | MEDIUM |
| 6 | Relative embedding dimension in Corollary 4.1.5 | V4 | C6 Correctness | E-NA | I2 | Q2 | R2 | D3 | P2 | HIGH |
| 7 | The geometric point ideal in Corollary 4.1.5 | V4 | C6 Correctness | E2 | I2 | Q2 | R2 | D3 | P2 | HIGH |
| 8 | Circular indices in Definition 5.1.3 | V4 | C1 Typo | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 9 | Character versus cocharacter lattice | V4 | C1 Typo | E-NA | I2 | Q0 | R1 | D3 | P2 | HIGH |
| 10 | Weight sign in Theorem 5.1.8 | V4 | C1 Typo | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 11 | Symmetric algebra needs completion | V4 | C4 Notation | E-NA | I2 | Q0 | R1 | D3 | P2 | HIGH |
| 12 | Finite freeness in Lemma 5.2.2 | V3 | C3 Elaboration | E1 | I0 | Q1 | R0 | D0 | P4 | HIGH |
| 13 | Integral eigen-coordinates in Section 5.3 | V4 | C6 Correctness | E4 | I2 | Q2 | R2 | D3 | P2 | HIGH |
| 14 | Tate-conjectural control of \(c(\rho_\ell)\) | V3 | C3 Elaboration | E3 | I1 | Q1 | R1 | D1 | P3 | MEDIUM |
No issue remains at I3 or higher. Comments 5 and 13 initially appeared capable of affecting main results, but the complete bounded repairs below preserve those results. In particular, the projector denominator in Comment 13 fits inside one unit of the coefficient bound that the printed proof does not use, so the stated radius \(s>3C(\alpha)\) is unchanged.
1. Theorem 1.1.13 needs an \(\ell\)-adic embedding
Comment ID: bdf27e34-8c0c-42be-a9bb-15e0c1bbc50d Location: PDF p. 6 (printed p. 1856), Theorem 1.1.13.
The displayed congruence and the tensor product \(\rho\otimes\overline{\mathbb Q}_\ell\) require an embedding \(\iota_\ell:\overline{\mathbb Q}\hookrightarrow \overline{\mathbb Q}_\ell\). Galois-conjugate representations of the finite group \(G\) can be nonisomorphic, so the realization is not literally choice-free.
This is, however, a standard suppressed convention. Once \(\iota_\ell\) is fixed, the proof reduces along the kernel of the resulting finite-image representation and proceeds unchanged. The constants depend only on \(G\), so no theorem hypothesis or conclusion changes.
- Classification:
V4/C4/E2 - Severity challenge:
Q1; the only missing datum is the customary embedding - Impact/repair:
I1/R1 - Disposition:
D1;P3/HIGH
Optional correction: begin the theorem with “For each \(\ell\), fix an embedding \(\iota_\ell:\overline{\mathbb Q}\hookrightarrow \overline{\mathbb Q}_\ell\), and write \(\rho_{\iota_\ell}=\rho\otimes_{\iota_\ell}\overline{\mathbb Q}_\ell\).”
2. Esnault-Kerz prove finiteness for irreducible sheaves
Comment ID: 6d008c16-b231-46f9-86d6-15956cb6495a Location: PDF p. 9 (printed p. 1859), Section 1.3.
The sentence attributes to [16, Theorem 2.1] finiteness, up to one base-field twist, of all semisimple representations of fixed rank and bounded ramification. The cited result is for irreducible lisse sheaves. This is also how the result is stated in the Esnault-Kerz preprint.
The distinction is material. For example, \(1\oplus\chi\), with \(\chi\) ranging through characters pulled back from \(W(\mathbb F_q)\), has trivial wild ramification. Modulo an overall twist, the relative character \(\chi\) remains, up to inversion, so infinitely many classes remain.
- Classification:
V4/C7/I2 - Dependency trace: this is a comparison-with-prior-work paragraph; the paper explicitly gives its own proof of Theorem 1.1.3
- Repair/disposition:
R2/D3 - Priority/confidence:
P2/HIGH
Correction: replace “semisimple” by “irreducible” in the first sentence of the paragraph. If a semisimple formulation is desired, state a constituent-wise twisting equivalence or add conditions controlling the relative twists.
3. The residue field is already \(\mathbb F_{\ell^r}\)
Comment ID: 6609c7b1-aa60-4ab7-8ad9-5c803f493d09 Location: PDF p. 10 (printed p. 1860), Section 2.1.
The comment is based on a transcription error in the Refine quotation. Visual inspection of the published page shows that the paper says
consistent with \(\Lambda=W(\mathbb F_{\ell^r})\) and the category \(\mathcal C_\Lambda\). There is no \(\mathbb F_\ell\) mismatch in the local PDF.
- Classification:
V0/C4/I0 - Repair/disposition:
R0/D0 - Priority/confidence:
P4/HIGH
4. Reference [15] has no Proposition 4.6
Comment ID: 0cc84971-2a05-4a28-add2-23ff880d8b71 Location: PDF p. 15 (printed p. 1865), Remark 3.1.2.
The pinpoint “[15, Proposition 4.6]” is incorrect. In the final Esnault-Groechenig paper, 4.6 is Definition 4.6, defining an \(f\)-periodic flat connection. Proposition 5.7 is a later projective-extension result, but it is not the cited item and is not an exact replacement for the argument in Proposition 3.1.1.
- Classification:
V4/C7/I2 - Dependency trace: Remark 3.1.2 is comparative only; Proposition 3.1.1 has a self-contained proof
- Repair/disposition:
R1/D3 - Priority/confidence:
P2/HIGH
Correction: remove the [15] pinpoint unless the author identifies the intended passage; retain [14, Proposition 3.1].
5. Lemma 4.1.3 needs the rescaled coordinate map
Comment ID: 62890e59-5868-467e-b736-3f12e95b0f4f Location: PDF p. 25 (printed p. 1875), proof of Lemma 4.1.3.
Both defects identified by Refine are present. A general automorphism of \(\Lambda[[\mathbf x]]\) need not have a uniformly bounded power equal to the identity modulo \(\varpi\); the reduction of \(x\mapsto x+x^\ell\) is a useful test. Also, \(\vartheta\) is an \(N\)-tuple of coordinate functions, so the printed expression \(\vartheta(\varpi^{-1}f,m)\) is not defined for a scalar \(f\).
Severity challenge and repair
The intended proof works after conjugating the point map on the small ball by the scaling \(\mathbf a=\varpi\mathbf y\).
- Choose rational \(0<c'<\min(c,1/2)\) and a totally ramified \(\Lambda'/\Lambda\) with \(|\varpi|_\ell=\ell^{-c'}\).
- If \(F\) is the coordinate self-map of points induced by \(\varphi\), set \(\widetilde F(\mathbf y)=\varpi^{-1}F(\varpi\mathbf y)\). Its constant and nonlinear terms vanish modulo \(\varpi\); its reduction is the invertible linear part of \(F\). A uniform exponent \(M_1\) of \(\mathrm{GL}_N(\mathbb F_{\ell^r})\) therefore gives \(\widetilde F^{M_1}\equiv\mathrm{id}\pmod\varpi\).
- A further uniform power \(M_2\) makes the map close enough to the identity for the analytic arc lemma. Let \(\vartheta(\mathbf y,m)=\widetilde F^{M_1M_2m}(\mathbf y)\).
- For the coordinate vector \(\mathbf a=z(\mathbf x)\), define \[ h_{f,j}(m)= f\!\left(\varpi\, \vartheta\!\left(\varpi^{-1}F^j(\mathbf a),m\right)\right). \] This is a scalar analytic function of \(m\), and its zeros are exactly the visits of the \(j\)-th residue-class subsequence to \(V(f)\).
The example \(x+x^\ell\) no longer obstructs the proof: after scaling, its nonlinear term acquires the factor \(\varpi^{\ell-1}\) and disappears in the reduction. Uniformity depends only on \(c,\ell^r,N\), as required.
- Classification:
V4/C6/E4 - Severity status:
Q2; the corrected coordinate calculation is complete and bounded to this proof - Impact:
I2, notI3 - Dependency trace: the repaired lemma restores Corollaries 4.1.5 and 4.1.6, Theorem 1.1.3, and Corollary 1.1.5 without changing their statements
- Repair/disposition:
R2/D3 - Priority/confidence:
P1/MEDIUM
6. Corollary 4.1.5 needs relative embedding dimension
Comment ID: 9161d256-6a01-403a-88d5-9f3badaa5de3 Location: PDF p. 26 (printed p. 1876), proof of Corollary 4.1.5.
For \(S=\Lambda[[x_1,\ldots,x_N]]\), the absolute cotangent space \(\mathfrak m_S/\mathfrak m_S^2\) also contains the class of \(\ell\). Thus the printed choice \(N=\dim\mathfrak m_R/\mathfrak m_R^2\), together with \(\mathcal J\subset\mathfrak m_S^2\), is not the correct minimal \(\Lambda\)-presentation.
Local repair
Use
and a relative Cohen presentation \(S=\Lambda[[x_1,\ldots,x_N]]\twoheadrightarrow R\). No assertion \(\mathcal J\subset\mathfrak m_S^2\) is needed: any kernel cuts out a closed analytic subspace of the open ball. A lift of \(\varphi\) has invertible linear part on the relative cotangent space and is therefore an automorphism of \(S\).
The test ring \(\Lambda[[x]]/(\ell-x^2)\) shows why retaining the printed \(\mathcal J\subset\mathfrak m_S^2\) formulation is unsafe.
- Classification:
V4/C6/I2 - Severity/repair:
Q2/R2 - Dependency trace: this repairs only the presentation and lift in Corollary 4.1.5; the uniform-period conclusion and all later uses are unchanged
- Disposition:
D3;P2/HIGH
7. The contracted kernel need not isolate the geometric point
Comment ID: ddf3c379-7fd8-44ee-a565-229218344ded Location: PDF p. 26 (printed p. 1876), final paragraph of Corollary 4.1.5.
The kernel of \(S\to R\xrightarrow z L\) need not be a maximal ideal of the integral ring \(S\), and its \(\mathbb C_\ell\)-zero locus can contain conjugate points. Hitting that larger locus does not imply returning to the chosen geometric point.
Lemma 4.1.3 already permits ideals in \(\mathcal O_{\mathbb C_\ell}[[\mathbf x]]\). Apply it instead to
Then \(V(\mathcal I_z)=\{z\}\). If \(z\) has exact period \(q\), the return-time set is \(q\mathbb Z_{\ge0}\). Being semilinear with the lemma's uniform period \(M\) forces \(q\mid M\), hence \(\varphi^M(z)=z\).
- Classification:
V4/C6/E2/I2 - Severity/repair:
Q2/R2 - Dependency trace: Corollaries 4.1.5-4.1.6 and the finiteness proof then work exactly as stated
- Disposition:
D3;P2/HIGH
8. The recursive filtration formula is circular as printed
Comment ID: 08e53838-d0b1-4a28-a4ec-e16dc7c9ac12 Location: PDF pp. 30-31 (printed pp. 1880-1881), Definition 5.1.3.
The sum \(\sum_{i+j=m}W^{-i}S_\rho\cdot W^{-j}S_\rho\) includes \((i,j)=(0,m)\) and \((m,0)\), while \(W^0S_\rho=S_\rho\). The right-hand side therefore contains the object being defined.
The intended multiplicative recursion is clear from Remark 5.1.4 and later uses: require \(i,j\ge1\). This yields the smallest multiplicative filtration generated by its degree-one and degree-two pieces.
- Classification:
V4/C1withmeaning_changing_typo/I2 - Dependency trace: later arguments use the intended multiplicative filtration
- Repair/disposition:
R2/D3 - Priority/confidence:
P2/HIGH
9. Step 2 uses character lattices
Comment ID: 2f2fe13e-5c5b-4c16-84ed-947f77e5a2ea Location: PDF p. 33 (printed p. 1883), proof of Lemma 5.1.5.
For \(T\hookrightarrow D\), the surjection with kernel \(K\) is
The vectors \((a_i)\), the monomial relations \(\prod_i\lambda_i^{a_i}=1\), and the equations cutting out \(T\) are all characters. The induced map on cocharacters goes in the opposite, injective direction.
The calculation itself is the correct character-lattice calculation, so this is a one-word terminology error.
- Classification:
V4/C1/I2 - Repair: replace “cocharacter lattices” by “character lattices” and, ideally, write \(X^*(-)\)
- Disposition:
R1/D3;P2/HIGH
10. Theorem 5.1.8 has the wrong filtration sign
Comment ID: 3fa1cf9a-dc20-4f09-8296-0e992697e4ea Location: PDF p. 34 (printed p. 1884), Theorem 5.1.8; compare pp. 31 and 37-40.
The nonzero pieces of \(S_\rho\) are \(\operatorname{gr}_W^{-i}\) for \(i\ge0\). The proof and Lemma 5.2.2 use an element acting on those pieces by \(\alpha^i\). The theorem instead prints \(\operatorname{gr}_W^iS_\rho\) with scalar \(\alpha^i\).
The theorem can be reparameterized using \(\alpha^{-1}\), so this does not invalidate the existence assertion. It does, however, conflict with the “inverse to the element in Lemma 5.1.5” sentence and with the indexing used in the quantitative argument.
- Classification:
V4/C1withmeaning_changing_typo/I2 - Dependency trace: the proof and Section 5.2 already use the intended \(-i\) indexing
- Repair/disposition:
R2/D3 - Priority/confidence:
P2/HIGH
Correction: state that \(\sigma_\alpha\) acts on \(\operatorname{gr}_W^{-i}S_\rho\) via \(\alpha^i\operatorname{Id}\), for \(i\ge0\).
11. The symmetric algebra must be completed
Comment ID: 55a3b820-7d52-4ef8-a5aa-c81bd3fb0bba Location: PDF p. 37 (printed p. 1887), Step 1 of Theorem 5.1.8.
An ordinary symmetric algebra is a polynomial ring, whereas \(S_\rho\) is a complete formal power-series ring. The equivariant splitting gives
as complete local algebras. The preceding sentence already invokes completeness, so the intended topology is unambiguous.
- Classification:
V4/C4/I2 - Dependency trace: every eigenvalue and filtration calculation is degreewise and remains valid after completion
- Repair/disposition:
R1/D3 - Priority/confidence:
P2/HIGH
12. Lemma 5.2.1 does not need finite freeness here
Comment ID: 69e9a63a-fac8-452a-b60c-36da6d9b74d6 Location: PDF pp. 40-41 (printed pp. 1890-1891), proof of Lemma 5.2.2.
The displayed modules over \(\overline{\mathbb Z}_\ell\) are not shown to be finite free, so the literal hypotheses of Lemma 5.2.1 are not checked. Nevertheless, the calculation needed in the induction works for arbitrary torsion-free lattices.
If \(T|_W=\alpha\,\mathrm{id}\), \(T\) acts by \(\beta\) on \(V/W\), and \(v=y+w\) with \(y\in V\), \(w\in W\otimes\overline{\mathbb Q}_\ell\), then
Thus \((\alpha-\beta)v\in V\), which is exactly the step used in Lemma 5.2.2. No Noetherian or basis hypothesis enters.
- Classification:
V3/C3/E1 - Severity status:
Q1; this is a verified routine generalization, not a proof gap - Impact/repair/disposition:
I0/R0/D0 - Priority/confidence:
P4/HIGH
An optional sentence could state the torsion-free version, but no erratum is needed for correctness.
13. The integral eigenbasis claim needs a projector-denominator argument
Comment ID: b8a15f1f-6f16-4890-89ed-9192bcd9bb3a Location: PDF p. 42 (printed p. 1892), proof of Theorem 1.1.11.
Theorem 5.1.8 splits the tangent space over \(L\), with eigenvalues \(\alpha\) and \(\alpha^2\). Because \(\alpha-\alpha^2=\alpha(1-\alpha)\) is not a unit when \(\alpha\) is close to one, the integral lattice need not have a basis of eigenvectors.
The elementary matrix
is a failure test: it is diagonalizable over \(L\), but an eigenvector can require division by \(\alpha-\alpha^2\). Consequently, the subsequent linear change of coordinates need not lie in \(\mathrm{GL}_m(\mathcal O_L)\), and the printed Gauss-norm estimate in the chosen integral ball does not follow.
Severity challenge and complete repair
Let \(\Lambda\) be the integral cotangent lattice and let \(V_1,V_2\) be the \(\alpha\)- and \(\alpha^2\)-eigenspaces in \(\Lambda\otimes_{\mathcal O_L}L\). Put
The spectral projectors are
Thus \(dP_i(\Lambda)\subseteq\Lambda_i\) and \(dv=dP_1(v)+dP_2(v)\), so
Choose integral bases of the two \(\Lambda_i\)'s and lift them, as in the proof of Theorem 5.1.8, to eigenfunctions \(e_j\). If \(B\) is their integral linear-coefficient matrix, then \(B\) is invertible over \(L\) and \(dB^{-1}\) is integral. Passing from \(e\) to \(f=B^{-1}e\), whose linear term is the coordinate vector \(x\), therefore costs at most
There is exactly enough room in Lemma 5.2.2 for this cost. For an eigenfunction of weight \(w=1\) or \(2\), controlling its degree-\(q\) coefficient only requires taking \(r=2q-1\) in that lemma. The relevant denominator is
whose valuation is at most
After applying \(B^{-1}\), every degree-\(q\) coefficient of \(f\) therefore has valuation at least
which is precisely the printed estimate (5.3.1). The printed proof used the coarser product through \(2q-1\) before the coordinate normalization, leaving the needed one-\(C(\alpha)\) margin implicit.
Finally, the common zeros of the eigenfunctions \(e_j\) are the common zeros of \(f_i=x_i+\) higher terms. If \(t=\min_i v_\ell(x_i)\ge s>3C(\alpha)\), then for every \(q\ge2\)
the worst case is \(q=2\). Hence the ultrametric argument proving that the origin is the only common zero on the ball goes through verbatim.
- Classification:
V4/C6/E4 - Severity status:
Q2; the triangular failure test confirms the printed integral-basis sentence is false, while the eigenlattice calculation and the sharpened denominator count give a complete repair - Impact:
I2; Theorem 1.1.11 and Remark 5.3.2 retain the exact \(s>3C(\alpha)\) radius - Repairability/disposition:
R2/D3 - Priority/confidence:
P2/HIGH
14. Remark 5.4.1 suppresses the tensor and boundary reductions
Comment ID: a37008ab-4ec7-46fb-a2d7-b05be66e37c2 Location: PDF pp. 44-45 (printed pp. 1894-1895), Remark 5.4.1.
The index \(H_\ell\) displayed in the remark is not literally \(c(\rho_\ell)\). The latter is defined using the weight-graded pieces of \(H^1(Y_{\bar k},\rho_\ell\otimes\rho_\ell^\vee)\) and the boundary term in Lemma 5.1.5.
The reduction is standard but nontrivial. For a geometric compatible system, \(\rho_\ell\otimes\rho_\ell^\vee\) is obtained from the cohomology of a fiber square using Kunneth, duality, and geometric projectors. Taking \(p=1\) in the construction of \(H_\ell\) controls the two weights in its cohomology; the finitely many boundary systems are handled by the localization sequence and the corresponding fiber or inertia subquotients. Under the Tate conjecture, the projectors and these subquotients are compatible across \(\ell\). Intersecting the finitely many bounded-index homothety subgroups still gives uniformly bounded index. This is the comparison to which the cited argument of [33, Section 2.3] must be applied.
The paper calls this a conditional remark and cites the relevant method; no proof of Theorem 1.1.11 or 1.1.13 depends on it. The comment is therefore right that a comparison sentence would help, but overstates the omission as a defect in an established theorem.
- Classification:
V3/C3/E3 - Severity status:
Q1; the tensor, projector, and boundary route is reconstructible, with the Tate conjecture supplying the compatible projectors - Impact/repair/disposition:
I1/R1/D1 - Priority/confidence:
P3/MEDIUM
Author decisions and follow-up
- Add Comment 13's eigenlattice/projector calculation to a living errata list; it corrects the integral-basis sentence without changing the stated radius.
- For Comment 5, have the author check the rescaled-coordinate formula and replace the two malformed lines in Lemma 4.1.3.
- Add comments 2 and 4 to a living errata list as literature-reference corrections.
- Treat comments 1 and 14 as optional clarifications, and dismiss comments 3 and 12.
P18 VANISHING FOR FROBENIUS TWISTS OF AMPLE VECTOR BUNDLES1 detailed comments · 1 numbered correction 1 I2
These errata refer to the version published in Tohoku Mathematical Journal (2) 71 (2019), no. 4, 549--557, doi:10.2748/tmj/1576724793. Page references below are to that version.
Page 555, first paragraph of the proof of Theorem 3.0.2.
The assertion that one model works simultaneously for every $n>N_0$ and every $i$ does not follow directly from the definition of $\phi$: that definition supplies a model for each fixed vector bundle, whereas $n$ is later specialized to the unbounded residue characteristic. Replace the first paragraph of the proof, ending with “exists by the definition of $\phi$),” by the following.
Put $r=\operatorname{rk}(\mathscr E)$, and choose a very ample line bundle $L$ on $X$. Choose a finite-type $\mathbb Z$-algebra $R$, with a map $R\to k$, together with models $\mathcal X\to S:=\operatorname{Spec}(R)$, $\widetilde{\mathscr E}$, and $\widetilde L$ of $X$, $\mathscr E$, and $L$. After enlarging and localizing $R$, we may assume that $\mathcal X\to S$ is flat and projective, that $\widetilde L$ is relatively very ample, and that $\widetilde{\mathscr E}$ is a vector bundle. We may also assume that
\[ \mathcal O_{\mathbf P_{\mathcal X}(\widetilde{\mathscr E})}(1) \]is ample relative to $S$. Here we use the quotient convention, so that, for $N\geq 0$ and the projection $\pi\colon\mathbf P_{\mathcal X}(\widetilde{\mathscr E})\to\mathcal X$,
\[ \pi_*\mathcal O(N)=\operatorname{Sym}^N(\widetilde{\mathscr E}) \quad\text{and}\quad R^b\pi_*\mathcal O(N)=0\quad(b>0). \]There are integers $d$ and $C\geq 1$ such that, for every closed point $\mathfrak q\in S$,
\[ \dim(\mathcal X_{\mathfrak q})\leq d, \qquad \operatorname{Reg}_{\widetilde L_{\mathfrak q}} (\mathcal X_{\mathfrak q})\leq C, \]where
\[ \operatorname{Reg}_{\widetilde L_{\mathfrak q}} (\mathcal X_{\mathfrak q}) :=\max\!\left\{1, \operatorname{reg}_{\widetilde L_{\mathfrak q}} (\mathcal O_{\mathcal X_{\mathfrak q}})\right\}. \]Indeed, the dimensions are bounded in this projective family, and the regularity bound follows by applying relative Serre vanishing and cohomology and base change to the finitely many twists $\widetilde L^{-a}$, $1\leq a\leq d$.
Choose a positive integer
\[ M>C\max\{d-1,0\}. \]Let $P=\mathbf P_{\mathcal X}(\widetilde{\mathscr E})$ and let $h\colon P\to S$ be the structure morphism. Apply generic flatness, relative Serre vanishing, and cohomology and base change to the finite collection
\[ \pi^*\!\left( \bigwedge^i\widetilde{\mathscr E} \otimes\widetilde L^{-M-a} \right), \qquad 0\leq i\leq r,\quad 1\leq a\leq d. \]After one further localization of $R$, there is a single integer $N_1$ such that all positive higher direct images under $h$ of these sheaves tensored with $\mathcal O_P(N)$ vanish, and their formation commutes with base change, for every $N\geq N_1$. The projective-bundle identities above then give, for every closed $\mathfrak q\in S$, every $N\geq N_1$, every $0\leq i\leq r$, and every $a>0$,
\[ H^a\!\left( \mathcal X_{\mathfrak q}, \operatorname{Sym}^N(\widetilde{\mathscr E}_{\mathfrak q}) \otimes\bigwedge^i\widetilde{\mathscr E}_{\mathfrak q} \otimes\widetilde L_{\mathfrak q}^{-M-a} \right)=0; \]for $a>d$ this also follows from the dimension bound. Consequently
\[ \operatorname{reg}_{\widetilde L_{\mathfrak q}}\!\left( \operatorname{Sym}^N(\widetilde{\mathscr E}_{\mathfrak q}) \otimes\bigwedge^i\widetilde{\mathscr E}_{\mathfrak q} \right)\leq -M. \]By [Ara04, Lemma 3.3], this strict inequality implies
\[ \phi\!\left( \operatorname{Sym}^N(\widetilde{\mathscr E}_{\mathfrak q}) \otimes\bigwedge^i\widetilde{\mathscr E}_{\mathfrak q} \right)=0, \]because
\[ -M<- \operatorname{Reg}_{\widetilde L_{\mathfrak q}} (\mathcal X_{\mathfrak q}) \bigl(\dim(\mathcal X_{\mathfrak q})-1\bigr). \]Choose $N_0\geq N_1+r$. Then $n-i\geq N_1$ whenever $n>N_0$ and $0\leq i\leq r$, so, simultaneously for every closed $\mathfrak q\in S$,
\[ \phi\!\left( \operatorname{Sym}^{n-i}(\widetilde{\mathscr E}_{\mathfrak q}) \otimes\bigwedge^i\widetilde{\mathscr E}_{\mathfrak q} \right)=0 \qquad(n>N_0,\ 0\leq i\leq r). \]For $i>r$ the corresponding exterior power is zero. This is the uniform statement needed below when $n$ is taken to be $\operatorname{char}(\kappa(\mathfrak q))$.
With this replacement, the spectral-sequence argument proving Theorem 3.0.2 is unchanged, as are Remark 3.0.3 and the later applications of the theorem.
References
Donu Arapura, Frobenius amplitude and strong vanishing theorems for vector bundles, with an appendix by Dennis S. Keeler, Duke Math. J. 121 (2004), no. 2, 231--267, doi:10.1215/S0012-7094-04-12122-0.
Report metadata
| Field | Value |
|---|---|
| Category | Published |
| Processing status | completed |
| Detailed comments | 1 |
| Domain | stem/mathematics |
| Completed | 2026-07-29T16:47:28.018648+00:00 |
| Refine document ID | c97eeff6-ebfc-4b70-a5cd-410a55cce6f9 |
Refine summary
This paper proves asymptotic vanishing theorems for Frobenius twists of ample vector bundles in positive characteristic under the assumption that the vector bundles and the scheme lift to characteristics modulo $p^2$. A main contribution is the application of these techniques to generalize the Bott-Danilov-Steenbrink vanishing theorem for ample vector bundles on toric varieties.
Overall feedback
Deligne-Illusie splitting in the main induction
In the induction step of Theorem 2.2.1, the complex builds from $\mathscr{E}^{(p^{N+s})}$. Applying the results of Section 2.1 directly requires $\mathscr{E}^{(p^{N+s+1})}$ to lift.
The theorem explicitly assumes only that $\mathscr{E}^{(p^N)}$ lifts, which leaves open the question of whether the second spectral sequence degenerates for every $s$ based on the stated hypothesis. The mathematical logic resolves by either strengthening the lifting assumption or detailing how a valid decomposition at the initial Frobenius level cleanly transports through successive Frobenius pullbacks while preserving the required grading and tensor factors.
Coefficient rings in Corollary 2.2.3
Theorem 2.2.1 requires a lift over $W_2(k)$. By contrast, Corollary 2.2.3 assumes only a lift over $\mathbb{Z}/p^2\mathbb{Z}$ before invoking the theorem directly.
For general perfect $k$, these data are not interchangeable without an additional construction. Identifying the reduction of $F_2^*\mathscr{E}_2$ with the intended Frobenius twist relies heavily on a precise semilinearity convention. The corollary functions fully if it imposes compatible $W_2(k)$-data or rigorously establishes that the version of Deligne-Illusie used in Section 2 applies under the weaker assumptions.
Uniform arithmetic models in Theorem 3.0.2
The proof of Theorem 3.0.2 invokes the definition of $\phi$ to select an arithmetic model where $\phi(\operatorname{Sym}^{n-i}\mathscr{E}\otimes\bigwedge^i\mathscr{E})=0$ for all $n>N_0$, all relevant $i$, and all closed fibers.
Because the definition supplies an arithmetic model bundle-by-bundle, rather than a universal model for this infinite family, uniform behavior cannot be assumed outright. A uniform relative-regularity argument following the spreading out and shrinking of the base addresses this. Furthermore, the fiberwise proof must handle arbitrary coherent sheaves on each closed fiber, rather than restricting scope to reductions of coherent sheaves selected on the characteristic-zero variety.
Spreading out in Theorem 4.0.2
The positive-characteristic branch of Theorem 4.0.2 requires that each reduced vector bundle lifts to the canonical toric $W_2$-lift. This specific behavior is not supplied merely by choosing an arbitrary finite-type $\mathbb{Z}$-model.
The characteristic-zero proof currently pivots on a one-line spreading out argument. Section 4 provides a cohesive picture when it spells out how the fan, the variety, and the bundles are spread out to guarantee that suitable closed fibers admit the required compatible $W_2$-lifts. Attention must also be given to how the sheaves $j_*\Omega_U^q$ on normal singular fibers behave under this process. Supplying the relevant base-change or semicontinuity argument connects the characteristic-zero case rigorously to the positive-characteristic branch.
Detailed comments
1. Section 3 uses one model for infinitely many bundles
- ID:
de22d381-3259-4d2d-8c9e-e000ee536db5 - Refine score:
0.61 - Original types: general
- Refine status: open
Comment
The appeal to the definition of $\phi$ does not by itself justify a single model on which the displayed amplitude-zero statement holds simultaneously for every $n>N_0$ and every $i$. Because the later argument takes $n=p$ while the residue characteristic varies, a uniform relative regularity or spreading-out argument is needed for this unbounded family.
Quoted passage
Now let $R$ be a finite-type $\mathbb{Z}$-algebra, with a map $R \rightarrow k$ and $(\mathcal{X}, \widetilde{\mathscr{E}})$ a finite-type $R$-scheme with a vector bundle so that $\mathcal{X}_{k} \simeq X, \widetilde{\mathscr{E}}_{k} \simeq \mathscr{E}$, and such that
$$ \phi\left(\operatorname{Sym}^{n-i}\left(\widetilde{\mathscr{E}}_{\mathfrak{q}}\right) \otimes \bigwedge^{i} \widetilde{\mathscr{E}}_{\mathfrak{q}}\right)=0 $$for all closed points $\mathfrak{q} \in \operatorname{Spec}(R)$ (such a model $(R, \mathcal{X}, \widetilde{\mathscr{E}})$ exists by the definition of $\phi$ ).
Scope
- Paper:
03 Published and Submitted Work/Published/P18_Litt_Vanishing_Frobenius_Twists.pdf - Refine report:
.refine/results/Published/P18_Litt_Vanishing_Frobenius_Twists.review.json - Rubric:
rubric/comment_triage_rubric.md - Version assessed: local published PDF, Tohoku Mathematical Journal 71 (2019), 549-557
- Detailed Refine comments assessed: 1
- Assessment date: 2026-07-30
The local published PDF is authoritative. PDF page 7 (printed p. 555) was rendered and visually inspected; the disputed quantifiers and parenthetical justification are present in the published typesetting.
Summary
| # | Short title | Validity | Category | Standardness | Impact | Challenge | Repair | Disposition | Priority | Confidence |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | One model for an unbounded family | V4 | C6 | E3 | I2 | Q2 | R2 | D3 | P2 | HIGH |
The comment initially merits serious scrutiny because the disputed uniformity is used in the proof of Theorem 3.0.2. The reconstruction below verifies a complete uniform relative-regularity repair, so no main statement changes. The printed appeal to the definition nevertheless makes a false quantifier inference, and the missing replacement is nontrivial; the final impact is therefore I2, not merely editorial and not I3.
1. The definition of \(\phi\) does not itself provide the asserted uniform model
Comment ID: de22d381-3259-4d2d-8c9e-e000ee536db5 Location: PDF p. 7 (printed p. 555), first two paragraphs of the proof of Theorem 3.0.2.
For each fixed \(n\) and \(i\), the characteristic-zero definition of \(\phi\) supplies a model on which
on every closed fiber after localization. It does not directly give one model for all \(n>N_0\). A finite common refinement is insufficient because the proof later takes \(n=p=\operatorname{char}\kappa(\mathfrak q)\), which is unbounded as \(\mathfrak q\) varies.
The Refine comment is therefore correct about the parenthetical “such a model exists by the definition of \(\phi\).” The theorem nevertheless has a uniform repair coming from the regularity proof of Lemma 3.0.1.
Severity reconstruction
After enlarging and localizing \(R\), spread out the following data:
- a flat projective morphism \(\mathcal X\to\operatorname{Spec}R\), a relatively very ample line bundle \(L\), and the vector bundle \(\widetilde{\mathscr E}\);
- relative ampleness of \(\mathcal O_{\mathbf P(\widetilde{\mathscr E})}(1)\) over \(\operatorname{Spec}R\);
- a common fiber dimension bound \(d\) and a uniform bound \[ \operatorname{Reg}_{L_{\mathfrak q}}(\mathcal X_{\mathfrak q})\le C \] for all closed \(\mathfrak q\).
Only \(0\le i\le r=\operatorname{rk}(\mathscr E)\) matter, since the higher exterior powers vanish. Choose \(M>C(d-1)\). On \(\mathbf P(\widetilde{\mathscr E})\), relative Serre vanishing, applied to the finite collection
gives a single \(N_1\) such that, for every closed \(\mathfrak q\), every \(N\ge N_1\), and every \(a>0\),
After generic flatness for this finite collection, relative Serre vanishing kills all positive higher direct images over \(\operatorname{Spec}R\). Cohomology and base change, together with the projective-bundle identities, then gives the displayed vanishing on every closed fiber. This is uniform in both \(N\) and \(\mathfrak q\), not merely pointwise on the base. Consequently,
uniformly in \(\mathfrak q,N,i\).
Arapura, Lemma 3.3 says that a coherent sheaf \(\mathscr G\) is Frobenius-ample when
The choices above therefore imply the desired amplitude-zero statement for all fibers and all \(N\ge N_1\). Replacing \(N_0\) by \(\max(N_0,N_1+r)\) makes \(n-i\ge N_1\) whenever \(n>N_0\), simultaneously for every relevant \(i\).
This is precisely the uniform statement needed when the proof specializes \(n\) to the varying residue characteristic \(p\).
Failure tests and dependencies
- Unbounded \(i\): absent, because \(\bigwedge^i\mathscr E=0\) for \(i>r\).
- Negative symmetric powers: avoided by taking \(N_0\ge N_1+r\).
- Varying fiber dimension or regularity: controlled by flat projective spreading and the uniform constant \(C\).
- Fiberwise base change: controlled by generic flatness and simultaneous vanishing of all positive relative higher direct images.
- Small residue characteristics: the proof already restricts to \(p>N_0\).
- Singular fibers: Arapura's regularity criterion and the argument above are stated for projective varieties and do not require smoothness.
- Downstream use: the spectral-sequence argument needs only this uniform amplitude-zero assertion; it is unchanged after the replacement justification.
- Classification:
V4/C6withuniform_spreading_out;relative_regularity;incorrect_justification;quantifier_exchange/E3 - Severity challenge:
Q2; the missing uniform spreading argument has been reconstructed completely from relative Serre vanishing, cohomology and base change, and the regularity criterion already underlying Lemma 3.0.1 - Impact:
I2 - Repair: replace the incorrect parenthetical appeal to the definition of \(\phi\) by the relative-regularity paragraph above, or cite a uniform relative version of Lemma 3.0.1
- Dependency trace: Theorem 3.0.2, its later citation in the paper, and all advertised results remain unchanged
- Disposition:
R2/D3;P2/HIGH
Action queue
- Add a public errata entry replacing the incorrect parenthetical with the uniform relative-regularity argument.
- Use the same replacement paragraph in any maintained manuscript version.
P19 Arithmetic representations of fundamental groups I10 detailed comments · 7 numbered corrections 3 I16 I21 I3
These errata refer to the version published in Inventiones mathematicae 214 (2018), no. 2, 605--639, doi:10.1007/s00222-018-0810-4. Page references below are to that version.
Page 606, Definition 1.1; pages 631--633, Lemma 4.1 and the proof of Theorem 1.2.
Definition 1.1 assigns the same rank to $\rho$ and to its ambient arithmetic representation. Over $\mathbb Q_\ell$, a rank-$n$ subquotient of a rank-$n$ representation has full dimension, so this excludes the intended case in which $\rho$ is a lower-dimensional constituent of a larger representation. In Definition 1.1, replace the second displayed map by
\[ \widetilde\rho\colon \pi_1^{\mathrm{\acute et}}(X_{k'},\bar x) \longrightarrow \operatorname{GL}_m(\mathbb Z_\ell) \qquad\text{for some }m\geq 1. \]Thus $m$ is independent of the rank $n$ of $\rho$.
Make the corresponding rank change in Lemma 4.1: its conclusion should read that there are a finite extension $k\subset k'$, an integer $m\geq1$, and a representation
\[ \beta\colon \pi_1^{\mathrm{\acute et}}(X_{k'},\bar x) \longrightarrow \operatorname{GL}_m(\mathbb Z_\ell) \]such that $\rho$ is a subquotient of $\beta|_{\pi_1^{\mathrm{\acute et}}(X_{\bar k},\bar x)}$ and this restriction is trivial modulo $\ell^r$. In the first paragraph of its proof, likewise write
\[ \gamma\colon \pi_1^{\mathrm{\acute et}}(X_{k'},\bar x) \longrightarrow \operatorname{GL}_m(\mathbb Z_\ell) \]for the ambient representation.
Finally, in the proof of Theorem 1.2, the representation produced by Lemma 4.1 has target $\operatorname{GL}_m(\mathbb Z_\ell)$, and the subsequent displays are
\[ \ker\!\left( \operatorname{GL}_m(\mathbb Z_\ell) \longrightarrow \operatorname{GL}_m(\mathbb Z/\ell^N\mathbb Z) \right) \]and
\[ \mathbb Q_\ell[[\pi_1^\ell(X_{\bar k},\bar x)]]^{\leq\ell^{-r}} \xrightarrow{\ \widetilde\beta\ } M_m(\mathbb Q_\ell). \]Every subsequent occurrence of $M_n(\mathbb Q_\ell)$ referring to $\beta$ must accordingly be replaced by $M_m(\mathbb Q_\ell)$. The construction in Lemma 4.1 already takes the full span of the conjugates of $\rho$, so these changes do not alter Theorems 1.2 or 1.4 or Corollary 1.6.
Page 616, Step 1 of the proof of Lemma 2.10.
The displayed equality-to-one character relations cut out the full Zariski closure of the cyclic group generated by $\gamma$, which need not be connected; they do not necessarily cut out its identity component $T$. Replace the paragraph beginning “Let $X^*(D),X^*(T)$ be the character lattices” and ending “as desired” by the following.
Put
\[ H=\overline{\{\gamma^n:n\in\mathbb Z\}}\subset D, \qquad T=H^\circ. \]Identify $X^*(D)$ with $\mathbb Z^m$ using the basis $\{e_i\}$. The kernel of the restriction $X^*(D)\to X^*(H)$ is
\[ K_H= \left\{(a_1,\ldots,a_m)\in\mathbb Z^m: \prod_{i=1}^m\lambda_i^{a_i}=1\right\}. \]For $(a_1,\ldots,a_m)\in K_H$, multiplicativity of the complex absolute value gives
\[ 2\sum_{i=1}^r a_i+\sum_{i=r+1}^m a_i=0. \]Consequently every character in $K_H$ is trivial on the torus
\[ T'=\left\{ \alpha\,\operatorname{Id}_{\operatorname{gr}^{-1}_W} \oplus \alpha^2\,\operatorname{Id}_{\operatorname{gr}^{-2}_W} :\alpha\in\mathbb G_m \right\}. \]The character equations defining $H$ therefore give $T'\subset H$. Since $T'$ is connected and contains the identity, it follows that $T'\subset H^\circ=T$. The splitting (2.1) is defined over $\mathbb Q_\ell$, so $T'$ is defined over $\mathbb Q_\ell$ and is contained in $\overline{\operatorname{im}(\rho)}$, as required.
This restores Step 1 and hence Lemma 2.10. The elements $\sigma_\alpha$ used in Theorems 2.8, 2.12, and 3.6 are unchanged, as are the main results.
Page 623, Example 3.1.
The assertion about integral eigenvectors is false when $\chi(\sigma)$ is a nontrivial root of unity. For example, if $\chi(\sigma)=-1$, then
\[ \sigma(T)=-\frac{T}{1+T} \qquad\text{and}\qquad \frac{T^2}{1+T}\in\mathbb Z_\ell[[T]] \]is a nonconstant invariant. Replace the final sentence of the example by:
On the other hand, if $\chi(\sigma)$ has infinite order, the integral $\sigma$-eigenvectors in $\mathbb Z_\ell[[T]]$ are precisely the constant series.
The later elements $\sigma_\alpha$ have $\alpha$ of infinite order, so no subsequent result is affected.
Page 626, final paragraph of the statement of Theorem 3.6.
Pointwise linear growth of the denominators of one eigenvector is not equivalent to the existence of one common radius together with Gauss-norm density of the full eigenvector span. Replace the paragraph beginning “Equivalently” by:
In addition, for this same $r_\alpha$, every $\sigma_\alpha$-eigenvector
\[ y\in\mathbb Q_\ell[[\pi_1^\ell(X_{\bar k},\bar x)]] \]belongs to $\mathbb Q_\ell[[\pi_1^\ell(X_{\bar k},\bar x)]]^{\leq\ell^{-r}}$ for every $r>r_\alpha$. In particular, $-v_n(\pi_n(y))$ grows at most linearly in $n$.
The uniform-radius and density assertions in the preceding sentences remain separate conclusions of the theorem and are proved by the corrected argument in the next item.
Pages 630--631, density portion of the proof of Theorem 3.6.
The printed estimate uses a one-step comparison between the weight and augmentation filtrations. Proposition 2.7 gives only
\[ \mathscr I^n\subset W^{-n}, \qquad W^{-2n-1}\subset\mathscr I^n, \]so a residual need not contract after each individual weight step. Replace the proof from the paragraph beginning “Note that, by the estimates in the previous two paragraphs” through the end of the proof by the following.
It remains to prove the density assertions. Put
\[ A_{\mathbb Z_\ell} =\mathbb Z_\ell[[\pi_1^\ell(X_{\bar k},\bar x)]], \qquad A=\mathbb Q_\ell[[\pi_1^\ell(X_{\bar k},\bar x)]], \]with their completed weight filtrations. By Theorems 2.8 and 2.12 and Proposition 2.7, the action of $\sigma_\alpha$ on $A/W^{-m}$ is semisimple, and all its eigenvalues belong to the distinct set
\[ 1,\alpha,\ldots,\alpha^{m-1}. \]For $0\leq i<m$, let
\[ P_{i,m}(\sigma_\alpha) =\prod_{\substack{0\leq j<m\\j\neq i}} \frac{\sigma_\alpha-\alpha^j}{\alpha^i-\alpha^j} \]be the projector onto the $\alpha^i$-eigenspace. These projectors are compatible under the quotient maps $A/W^{-m'}\to A/W^{-m}$ for $m'>m$.
Let $\Lambda_m$ be the image of $A_{\mathbb Z_\ell}$ in $A/W^{-m}$. The numerator of $P_{i,m}(\sigma_\alpha)$ preserves $\Lambda_m$, while the valuation of its denominator is
\[\begin{aligned}D(i,m) &=\sum_{\substack{0\leq j<m\\j\neq i}} v_\ell(\alpha^i-\alpha^j)\\ &=\sum_{s=1}^{i}v_\ell(\alpha^s-1) +\sum_{s=1}^{m-i-1}v_\ell(\alpha^s-1)\\ &\leq C(\alpha,\ell,i)+C(\alpha,\ell,m-i-1),\end{aligned}\]where $C(\alpha,\ell,0)=0$. The last inequality is Lemma 3.10. Hence
\[ P_{i,m}(\sigma_\alpha)(\Lambda_m) \subset \ell^{-D(i,m)}\Lambda_m. \]For $z\in A_{\mathbb Z_\ell}$, compatibility of the projectors defines an element $w_i\in A$ whose image in $A/W^{-m}$, for every $m>i$, is
\[ P_{i,m}(\sigma_\alpha)(z\bmod W^{-m}). \]For $m\leq i$ this image is zero. Thus
\[ w_i\in W^{-i}A, \qquad \sigma_\alpha(w_i)=\alpha^i w_i, \qquad z\equiv\sum_{i=0}^{m-1}w_i\pmod{W^{-m}}. \]We now estimate $w_i$ in the $r$-Gauss norm. If $n<\lceil i/2\rceil$, then $w_i\in W^{-i}\subset\mathscr I^n$, so $\pi_n(w_i)=0$. Otherwise take $m=2n+1$. Since $W^{-2n-1}\subset\mathscr I^n$, the preceding denominator estimate gives
\[ -v_n(\pi_n(w_i)) \leq C(\alpha,\ell,i)+C(\alpha,\ell,2n-i). \]By the formula in Lemma 3.10, there are constants $c,b\geq0$ such that
\[ C(\alpha,\ell,k)\leq ck+b\quad(k\geq0), \qquad c\leq C(\alpha,\ell,1). \]For $r>r_\alpha=2C(\alpha,\ell,1)$ we therefore have $r>2c$ and
\[\begin{aligned}|w_i|_r &\leq \sup_{n\geq\lceil i/2\rceil} \ell^{C(\alpha,\ell,i)+C(\alpha,\ell,2n-i)-nr}\\ &\leq \ell^{-(r-2c)\lceil i/2\rceil+2b} \longrightarrow 0 \qquad\text{as }i\longrightarrow\infty.\end{aligned}\]For each fixed $i$, the same estimate also shows that $v_n(\pi_n(w_i))+nr\to\infty$; hence
\[ w_i\in \mathbb Q_\ell[[\pi_1^\ell(X_{\bar k},\bar x)]]^{\leq\ell^{-r}}. \]The estimate $|w_i|_r\to0$ makes $\sum_iw_i$ converge in the complete $r$-Gauss norm to some element $z'$. Gauss-norm convergence implies $W$-adic convergence in $A$, while the displayed congruences give $z'=z$. If $z\in W^{-a}A_{\mathbb Z_\ell}$, the projectors with $i<a$ vanish on $z$, so only the eigenvalues $\alpha^a,\alpha^{a+1},\ldots$ occur. Finally, the $\mathbb Q_\ell$-span of $W^{-a}A_{\mathbb Z_\ell}$ is dense in
\[ W^{-a} \mathbb Q_\ell[[\pi_1^\ell(X_{\bar k},\bar x)]]^{\leq\ell^{-r}}. \]Thus finite sums of the $w_i$ give both density assertions of the theorem.
This replacement retains $r_\alpha=2C(\alpha,\ell,1)$ and the stated eigenvalue ranges. It therefore supplies exactly the form of Theorem 3.6 used in Theorems 1.2 and 1.4 and Corollary 1.6.
Page 634, Remark 4.3.
The estimate used in the proof does not give $N(X,\ell)=1$ when $\ell=3$. In the final sentence of Remark 4.3, replace “$\ell>2$” by “$\ell\geq5$”, and insert after that sentence:
When $\ell=3$, the same estimate gives $N(X,3)=2$. Indeed, for a topological generator $\alpha\in\mathbb Z_3^\times$, one has $s=2$ and $v_3(\alpha^2-1)=1$, and hence
\[ C(\alpha,3,1) =\frac12\left(1+\frac1{3-1}\right)=\frac34, \qquad r_\alpha=\frac32. \]Thus the least integer strictly greater than $r_\alpha$ is $2$.
Only the numerical optimization in Remark 4.3 changes; the existence results in Theorems 1.2 and 1.4 are unaffected.
Page 637, Question 4.7.
For a general geometric basepoint, the Galois action on the geometric fundamental group is only outer, so the question does not specify an actual Frobenius operator whose eigenvectors are to be considered. Replace Question 4.7 by:
Question 4.7. Let $X$ be a smooth curve over a finite field $k$, let $x\in X(k)$, and let $\bar x$ be the associated geometric point. Let $\ell$ be a prime different from the characteristic of $k$. Does there exist an $r=r(X)$ such that
\[ \mathbb Q_\ell[[\pi_1^\ell(X_{\bar k},\bar x)]]^{\leq\ell^{-r}} \]admits a set of Frobenius eigenvectors with dense span?
This changes only the formulation of the final open question; no theorem or proof depends on it.
Report metadata
| Field | Value |
|---|---|
| Category | Published |
| Processing status | completed |
| Detailed comments | 10 |
| Domain | stem/mathematics |
| Completed | 2026-07-29T17:41:46.771823+00:00 |
| Refine document ID | a7188689-c0fc-410b-b4a1-1e4e2a69a324 |
Refine summary
This paper investigates continuous ℓ-adic representations of the étale fundamental group of normal algebraic varieties over finitely generated fields of characteristic zero. The main contribution is proving that any nontrivial, semisimple arithmetic representation is nontrivial modulo ℓ^N for a specific integer N, which also implies that ℓ-adic representations arising from geometry satisfy the same property.
Overall feedback
Here are some structural observations from a careful read of the manuscript.
Arithmetic subquotients and integral lattices
In Definition 1.1, both $\rho$ and its ambient extension take values in $GL_n$, but an ambient representation containing $\rho$ as a genuine subquotient can have a larger sequence rank. This tension surfaces structurally in Lemma 4.1: the Galois-orbit span $W$ can contain several conjugates of $\rho$, potentially giving it a rank strictly greater than $n$. Consequently, readers might scrutinize the assertion that its lattice $\beta$ fundamentally defines a $GL_n$-representation.
Because triviality modulo $\ell^r$ strictly depends on the chosen integral lattice—and the proof navigates through rational socles, intersections, and quotients—the argument requires a methodical separation of rational constituents from finite-free integral subquotients. Explicitly proving saturation, finite freeness, and the exact preservation of the original congruence level is essential to anchor this component of the deduction.
Density estimates in Theorem 3.6
The iterative density argument in Theorem 3.6 presents a complication regarding integrality. Because $w_0$ is generally nonintegral, the difference $z_1 = z_0 - w_0$ need not lie in the integral completed group ring. Nevertheless, the subsequent step invokes an estimate established exclusively for integral graded elements.
Furthermore, the displayed contraction by $\ell^{-r}$ relies on the assumption that a single step in the weight filtration strictly increases the augmentation degree. In the mixed weight $-1/-2$ setting, Proposition 2.7 only supplies the coarser relation $W^{-2n-1} \subset \mathscr{I}^n$. Bridging the central passage from finite-level eigenvectors to dense eigenvectors in the convergent ring requires explicit operator-norm and filtration-tail estimates. Structurally connecting the relevant $W^{-s}$ subspace inside the convergent ring with the corresponding Gauss-norm completion would solidify the mechanism that Section 4 relies upon to annihilate these elements.
Basepoint extension and uniformity in Theorem 1.4
A question of exact bounding arises with respect to the $N(\ell, m)$ parameter in Theorem 1.4. The proof of Theorem 1.2 replaces $k$ with a finite extension so the chosen geometric basepoint becomes rational. Conversely, Theorem 1.4 establishes a uniformity bound depending exclusively on $\ell$ and the original Galois-image index $m$.
Executing this basepoint extension can inflate the index of the Galois image by a factor uncontrolled by $m$. Because Remark 2.11 regulates the available $\alpha$ using only the index over the field currently in use, Remark 4.2 does not fully establish the stated uniformity across the extension. Resolving this requires either bounding the effect of the field extension uniquely in terms of $m$, or structurally bypassing the extension via an outer-action or path-torsor argument.
Period bounds in the mod-3 example
There is a numerical discrepancy between the theoretical period bound and the mod-3 sharpness application. Remark 4.3 proposes that index one and $\ell > 2$ permit $N=1$. Example 4.5 relies on this parameter to rule out the arithmeticity or geometricity of a representation trivial modulo 3.
However, for a punctured projective line, $H^1$ is pure of weight 2, and Remark 3.11 therefore provides the period relation $r_\alpha = C(\alpha^2, 3, 1)$. If $\alpha$ topologically generates $\mathbb{Z}_3^\times$, then $v_3(\alpha^2 - 1) = 1$, and Lemma 3.10 dictates $C(\alpha^2, 3, 1) = 3/2$. This yields an integer threshold of at least 2, rather than 1, meaning the argument does not analytically exclude the representation in Example 4.5 based solely on its mod-3 triviality. Establishing a sharper, case-specific estimate, or mathematically restricting the scope of Remark 4.3 and Example 4.5, is necessary to support this deduction.
Detailed comments
1. Dimension mismatch in Definition 1.1
- ID:
e848fe0d-1517-41b3-a481-bb00469d4a77 - Refine score:
0.52 - Original types: general
- Refine status: open
Comment
Definition 1.1 assigns both $\rho$ and the ambient representation $\tilde{\rho}$ rank $n$. As written, a rank-$n$ free subquotient of a rank-$n$ representation has the same full dimension after tensoring with $\mathbb{Q}_\ell$, so the definition excludes proper lower-dimensional constituents of larger arithmetic representations. This is inconsistent with the broader subquotient notion used in the abstract and the geometric application.
Quoted passage
Definition 1.1 Let $k$ be a field and $X$ a geometrically connected $k$-variety (an integral, separated scheme, finite type over $k$ ) with a geometric point $\bar{x}$. Then we say that a continuous $\ell$-adic representation of the geometric étale fundamental group of $X$
$$ \rho: \pi_{1}^{\text {ét }}\left(X_{\bar{k}}, \bar{x}\right) \rightarrow G L_{n}\left(\mathbb{Z}_{\ell}\right) $$is arithmetic if there exists a finite extension $k^{\prime} / k$ and a representation
$$ \tilde{\rho}: \pi_{1}^{\text {ét }}\left(X_{k^{\prime}}, \bar{x}\right) \rightarrow G L_{n}\left(\mathbb{Z}_{\ell}\right), $$such that $\rho$ is a subquotient of $\left.\tilde{\rho}\right|_{\pi_{1}^{\text {ét }}\left(X_{\bar{k}}, \bar{x}\right)}$.
2. Semisimplicity statement needs its finite-level scope
- ID:
a844ee35-dd45-45da-a512-4785ff531b56 - Refine score:
0.34 - Original types: general
- Refine status: open
Comment
The stated scope of Theorem 2.20 is too broad: it proves that Frobenius acts semisimply on each quotient $\mathbb{Q}_\ell[[\pi_1^\ell(Y_{\overline{\mathbb{F}_q}},\bar y)]]/\mathscr I^n$, not that the entire completed algebra is semisimple in the ordinary algebraic sense. At the inverse-limit level, the relevant consequence is a dense span of eigenvectors in the $\mathscr I$-adic topology.
Quoted passage
The key input here is a semi-simplicity result (Theorem 2.12). Arguments analogous to those of the proof of Theorem 2.12 prove Theorem 2.20: that if $Y$ is a smooth variety over $\mathbb{F}_{q}$, admitting a simple normal crossings compactification, then for $y \in Y\left(\mathbb{F}_{q}\right)$, Frobenius acts semisimply on $\mathbb{Q}_{\ell}\left[\left[\pi_{1}^{\ell}\left(Y_{\overline{\mathbb{F}_{q}}}, \bar{y}\right)\right]\right]$. Step 2 (Sect. 3). For each real number $r>0$, we construct certain Galois-stable normed $\mathbb{Q}_{\ell}$-subalgebras
3. Step 3 overstates the socle reduction
- ID:
08c49c98-4c46-435c-88c2-08ff350a24e6 - Refine score:
0.37 - Original types: general
- Refine status: open
Comment
Step 3 overstates Lemma 4.1: arithmeticity does not imply that the original $\rho$ extends after finite base change. In the full proof, one first reduces to an irreducible constituent and then replaces it by an arithmetic representation $\beta$ whose geometric restriction remains congruent to the identity and has that constituent as a subquotient. The equivariant map is constructed for $\beta$; unipotence then passes through subquotients and the socle filtration.
Quoted passage
Suppose
$$ \rho: \pi_{1}^{\ell}\left(X_{\bar{k}}, \bar{x}\right) \rightarrow G L(V) $$is an arithmetic representation on a finite free $\mathbb{Z}_{\ell}$-module $V$. Then by a socle argument (Lemma 4.1), we may assume that $\rho$ extends to a representation of $\pi_{1}^{\ell}\left(X_{k^{\prime}}, \bar{x}\right)$ for some $k^{\prime} / k$ finite. In particular, for $m$ such that $\sigma_{\alpha}^{m} \in G_{k^{\prime}} \subset$ $G_{k}, \sigma_{\alpha}^{m}$ acts on $\operatorname{End}(V)$ so that the morphism
4. Identity component is described by the wrong relations
- ID:
96e3a184-d80e-4824-9917-9725acbc4fe3 - Refine score:
0.32 - Original types: general
- Refine status: open
Comment
In Step 1 of Lemma 2.10, the displayed condition describes the characters trivial on the full Zariski closure $H=\overline{\{\gamma^n\}}$, not necessarily those trivial on its identity component $T=H^\circ$. A character is trivial on $T$ when its value $\prod_i\lambda_i^{a_i}$ at $\gamma$ is a root of unity, rather than only when it equals $1$. The desired conclusion is recoverable: the proposed one-parameter subgroup lies in $H$ by the displayed relations and, being connected and containing the identity, lies in $H^\circ=T$. Additionally, the weight and absolute-value argument still places $T'$ in $T$, since every root of unity has complex absolute value $1$, meaning this is a local gap rather than a failure of the lemma. Nevertheless, the finite component group must be accounted for in the stated character-lattice argument.
Quoted passage
Let $X^{*}(D), X^{*}(T)$ be the character lattices of $D, T$ respectively; identify $X^{*}(D) \simeq \mathbb{Z}^{m}$ via the basis $\left\{e_{i}\right\}$. The inclusion $T \hookrightarrow D$ induces a surjection $X^{*}(D) \rightarrow X^{*}(T)$, with kernel $K$ given by $\underline{a} \in \mathbb{Z}^{m}$ such that
$$ \prod_{i=1}^{m} \lambda_{i}^{a_{i}}=1 . $$$T$ is precisely the torus cut out by the characters in $K$, i.e. the subtorus given by diagonal matrices $M$ such that $\chi(M)=1$ for all $\chi \in K$. But in particular, this holds for the matrices
5. The inertia logarithm is not a finite sum
- ID:
2f7f7b3b-6a19-4858-85f7-91f622e7a26a - Refine score:
0.23 - Original types: general
- Refine status: open
Comment
The statement that the logarithm is a finite sum merely because $\iota_{x_i}(\gamma)-1\in\mathscr I$ is literally incorrect: membership in the augmentation ideal gives $\mathscr I$-adic convergence, not nilpotence. The construction remains valid because its target is $\mathscr I/\mathscr I^n$, where the series truncates after degree $n-1$; equivalently, one may form the convergent logarithm in the completed Mal’cev algebra and then project to that quotient.
Quoted passage
Here $\log \left(\iota_{x_{i}}(\gamma)\right)$ is the power series
$$ \log \left(\iota_{x_{i}}(\gamma)\right)=\log \left(1+\left(\iota_{x_{i}}(\gamma)-1\right)\right)=\sum_{j=1}^{\infty}(-1)^{j+1} \frac{\left(\iota_{x_{i}}(\gamma)-1\right)^{j}}{j}, $$which is in fact a finite sum because $\left(\iota_{x_{i}}(\gamma)-1\right) \in \mathscr{I}$.
6. Example 3.1 overstates integral eigenvectors
- ID:
a60e23bb-d311-4322-b2d7-2f9de7d96091 - Refine score:
0.25 - Original types: general
- Refine status: open
Comment
The assertion about integral eigenvectors fails when $\chi(\sigma)$ is a nontrivial root of unity. If $\chi(\sigma)=-1$, for example, then $T^2/(1+T)\in\mathbb{Z}_\ell[[T]]$ is a nonconstant $\sigma$-invariant eigenvector. Moreover, all nonzero constants are eigenvectors, so even in the infinite-order case the statement is accurate only up to scalar multiples of $1$.
Quoted passage
The elements
$$ (\log (1+T))^{n} \in \mathbb{Q}_{\ell}[[T]], \quad n \in \mathbb{Z}_{\geq 0} $$are $\sigma$-eigenvectors with eigenvalue $\chi(\sigma)^{n}$; their span is dense in $\mathbb{Q}_{\ell}[[T]]$ for the $\left(T\right.$ )-adic topology, as $(\log (1+T))^{n}$ has leading term $T^{n}$. On the other hand, if $\chi(\sigma) \neq 1$, the only $\sigma$-eigenvector in $\mathbb{Z}_{\ell}[[T]]$ is 1 .
7. Theorem 3.6’s claimed equivalence loses density
- ID:
fcaa1617-9e18-4113-b4f0-3fbf80c6e5bb - Refine score:
0.28 - Original types: general
- Refine status: open
Comment
The sentence beginning “Equivalently” does not capture the full Gauss-norm density assertion. Pointwise linear denominator growth only places each eigenvector in some convergent group ring; it neither supplies a uniform radius for all eigenvectors nor implies density in the stronger Gauss-norm topology. The proof correctly establishes these additional properties separately.
Quoted passage
Moreover $W^{-n} \mathbb{Q}_{\ell}\left[\left[\pi_{1}^{\ell}\left(X_{\bar{k}}, \bar{x}\right)\right]\right]^{\leq \ell^{-r}}$ admits a set of $\sigma_{\alpha}$-eigenvectors with eigenvalues in $\left\{\alpha^{n}, \alpha^{n+1}, \ldots\right\}$ and with $\mathbb{Q}_{\ell}$-span dense in the topology defined by the $r$-Gauss norm.
Equivalently, if $y$ is a $\sigma_{\alpha}$-eigenvector in $\mathbb{Q}_{\ell}\left[\left[\pi_{1}^{\ell}\left(X_{\bar{k}}, \bar{x}\right)\right]\right],-v_{n}\left(\pi_{n}(y)\right)$ grows at most linearly in $n$. We require several lemmas before giving the proof.
8. Gauss-norm decay in Theorem 3.6 is not justified
- ID:
50f6af55-414d-4846-a233-a41dcf54ac80 - Refine score:
0.74 - Original types: general
- Refine status: open
Comment
The Gauss-norm decay used to prove density in Theorem 3.6 is not valid as written. Proposition 2.7 does not justify restricting the supremum to $n\geq i$ or gaining a factor of $\ell^{-r}$ after every single weight step. For $X=\mathbb G_m$, one has $W^{-2d}=W^{-2d+1}=(T^d)$; taking $z=T$ and $w_0=\log(1+T)$ gives $w_1=0$ and $z_2=z_1\neq0$, contradicting the asserted one-step contraction. Thus the displayed estimates do not establish Gauss-norm density without a revised argument accounting for the approximately two-to-one comparison between the weight and augmentation filtrations.
Quoted passage
Note that, by the estimates in the previous two paragraphs, we have that if $\tilde{y} \in \operatorname{gr}_{W}^{-i} \mathbb{Z}_{\ell}\left[\left[\pi_{1}^{\ell}\left(X_{\bar{k}}, \bar{x}\right)\right]\right]$, then
$$ \begin{aligned} \left|s_{i}(\tilde{y})\right|_{r} & =\sup _{n}\left\{\ell^{-v_{n}\left(\pi_{n}\left(s_{i}(\tilde{y})\right)\right)-n r}\right\} \\ & \leq \sup _{n \geq i}\left\{\ell^{C(\alpha, \ell, 2 n-i-1)-n r}\right\} \\ & =\ell^{C(\alpha, \ell, i-1)-i r} \end{aligned} $$Note that this value tends to zero monotonically with $i$. We now check that the $\mathbb{Q}_{\ell}$-span of the $\sigma_{\alpha}$-eigenvectors in $W^{-i} \mathbb{Q}_{\ell}\left[\left[\pi_{1}^{\ell}\left(X_{\bar{k}}, \bar{x}\right)\right]\right]^{\leq \ell^{-r}}$ are dense in the topology defined by the Gauss norm. It suffices to show that $W^{-i} \mathbb{Z}_{\ell}\left[\left[\pi_{1}^{\ell}\left(X_{\bar{k}}, \bar{x}\right)\right]\right]$ is in the closure of the $\mathbb{Q}_{\ell}$-span of the $\sigma_{\alpha}$-eigenvectors, as the $\mathbb{Q}_{\ell}$-span of $W^{-i} \mathbb{Z}_{\ell}\left[\left[\pi_{1}^{\ell}\left(X_{\bar{k}}, \bar{x}\right)\right]\right]$ is evidently dense in the Gauss norm topology. Given $z \in W^{-i} \mathbb{Z}_{\ell}\left[\left[\pi_{1}^{\ell}\left(X_{\bar{k}}, \bar{x}\right)\right]\right]$, let $z_{0}=z, w_{0}=s_{i}\left(z_{0} \bmod W^{-i-1}\right)$ and in general,
$$ z_{j}=z_{j-1}-w_{j-1}, w_{j}=s_{i+j}\left(z_{j} \bmod W^{-i-j-1}\right) . $$Then the $w_{j}$ are $\sigma_{\alpha}$-eigenvectors, so it suffices to show that
$$ \sum w_{j} \rightarrow z, $$or equivalently that $\left|z_{j}\right|_{r} \rightarrow 0$. This follows from the estimates in the previous paragraph, which yield
$$ \begin{aligned} \left|z_{j}\right|_{r} & \leq \max \left\{\left|z_{j-1}\right|_{r} \cdot \ell^{-r},\left|w_{j-1}\right|\right\} \\ & \leq \max \left\{\left|z_{j-1}\right|_{r} \cdot \ell^{-r},\left|z_{j-1}\right|_{r} \cdot \ell^{C(\alpha, \ell, i+j)-(i+j) r}\right\} . \end{aligned} $$ $\square$
9. Remark 4.3 does not yield N=1 when ell=3
- ID:
6ca5e44d-d719-49b8-b56c-85ca0c2497a3 - Refine score:
0.48 - Original types: general
- Refine status: open
Comment
The stated deduction of $N(X,\ell)=1$ is not supported by Remark 3.11 for a mixed-weight affine curve when $\ell=3$. For a topological generator $\alpha\in\mathbb{Z}_3^\times$, the available general estimate is $r_\alpha=2C(\alpha,3,1)=3/2$; the proof of Theorem 1.2 therefore gives $N=2$, not $N=1$. The $N=1$ assertion in this case requires a sharper estimate or an additional qualification.
Quoted passage
If this index is 1 for some $\ell>2$ (as is expected to be the case for almost all $\ell$ (see e.g. [37, § 10], [41], and [7] for discussion of the case where $H^{1}\left(X_{\bar{k}}, \mathbb{Z}_{\ell}\right)$ is pure of weight 1 -as far as we know, the mixed case has not been conjectured in the literature, though it seems natural to do so), we may take $N(X, \ell)=1$, by choosing $\alpha$ in Theorem 2.8 to be a topological generator of $\mathbb{Z}_{\ell}^{\times}$, and using Remark 3.11.
10. Question 4.7 lacks a Frobenius action at the basepoint
- ID:
36239f61-6ca2-4ad1-8656-e2579e60d09f - Refine score:
0.3 - Original types: general
- Refine status: open
Comment
As stated, Question 4.7 does not specify an actual Frobenius endomorphism of the based pro-$\ell$ group. For a general geometric basepoint, $G_k$ acts only outerly; a Frobenius lift or compatible basepoint/path produces an endomorphism, but different choices differ by inner twists. Because dense spanning by literal eigenvectors is a property of an actual linear operator and is not shown to be invariant under those twists, the intended Frobenius action is not fully defined.
Quoted passage
A final remark: it is natural to ask whether the hypothesis (†) in Theorem 1.4 is necessary; it fails, for example, for many curves over finite fields. We could salvage this situation if the following question has a positive answer:
Question 4.7 Let $X$ be a smooth curve over a finite field $k$, and let $\ell$ be a prime different from the characteristic of $k$. Does there exist an $r=r(X)$ such that $\mathbb{Q}_{\ell}\left[\left[\pi_{1}^{\ell}\left(X_{\bar{k}}, \bar{x}\right)\right]\right]^{\leq \ell^{-r}}$ admits a set of Frobenius eigenvectors with dense span?
Scope
- Paper:
03 Published and Submitted Work/Published/P19_Litt_Arithmetic_Representations_I.pdf - Refine report:
.refine/results/Published/P19_Litt_Arithmetic_Representations_I.review.json - Rubric:
rubric/comment_triage_rubric.md - Version assessed: local published PDF, Inventiones mathematicae 214 (2018), 605-639
- Detailed Refine comments assessed: 10
- Assessment date: 2026-07-30
The local published PDF is authoritative. PDF pages 2, 4-5, 12, 17, 19, 22, 26-27, 30, and 33 were rendered and visually inspected. The questioned ranks, character relations, logarithm wording, filtration indices, norm estimates, and basepoint language are all present in the published typesetting.
Summary
| # | Short title | Validity | Category | Standardness | Impact | Challenge | Repair | Disposition | Priority | Confidence |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Ambient rank in arithmeticity | V4 | C4 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 2 | Finite-level semisimplicity | V3 | C9 | E-NA | I1 | Q0 | R1 | D1 | P3 | HIGH |
| 3 | Socle reduction in the proof sketch | V4 | C9 | E-NA | I1 | Q0 | R1 | D1 | P3 | HIGH |
| 4 | Relations cutting out \(H^\circ\) | V4 | C6 | E4 | I2 | Q2 | R2 | D3 | P2 | HIGH |
| 5 | Logarithm is convergent, not finite | V4 | C4 | E-NA | I1 | Q0 | R1 | D1 | P3 | HIGH |
| 6 | Integral eigenvectors in Example 3.1 | V4 | C8 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 7 | False equivalence in Theorem 3.6 | V4 | C9 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 8 | Gauss-norm density argument | V4 | C6 | E4 | I3 | Q3 | R3 | D4 | P0 | HIGH |
| 9 | The \(\ell=3\) bound in Remark 4.3 | V4 | C5 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 10 | Frobenius action in Question 4.7 | V4 | C4 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
Independent challenges are complete for Comments 4 and 8. Comment 4 has a complete local repair (Q2). Comment 8 requires replacing the density portion of the proof of Theorem 3.6 with a spectral-projector argument (Q3); it remains I3 because Theorem 3.6 is a central input to the main results and the repair is substantial.
1. Definition 1.1 needs an independent ambient rank
Comment ID: e848fe0d-1517-41b3-a481-bb00469d4a77 Location: PDF p. 2, Definition 1.1; compare the abstract, Definition 1.5, Lemma 4.1, and the proof of Theorem 1.2.
Both \(\rho\) and \(\widetilde\rho\) are printed with target \(\operatorname{GL}_n(\mathbb Z_\ell)\). After tensoring with \(\mathbb Q_\ell\), a rank-\(n\) free subquotient of a rank-\(n\) representation has full dimension, so this does not express the advertised notion of a constituent of a larger arithmetic representation. Definition 1.5 and the geometric corollary require the larger ambient rank.
The proof also reveals the intended correction. In Lemma 4.1, the \(\pi_1(X_{k'})\)-span of the conjugates of an irreducible rank-\(n\) constituent can have rank larger than \(n\). The resulting representation \(\beta\) is then used only as an ambient representation from which \(\rho\) is recovered as a subquotient.
- Classification:
V4/C4/I2 - Repair: write \[ \widetilde\rho:\pi_1^{\mathrm{ét}}(X_{k'},\bar x) \longrightarrow\operatorname{GL}_m(\mathbb Z_\ell) \] for an unrestricted \(m\), and make the same rank change for \(\gamma,\beta\) and \(M_m\) in Lemma 4.1 and the proof of Theorem 1.2
- Dependency trace: the proof already works with the larger \(\pi_1(X_{k'})\)-span; the main theorem statements and geometric application retain their advertised scope
- Disposition:
R2/D3;P2/HIGH
2. The introduction uses a topological/pro-semisimple shorthand
Comment ID: a844ee35-dd45-45da-a512-4785ff531b56 Location: PDF p. 4, Step 1 of the proof overview; Theorem 2.20 on PDF p. 18.
The formal statement of Theorem 2.20 is that Frobenius acts semisimply on every quotient
The overview suppresses “modulo \(\mathscr I^n\)” and says it acts semisimply on the completed algebra. In the ordinary algebraic sense, the latter would require the entire infinite-dimensional vector space to be the direct sum of its eigenspaces, which is stronger and generally false. What is true is finite-level semisimplicity, or topological/pro-semisimplicity with a dense \(\mathscr I\)-adic eigenvector span.
The comment is classified V3, rather than V4, because “semisimple” for this pro-object can naturally be read in precisely that finite-level/topological sense, and the cited theorem states the scope unambiguously. The overview is imprecise, not a false mathematical dependency.
- Classification:
V3/C9/I1 - Repair: say “Frobenius acts semisimply on every finite quotient; equivalently for the present purpose, the completed algebra has an \(\mathscr I\)-adically dense span of eigenvectors”
- Disposition:
R1/D1;P3/HIGH
3. Step 3 suppresses the replacement by an ambient representation
Comment ID: 08c49c98-4c46-435c-88c2-08ff350a24e6 Location: PDF p. 5, Step 3 of the proof overview; full argument on PDF pp. 27-29.
The overview says that a socle argument lets one assume the original \(\rho\) extends to \(\pi_1(X_{k'})\). Arithmeticity gives only that \(\rho\) is a subquotient of such a representation. The full proof correctly:
- passes through the socle filtration;
- reduces to an irreducible constituent;
- applies Lemma 4.1 to construct an ambient \(\beta\) whose geometric restriction is still trivial modulo \(\ell^N\);
- proves \(\beta\) unipotent and then passes unipotence to \(\rho\).
- Classification:
V4/C9/I1 - Repair: replace “we may assume that \(\rho\) extends” by “after reducing to an irreducible constituent, Lemma 4.1 lets us replace \(\rho\) by an ambient \(\beta\) which extends and has \(\rho\) as a subquotient”
- Dependency trace: the full proof already contains the correct reduction
- Disposition:
R1/D1;P3/HIGH
4. Step 1 of Lemma 2.10 ignores the finite component group
Comment ID: 96e3a184-d80e-4824-9917-9725acbc4fe3 Location: PDF p. 12, Step 1 of Lemma 2.10.
Let
The printed lattice
cuts out \(H\), not necessarily \(T\). A character is trivial on \(T\) exactly when its value on \(\gamma\) is a root of unity. The distinction matters when \(H/T\) is nontrivial.
For an explicit disconnected test, take \(\gamma=(2,-2)\in\mathbb G_m^2\). Then
The character \(y/x\) is trivial on \(H^\circ\) but takes the root-of-unity value \(-1\) on \(\gamma\); only its square belongs to the printed equality-to-one lattice. Thus the finite-component issue is real and cannot be removed by interpreting the printed \(K\) as the character kernel of \(H^\circ\).
Severity challenge and local repair
Keep the displayed lattice, but call it \(K_H\). If \(\prod_i\lambda_i^{a_i}=1\), taking complex absolute values gives
Therefore every character in \(K_H\) vanishes on
It follows first that \(T'\subset H\). Since \(T'\) is connected and contains the identity, \(T'\subset H^\circ=T\), which is the required conclusion.
Equivalently, one can use the actual kernel for \(T\): its characters take root-of-unity values on \(\gamma\), and the same absolute-value calculation works because every root of unity has complex absolute value one.
- Classification:
V4/C6/E4 - Severity challenge:
Q2; both repairs fully account for the component group - Impact:
I2 - Dependency trace: the repaired Lemma 2.10 still supplies every \(\sigma_\alpha\) used by Theorems 2.12 and 2.8. Theorem 3.6's eigenvector-density statement, and hence Theorems 1.2 and 1.4 and Corollary 1.6, use only that unchanged conclusion. All remain valid once the intermediate containment is written as \(T'\subset H\), then \(T'\subset H^\circ\)
- Disposition:
R2/D3;P2/HIGH
5. The inertia logarithm is infinite but convergent
Comment ID: 2f7f7b3b-6a19-4858-85f7-91f622e7a26a Location: PDF p. 17, proof of Theorem 2.12.
Membership of \(\iota_{x_i}(\gamma)-1\) in the augmentation ideal gives \(\mathscr I\)-adic convergence of
not nilpotence in the completed algebra. The intended map has target \(\mathscr I/\mathscr I^n\), where all terms of degree at least \(n\) vanish, so the projected sum is finite and every subsequent calculation is valid.
- Classification:
V4/C4/I1 - Repair: replace “is in fact a finite sum” by “converges \(\mathscr I\)-adically; its image modulo \(\mathscr I^n\) is the finite sum through degree \(n-1\)”
- Disposition:
R1/D1;P3/HIGH
6. Example 3.1 needs an infinite-order hypothesis
Comment ID: a60e23bb-d311-4322-b2d7-2f9de7d96091 Location: PDF p. 19, Example 3.1.
When \(\chi(\sigma)=-1\), the action sends
and the nonconstant integral series
is fixed. Thus the assertion fails for nontrivial torsion values of \(\chi(\sigma)\). Even when \(\chi(\sigma)\) has infinite order, every constant in \(\mathbb Z_\ell\), not only \(1\), is an eigenvector.
- Classification:
V4/C8/I2 - Repair: say “if \(\chi(\sigma)\) has infinite order, the integral eigenvectors are precisely the constant series” or explicitly work only up to nonzero scalar
- Dependency trace: the elements \(\sigma_\alpha\) used later have \(\alpha\) not a root of unity, so no theorem uses the false torsion case
- Disposition:
R2/D3;P2/HIGH
7. Linear denominator growth is not equivalent to Gauss-density
Comment ID: fcaa1617-9e18-4113-b4f0-3fbf80c6e5bb Location: PDF p. 22, statement of Theorem 3.6.
For a fixed eigenvector \(y\), linear growth of \(-v_n(\pi_n(y))\) says that \(y\) belongs to some convergent group ring. It does not provide one common radius for all eigenvectors, and it does not show that their span is dense in the corresponding Gauss norm. Those are separate uniform and approximation assertions.
- Classification:
V4/C9/I2 - Repair: replace “Equivalently” by “In particular”; retain the uniform-radius and Gauss-density clauses as separate conclusions
- Dependency trace: the main proof uses the actual uniform density conclusion, not the stated false converse
- Disposition:
R2/D3;P2/HIGH
8. The one-weight-step Gauss-norm estimate is false
Comment ID: 50f6af55-414d-4846-a233-a41dcf54ac80 Location: PDF pp. 26-27, density portion of the proof of Theorem 3.6.
Proposition 2.7 gives
Thus an element of \(W^{-i}\) projects automatically to zero modulo \(\mathscr I^n\) only for roughly \(n<i/2\), not for all \(n<i\). The restriction of the norm supremum to \(n\ge i\) and the claimed factor \(\ell^{-r}\) after each single weight step are unsupported.
The example \(X=\mathbb G_m\) makes the failure literal:
Starting with \(z=T\), the first eigencomponent is \(\log(1+T)\); the next odd-weight component is zero, so the residual does not contract on that step.
Severity challenge: replacement by spectral projectors
Let
with its completed filtrations. On \(A/W^{-m}\), Theorems 2.8 and 2.12 give a semisimple \(\sigma_\alpha\)-action with distinct eigenvalues \(1,\alpha,\ldots,\alpha^{m-1}\). For \(0\le i<m\), use the spectral projector
Its numerator preserves the integral lattice. Its denominator has valuation
by Lemmas 3.9-3.10.
For \(z\in A_{\mathbb Z_\ell}\), the compatible projectors define its \(\alpha^i\)-eigencomponent \(w_i\in W^{-i}A\). To estimate \(\pi_n(w_i)\), use \(m=2n+1\), because \(W^{-2n-1}\subset\mathscr I^n\). For \(n<\lceil i/2\rceil\), this projection is zero; otherwise
Writing \(C(\alpha,\ell,k)\le ck+b\) with \(c\le C(\alpha,\ell,1)\), one obtains, for \(r>2C(\alpha,\ell,1)\),
The finite sums \(\sum_{i<m}w_i\) agree with \(z\) modulo \(W^{-m}\). The bound above makes the series \(\sum_iw_i\) converge to \(z\) in the \(r\)-Gauss norm. If \(z\in W^{-a}\), only the components with \(i\ge a\) occur, giving exactly the eigenvalue range \(\{\alpha^a,\alpha^{a+1},\ldots\}\). Finally, \(A_{\mathbb Z_\ell}\otimes\mathbb Q_\ell\) is dense in the convergent group ring, so these finite eigenvector sums prove the full density statement.
Failure tests and dependencies
- Two-to-one filtration comparison: handled by \(m=2n+1\) and the lower limit \(n\ge\lceil i/2\rceil\).
- Repeated weight steps for \(\mathbb G_m\): harmless; the missing odd components simply vanish, while the even components still decay.
- \(\ell=2\): Lemma 3.10 has an additive constant, absorbed into \(b\); its slope remains below the chosen \(r/2\).
- Coincident eigenvalues: excluded because \(\alpha\) is not a root of unity.
- Weight-truncated density: preserved because the spectral projectors respect \(W^\bullet\).
- Classification:
V4/C6/E4 - Severity challenge:
Q3; the replacement argument is complete but substantially different from the printed recursion - Impact:
I3 - Repairability:
R3 - Dependency trace: the repaired Theorem 3.6 supplies the same radius \(r_\alpha=2C(\alpha,\ell,1)\) and the same density/eigenvalue conclusions used in Theorems 1.2 and 1.4 and Corollary 1.6
- Disposition:
D4;P0/HIGH; author and specialist review recommended. The independent challenge in this assessment confirms the projector compatibility, denominator bound, half-weight cutoff, and \(\ell=2\) case.
9. Remark 4.3's bound fails at \(\ell=3\)
Comment ID: 6ca5e44d-d719-49b8-b56c-85ca0c2497a3 Location: PDF p. 30, Remark 4.3; compare Lemma 3.10 and Remark 3.11 on PDF pp. 23-27.
For a topological generator \(\alpha\in\mathbb Z_3^\times\), its order modulo three is \(s=2\) and \(v_3(\alpha^2-1)=1\). Hence
The proof of Theorem 1.2 takes \(N\) to be the least integer strictly greater than \(r_\alpha\), namely \(N=2\), not \(1\).
- Classification:
V4/C5/I2 - Repair: change “for some \(\ell>2\)” to “for some \(\ell\ge5\),” or state separately that the displayed general estimate only gives \(N(X,3)=2\); an \(N=1\) assertion at \(\ell=3\) would need a sharper bound
- Dependency trace: only the explicit optimal bound in Remark 4.3 changes; Theorems 1.2 and 1.4 still provide an \(N\)
- Disposition:
R2/D3;P2/HIGH
10. Question 4.7 does not specify a based Frobenius operator
Comment ID: 36239f61-6ca2-4ad1-8656-e2579e60d09f Location: PDF p. 33, Question 4.7.
For a general geometric basepoint \(\bar x\), the arithmetic fundamental-group sequence gives an outer \(G_k\)-action on \(\pi_1^\ell(X_{\bar k},\bar x)\), not a distinguished endomorphism. A \(k\)-rational basepoint, a rational tangential basepoint, or a chosen lift of Frobenius together with a path supplies an actual operator. Different choices can differ by inner automorphisms, and the paper does not show that dense span by literal eigenvectors is invariant under such a change.
- Classification:
V4/C4/I2 - Repair: assume \(x\in X(k)\) and take \(\bar x\) to be its associated geometric point, or explicitly include a chosen Frobenius lift/path in the question
- Dependency trace: this affects only the formulation of the final open question, not a theorem or proof
- Disposition:
R2/D3;P2/HIGH
Action queue
- Send comment 8 and its independently challenged spectral-projector repair for author/specialist review before choosing the form of a corrigendum.
- Independently challenge comment 4's component-group repair.
- Add living-errata entries for comments 1, 4, and 6-10.
- Treat comments 2, 3, and 5 as maintained-copy editorial improvements.
P20 NON-ABELIAN LEFSCHETZ HYPERPLANE THEOREMS32 detailed comments · 30 numbered corrections 1 I01 I126 I24 I4
These errata refer to the version published in Journal of Algebraic Geometry 27 (2018), 593--646, doi:10.1090/jag/704. Page numbers below refer to that version.
Page 595, final bullet in the list of applications. The K3-type alternative is too broad: Theorem 6.22 does not cover general noncompact orthogonal quotients. Delete the sub-bullet “$\mathscr H$ is of K3 type.” After the remaining sub-bullets, insert:
For variations arising from the polarized hyperk\"ahler moduli situation, the analogous extension statement follows from Theorem 6.16. In particular, the K3-type case asserted here is covered when the relevant quotient is compact, or when it arises from that hyperk\"ahler moduli construction.
No assertion is made here for a general noncompact K3-type orthogonal quotient.
Pages 598--599, Theorem 1.10. The positive-characteristic alternative omits the dimension bound used in Theorems 4.11, 4.29, and 5.7. Replace
“$k$ is perfect of characteristic $p>0$, and $X$ lifts to $W_2(k)$”
by
“$k$ is perfect of characteristic $p>\dim(X)$, and $X$ lifts to $W_2(k)$.”
The same inequality is to be included whenever the positive-characteristic branch of Theorem 1.10 is summarized. With this change, the cited vanishing and uniqueness results apply.
Pages 602--603 and 625, Theorems 1.20 and 4.29. These statements omit the projectivity of $X$, which is required by the vanishing, algebraization, and extension results used in their proofs. In the opening sentence of each theorem, replace “a smooth $L$-variety” by “a smooth projective $L$-variety.” The remaining occurrences of $X$ in these two statements and proofs are to be understood with this hypothesis.
Pages 602--603, 617, and 625, Theorems 1.19, 1.20, 4.12, and 4.29. The positive-characteristic statements mix the base fields $k$ and $L$. Make the following replacements:
In Theorems 1.19 and 4.12, replace “$Y$ a smooth $k$-variety” and “$f\colon D\to Y$ a morphism” by “$Y$ a smooth $L$-variety” and “$f\colon D\to Y$ an $L$-morphism.”
In Theorems 1.20 and 4.29, make the same replacements, retaining the projectivity correction to $X$ above.
Thus all relative Frobenius maps $F_{Y/L}$ and twists $Y^{(p^k)}$ in these statements and proofs are formed over $L$.
Pages 603 and 625, Theorem 1.20(2) and Theorem 4.29(3). The no-rational-curves alternative also requires the target to be proper, as is needed in Proposition 3.5. Replace it in both statements by:
$Y$ is proper and $\overline Y_L$ contains no rational curves.
Here $Y$ has the $L$-structure specified in the base-field correction above. With properness added, Proposition 3.5 supplies the final extension step.
Page 604, paragraph preceding Theorem 1.22; compare pages 637--638 and 642. Theorem 1.22 is not literally Theorem 6.32: its finite-cover hypothesis and its bound on $\dim(Y)$ differ from the hypotheses of Theorem 6.32. Replace “In Theorem 6.32, we improve this result to” by the following argument:
The following finite-cover variant is obtained as follows. Given a finite surjective \'etale morphism $Y'\to Y$ with $Y'$ a scheme, form $D'=D\times_Y Y'$. By the Lefschetz theorem for finite \'etale covers, $D'\to D$ extends uniquely to a finite \'etale cover $X'\to X$. The map $D'\to Y'$ satisfies
\[ \dim(\operatorname{im}(D'\to Y')) \leq \dim(Y)<\dim(D')-1. \]Lemma 6.31 gives
\[ \phi\bigl(\NN_{D'/X'}\otimes f'^*\Omega^1_{Y'}\bigr) <\dim(D')-1, \]so Theorem 5.1(1) extends $f'\colon D'\to Y'$ uniquely to $X'$. On $X'\times_X X'$, the two pullbacks agree along the inverse image of $D$ and hence agree everywhere by the uniqueness part of Theorem 5.1. The cocycle condition follows in the same way, and finite \'etale descent gives a unique map $X\to Y$ extending $f$.
This proves Theorem 1.22, subject to the reducedness qualification for existence statements recorded below, without identifying it with Theorem 6.32.
Pages 604--605, Remark 1.24. The torsion condition on the displayed cokernel implies that global one-forms generically span the cotangent bundle; it does not imply that the Albanese map is finite. Replace “finite-to-one map to an Abelian variety” by “generically finite map onto its image in an Abelian variety” in the first sentence. Make the corresponding replacement in the last paragraph of the remark: such targets need not admit finite maps to Abelian varieties.
Pages 605--606, Lemma 2.1. The last biduality step uses the perfectness of $Rf_*(\mathscr F\otimes(\mathscr L^\vee)^{\otimes n})$, which does not follow from the printed hypotheses. Replace the opening sentence by:
Let $S$ be a Noetherian scheme. Let $f\colon X\to S$ be a projective perfect morphism with dualizing complex $\omega_{X/S}=f^!\OO_S$ having coherent cohomology concentrated in degrees $[-n,-m]$.
Here “perfect” may equivalently be replaced by “of finite Tor-dimension.” Proper perfect pushforward preserves perfect complexes, so the existing biduality argument applies. The later applications over a field satisfy this hypothesis.
Page 608, Corollary 2.6. Corollary 2.5 identifies relative cohomology sheaves, rather than global Ext groups over an arbitrary base $S$. Replace the final displayed map and the sentence following it by:
Writing $h=f\circ g\colon Y\to S$ and $\widehat h\colon\widehat Y_D\to S$ for the induced morphism, the natural map of relative Ext sheaves
\[ \mathcal H^i\!\left( Rh_*R\mathcal Hom_Y(\mathscr F,\mathscr G)\right) \longrightarrow \mathcal H^i\!\left( R\widehat h_*R\mathcal Hom_{\widehat Y_D} (\widehat{\mathscr F},\widehat{\mathscr G})\right) \]is an isomorphism for $0\leq i\leq m-a-2$.
When $S=\Spec(k)$, this is the global Ext statement used later.
Page 609, proof of Corollary 2.7. The positivity sign in the choice of presentation is reversed. Replace “with $m_1-m_2\gg0$” by “with $m_2-m_1\gg0$.” This agrees with the preceding construction and makes the relevant Hom bundle sufficiently positive for Serre vanishing.
Pages 611--612, Corollary 3.4. The proof invokes Corollary 3.2, whose source is locally $\mathbb Q$-factorial. Replace the opening words by:
Let $X$ be a normal, locally $\mathbb Q$-factorial projective $k$-variety, and let $Y$ be a quasi-projective $k$-variety.
The later applications have $X$ smooth and therefore satisfy the added hypothesis.
Page 612, proof of Proposition 3.5. The printed proof neither justifies the quasi-finiteness of $b\colon X'\to X$ nor names the correct base of the ensuing finite morphism. Replace the paragraph beginning “Let $X'\to Y$ be the map given by $b'\circ\widetilde s$” through the end of the proof by:
Let $j\colon X'\to Y$ be the morphism induced by the normalized closure. This morphism is finite. If $b$ had an exceptional divisor, a rational curve $C$ through its general point would be contracted by $b$. Since $f\circ j=b$, the curve $j(C)$ would lie in a geometric fiber of $f$. That fiber contains no rational curves, so $j(C)$ would be a point, contradicting the finiteness of $j$. Thus $b$ has no exceptional divisor. Purity then implies that $b$ is quasi-finite. Since $b$ is proper, it is finite over $X$; since it is also birational and $X$ is normal, it is an isomorphism. The morphism $j$ therefore gives the desired section. \qed
Page 613, Corollary 4.2. The statement omits the embedding and thickening hypotheses needed to define the normal bundle and obstruction class. Replace its opening sentence by:
Let $D\hookrightarrow X$ be a closed lci subscheme of schemes over a field $k$, with ideal sheaf $\II_D$, and let $D_2=V(\II_D^2)$. Let $Y$ be an arbitrary smooth $k$-scheme.
The remainder of the statement then applies to a morphism $s\colon D\to Y$ and its extensions to this specified first infinitesimal thickening $D_2$.
Page 615, Theorem 4.4. The range in the second Le Potier vanishing is misstated. Replace
\[ H^i(X,\Omega_X^p\otimes E)=0\qquad\text{for }i+p\geq n-e \]by
\[ H^i(X,\Omega_X^p\otimes E)=0\qquad\text{for }i+p\geq n+e. \]The argument following the theorem uses the other displayed Le Potier vanishing and is unchanged.
Pages 615--616, Theorem 4.6. The opening bound $\dim(D)\geq2$ makes the first bullet's dimension-one case vacuous, and the analytic positivity input requires the complex setting. Replace the statement by:
Theorem 4.6. Let $X$ be a projective complex variety, and let $D\subset X$ be a smooth lci subscheme with ample normal bundle. Let $\widehat D$ be the formal scheme obtained by completing $X$ at $D$. Let $Y$ be a smooth complex variety with Nakano semipositive cotangent bundle. Given a morphism $f\colon D\to Y$,
there is at most one extension of $f$ to a morphism $\widehat D\to Y$ if $\dim(D)\geq1$; and
such an extension exists if $\dim(D)\geq2$.
The printed proof applies separately in these two ranges.
Page 619, final paragraph of the proof of Theorem 4.12. The displayed duality has the wrong dual and cohomological degree for the normalization of $K_D$. Let
\[ A=\operatorname{Frob}_p^{k*} \bigl(f^*T_Y\otimes\OO_D(-D)\bigr)\otimes\OO_D(-sD). \]Replace the paragraph beginning “Recall also that $\OO_D(-D)=\NN_{D/X}^{\vee}$” by:
Recall that $\OO_D(-D)=\NN_{D/X}^{\vee}$. Grothendieck duality gives
\[ H^i(D,A)^\vee \simeq \mathbb H^{-i}\bigl(D,K_D\otimes A^\vee\bigr), \]where
\[ A^\vee= \OO_D(sD)\otimes\operatorname{Frob}_p^{k*} \bigl(f^*\Omega_Y^1\otimes\NN_{D/X}\bigr). \]In the hypercohomology spectral sequence
\[ H^a\bigl(D,\mathcal H^b(K_D)\otimes A^\vee\bigr) \Longrightarrow \mathbb H^{a+b}\bigl(D,K_D\otimes A^\vee\bigr), \]a term contributing to total degree $-i$ has $a=-i-b\geq r-i$, because $K_D$ is supported in degrees $[-d,-r]$. For $i=0,1$, the hypothesis
\[ \phi\bigl(f^*\Omega_Y^1\otimes\NN_{D/X}\bigr)<r-1 \]therefore kills every such term for $k\gg0$ (and every $s\geq0$). Hence $H^i(D,A)=0$ for $i=0,1$, as required. \qed
Page 623, Lemma 4.26. Absolute Frobenius is not a $k$-morphism over a general perfect field. Replace the statement by:
Lemma 4.26. Let $k$ be a perfect field of characteristic $p>0$, let $X$ be a normal $k$-variety, and let $Y$ be a $k$-variety. A $k$-morphism $f\colon X\to Y^{(p)}$ factors uniquely as
\[ X\xrightarrow{\bar f}Y\xrightarrow{F_{Y/k}}Y^{(p)} \]if and only if the induced map $f^*\Omega^1_{Y^{(p)}/k}\to\Omega_X^1$ is zero.
In the affine proof, write $Y=\Spec(B)$ and choose $p$th roots of the images of a set of $k$-algebra generators of $B$. The resulting homomorphism $B\to A$ is a $k$-algebra homomorphism and gives $\bar f$; reducedness of $A$ shows that the defining ideal is killed. Equivalently, if perfectness is used to identify the twist with the underlying scheme $Y$, the corresponding semilinear factor sends a scalar $c$ to $c^{1/p}$. This is the scalar rule missing from the printed affine calculation. All subsequent factorizations are to be read using the relative Frobenius $F_{Y/k}$ and its twists.
Pages 624--630, Lemma 4.28 and the existence theorems using Frobenius descent. Lemma 4.28 is false for a nonreduced ample Cartier divisor. Add the hypothesis that $D$ is reduced, and replace its proof by:
Since $D$ is reduced, Frobenius $F_D\colon D\to D$ is a universal homeomorphism and the map on structure sheaves is injective. It is therefore an epimorphism of schemes. Consequently, if $g_1\circ F_D=g_2\circ F_D$, then $g_1=g_2$. \qed
In the existence part of the proof of Theorem 4.29, after obtaining $\bar f=F_{Y/L}\circ h$, replace the appeal to Lemma 4.28 by:
Restriction to $D$ and universal commutativity of Frobenius give
\[ (h|_D)\circ F_D =\bigl(F_{Y/L}^{,k-1}\circ f\bigr)\circ F_D. \]Because $D$ is reduced, $F_D$ is an epimorphism; hence
\[ h|_D=F_{Y/L}^{,k-1}\circ f. \]This completes the induction on $k$.
Add “$D$ reduced” to the existence statements in Theorems 1.20, 4.22, 4.29, 4.31, and 5.1, and to the corresponding existence clauses or applications in Theorems 1.8, 1.10, 1.11, 1.22, 6.1, 6.2, and 6.32. The uniqueness-only statements in Theorems 4.21 and 5.7, and the “at most one” clauses elsewhere, do not require this addition. Theorem 6.35 is also unaffected because it uses a separate deformation argument.
For the characteristic-zero spreading argument in Theorem 4.22, take a flat model of the reduced divisor and shrink the base so that its closed fibers are geometrically reduced. The corrected positive-characteristic theorem then applies to those fibers.
Pages 626--627, final paragraph of the proof of Theorem 4.22. The obstruction module has the wrong denominator, and fiberwise vanishing does not by itself identify a relative obstruction class. Replace the paragraph beginning “In particular, letting $D_n$ be” by:
Let $D_n$ be the $n$th infinitesimal neighborhood of $D$. The obstruction to extending a map from $D_{n-1}$ to $D_n$ lies in
\[ H^1\!\left(D, f^*T_Y\otimes \II_D^{,n-1}/\II_D^{,n}\right). \]After shrinking $S$, cohomology and base change identifies the fibers of the corresponding relative coherent cohomology sheaf with these obstruction groups. The relative obstruction is a section of that sheaf, and its value at every closed point of $S$ is zero because the corresponding fiber map extends by Theorem 4.29. A section of a coherent sheaf that vanishes at every closed point is zero. Thus the obstruction vanishes. Induction on $n$ gives the required morphism $\widehat D\to Y$.
Pages 627--630, Theorem 5.1. The proof uses extension and algebraization results whose source is projective. In the opening sentence, replace “$X$ a smooth $k$-variety” by “$X$ a smooth projective $k$-variety.” The headline Theorem 1.10 already assumes projectivity, so its applications are unchanged.
Pages 628--630, Theorem 5.1(2); compare page 642. The proof of case (2) invokes Corollary 3.4, which only requires a quasi-projective target, and the later statements use that weaker condition. Replace “the coarse space of $\mathscr Y$ is projective” by “the coarse space of $\mathscr Y$ is quasi-projective.” This makes Theorems 5.1, 1.10, and 6.32 consistent.
Pages 629--630, Lemma 5.4. In case (2), the proof extends only an object whose restriction has small image; it does not prove the unrestricted equivalence stated in the lemma. Replace the statement by:
Lemma 5.4. Let $X,D,\mathscr Y$ be as in Theorem 5.1, and let $U\subset X$ be a Zariski-open subset containing $D$. In cases (1) and (3) of that theorem, the restriction functor
\[ \mathscr Y(X)\longrightarrow\mathscr Y(U) \]is an equivalence. In case (2), it is fully faithful, and an object $\xi\in\mathscr Y(U)$ is in its essential image provided that the image of $\xi|_D$ in the coarse space has dimension at most $\dim(D)-2$.
For essential surjectivity in case (2), apply Corollary 3.4 to the particular coarse-space map satisfying this image bound, then use the printed normalization and descent construction. For full faithfulness, extend an isomorphism formally by Corollary 2.10 and then across the complement by the same diagonal and purity argument used in the proof. This is the form needed in Theorem 5.1.
Pages 633--634, Theorem 6.5. The theorem is false for a nonproper smooth morphism: relative global generation alone does not make the pushforward a finite-rank nef Hodge bundle. Replace its statement by:
Theorem 6.5. Let $k$ be a field of characteristic zero, and let $f\colon Y\to X$ be a smooth proper morphism of smooth $k$-varieties. Assume that $f_*\Omega^1_{Y/X}$ is locally free and nef and that its formation commutes with base change. If the evaluation map
\[ f^*f_*\Omega^1_{Y/X}\longrightarrow\Omega^1_{Y/X} \]is surjective, then $\Omega^1_{Y/X}$ is nef.
Indeed, the source of the evaluation map is the pullback of a nef vector bundle, and a quotient of a nef vector bundle is nef. Corollary 6.6 and the subsequent intended applications are in this smooth proper Hodge-theoretic setting.
Page 634, proof of Lemma 6.9. Relative duality uses $R^{n-1}$, rather than $R^1$, on the right. Replace the proof through its final displayed calculation by:
Relative duality gives
\[ (R^1f_*T_{Y/X})^\vee \simeq R^{n-1}f_*(\Omega^1_{Y/X}\otimes\omega_{Y/X}). \]Since the geometric fibers have trivial cotangent bundle, the evaluation map identifies $\Omega^1_{Y/X}$ with $f^*f_*\Omega^1_{Y/X}$. The projection formula and relative duality therefore give
\[\begin{aligned}(R^1f_*T_{Y/X})^\vee &\simeq f_*\Omega^1_{Y/X}\otimes R^{n-1}f_*\omega_{Y/X}\\ &\simeq f_*\Omega^1_{Y/X}\otimes(R^1f_*\OO_Y)^\vee.\end{aligned}\]Both factors are nef Hodge bundles, so their tensor product is nef. \qed
Page 635, proof of Lemma 6.12. The proof reverses the relevant duals and writes $f^*\Omega^1_M$ where the classifying pullback is meant. Let $g\colon X\to M$ be the classifying map and put
\[ W=(f_*\Omega^1_{Y/X})^\vee\otimes R^1f_*\OO_Y. \]Replace the final three sentences of the proof by:
The corrected Lemma 6.9 shows that $W^\vee$ is nef. By assumption, there are injections
\[ R^1f_*T_{Y/X}\hookrightarrow W, \qquad g^*T_M\hookrightarrow R^1f_*T_{Y/X}. \]Dualizing produces surjections
\[ W^\vee\twoheadrightarrow(R^1f_*T_{Y/X})^\vee \twoheadrightarrow g^*\Omega_M^1. \]Thus $g^*\Omega_M^1$ is a quotient of a nef vector bundle and is nef. \qed
Pages 635--636, proof of Theorem 6.16. The displayed chain contains an incorrect pullback and undefined relative tangent and pushforward terms. Let $f\colon Y\to X$ be the hyperk\"ahler family, $g\colon X\to\MM$ its classifying map, and $a\colon A\to X$ its Kuga--Satake Abelian scheme. Replace the two displayed chains in the proof by:
\[ g^*T_{\MM} \longrightarrow R^1f_*T_{Y/X} \longrightarrow R^1a_*T_{A/X} \longrightarrow (a_*\Omega^1_{A/X})^\vee\otimes R^1a_*\OO_A. \]The Kuga--Satake map on Hodge structures, followed by local Torelli, makes this composite injective. The target has nef dual by the corrected Lemma 6.9. Dualizing the injection therefore makes $g^*\Omega^1_{\MM}$ a quotient of a nef bundle, so $g^*\Omega^1_{\MM}$ is nef, as required.
Pages 637--638, proof of Theorem 6.21. Theorem 5.1 only applies when $\dim(X)\geq3$ and therefore does not prove the surjectivity assertion for surfaces. Replace the proof by:
The two assertions are the Lefschetz theorem cited in the statement, [19, Th\'eor\`eme X.3.10]. For $\dim(X)\geq3$, they may also be recovered from Theorem 5.1 by taking $\mathscr Y=BG$ for a finite \'etale group scheme $G$. When $\dim(X)=2$, surjectivity is supplied by the cited theorem; equivalently, one may apply the dimension-two full-faithfulness result of Theorem 5.7 to finite \'etale torsors. \qed
Page 639, proof of Proposition 6.25. The normalization of a rational curve need not be unramified. Replace the second and third sentences of the proof by:
Let $C$ be the image of a nonconstant morphism $\mathbb P^1\to X$, and let $\iota\colon\mathbb P^1\to C\hookrightarrow X$ be its normalization map. The differential
\[ \iota^*\Omega_X^1\longrightarrow\Omega_{\mathbb P^1}^1 \]is generically nonzero. Its image is therefore $\Omega_{\mathbb P^1}^1(-R)$ for an effective ramification divisor $R$. This line bundle has degree $-2-\deg(R)<0$, but it is a quotient of $\iota^*\Omega_X^1$, contradicting the nefness of $\Omega_X^1$. \qed
Page 643, proof of Theorem 6.35. For $D\in|\mathscr L^{\otimes n}|$, the kernel of restriction has twist $-(n'+n)$, not $-(n'+1)$. Put $E=(s^*\Omega^1_{A/X})^\vee$ and write $E(-r)=E\otimes\mathscr L^{-r}$. Replace the short exact sequence by
\[ 0\longrightarrow E(-(n'+n)) \longrightarrow E(-n') \longrightarrow E(-n')|_D \longrightarrow0. \]The chosen Serre vanishing still applies, since $n'\geq n$ implies $n'+n\geq n$. Thus equation (6.1) and the remainder of the proof are unchanged.
Page 643, Remark 6.36. Corollary 6.6 proves the asserted improvement only in characteristic zero. Replace the remark by:
Remark 6.36. If $\operatorname{char}(k)=0$, then by Corollary 6.6 we may take $n=1$ above if
\[ \operatorname{rel.dim.}(f)<\dim(X)-1. \]
Report metadata
| Field | Value |
|---|---|
| Category | Published |
| Processing status | completed |
| Detailed comments | 32 |
| Domain | stem/mathematics |
| Completed | 2026-07-29T17:48:49.006591+00:00 |
| Refine document ID | 8cae163e-9db1-4bc1-be6f-cd96e6d7ba9f |
Refine summary
This paper investigates the conditions under which the restriction map of morphisms from a projective variety to a given space, and from an ample divisor to that space, is an isomorphism. By utilizing positive characteristic methods and deformation theory, the author provides an exhaustive framework that yields both classical and novel Lefschetz hyperplane theorems for various target spaces, including Deligne-Mumford stacks.
Overall feedback
Positive-characteristic deformation arguments
In Theorem 4.12, the proof handles the dualizing complex $K_D$ as though it were a shifted sheaf when converting the obstruction groups by duality. Since $K_D$ is only assumed to have cohomology in degrees $[-\dim D, -r]$, the hypercohomology spectral sequence must be invoked to derive the threshold $\phi<r-1$. The displayed equality does not intrinsically establish the required vanishing for an arbitrary singular divisor.
Independently, Lemma 4.28 processes the difference of two morphisms as a global section of the nilpotent ideal. However, infinitesimal differences are governed locally by derivations, such as sections of $g^*T_Y\otimes J$, and the reducedness of $\Gamma(D,\mathcal O_D)$ does not naturally eliminate these differences. Because Theorem 4.29 builds on Lemma 4.28 to descend from a Frobenius-composed map when $D$ is nonreduced, ensuring the claimed treatment of arbitrarily singular divisors will require an alternative descent argument or an explicit reducedness restriction.
Characteristic-zero spreading-out construction
For Theorem 4.22, the characteristic-zero spreading-out argument uses the existence of extensions on every closed fiber to conclude that the relative obstruction module vanishes. A structural question arises here: fiberwise extension demonstrates that an appropriate obstruction class specializes to zero, but this specializes-to-zero property neither forces the entire $H^1$-module to be zero nor guarantees that the fiberwise lifts extend a prior choice of lift on $D_{n-1}$.
Completing this formal extension entails an inductive relative obstruction construction, alongside cohomology-and-base-change control to show that specialization accurately detects the class, ensuring compatible choices of lifts through all $D_n$. Furthermore, the argument requires checking that the spread-out closed fibers satisfy the $W_2$-lifting and the assorted hypotheses of Theorem 4.29.
Formal-to-algebraic transition scopes
In transitioning from the formal to the algebraic setting, Corollary 2.7 proposes replacing a quasi-projective morphism with a projective flat bundle. The progression would be fully bridged by explaining the mechanism transferring the formal sheaf with proper support to that bundle, and how the resulting algebraization returns to the original $Y$.
Similarly, in Corollary 2.8, pulling a formal subscheme to a Chow cover, algebraizing it, and taking a scheme-theoretic image relies on the flatness of completion. To confirm this step, an explicit proof that the completed image securely recovers the original formal subscheme with its scheme structure is indispensable.
Corollaries 2.9, Lemma 5.2, and the broader passage from formal maps to maps on neighborhoods for schemes and stacks depend inherently on these properties. Providing a precise proper-support algebraization and descent theorem, encompassing the compatibility of scheme-theoretic images with completion, is essential for the later geometric results to hold at the required generality.
Consistency of theorem statements and hypotheses
There is an opportunity to strictly align the matrix of theorems stated in the introduction with the proved variants in the text. Theorem 1.10 advertises a positive-characteristic Deligne–Mumford-stack theorem without the bound $p>\dim X$. Conversely, Theorem 5.1 remains restricted to characteristic zero, and the scheme theorem used as its positive-characteristic model, Theorem 4.29, specifically requires $p>\dim X$.
Theorem 1.10 also specifies a quasi-projective coarse space in case (2), whereas Theorem 5.1 requires a projective coarse space. Additionally, Theorems 4.12, 4.22, and 4.29 omit properness or projectivity assumptions that are functionally required by duality arguments, Deligne–Illusie vanishing, Corollary 2.9, and the rational-map extension results invoked during their proofs.
In Section 5, noting that positive-characteristic stack arguments are analogous leaves certain structural gaps, particularly when Lemma 5.4 uses descent from normalized scheme-theoretic images without verifying the required groupoid $2$-cocycle. Aligning the introductory claims with a fully specified and verified set of characteristic-dependent theorems will solidify the paper's overarching architecture.
Duality and Kuga-Satake arguments in moduli applications
Several flagship moduli applications in Section 6 utilize cotangent-positivity calculations that confront dimensional and factorization hurdles. Lemma 6.9 applies the relation $(R^1f_*T_{Y/X})^\vee=R^1f_*(\Omega^1_{Y/X}\otimes\omega_{Y/X})$. However, relative Serre duality in relative dimension $n$ generates degree $n-1$ on the right side. Consequently, the ensuing positivity calculation acts upon a different Hodge bundle except in highly specialized dimensions.
This identical degree discrepancy affects Theorem 6.17, impacting the Calabi–Yau application. Furthermore, the hyperkähler case within Theorem 6.16 relies on establishing the asserted global Kuga–Satake family and verifying the tangent-map factorization to secure the necessary geometry.
Because Corollary 6.19 and prominent assertions in the abstract rest on these calculations, the geometric architecture could be strengthened by separating the applications. Reserving one category for targets with independently established nef cotangent bundles, and isolating targets that rely directly on modified duality computations, Hodge-bundle positivity, and Torelli/Kuga–Satake arguments, will present the applications robustly.
Detailed comments
1. Calabi–Yau application is stated too broadly
- ID:
55e7a22a-0e78-409c-8b94-c8d52512acd0 - Refine score:
0.31 - Original types: general
- Refine status: open
Comment
The Calabi–Yau application is valid if “Calabi–Yau” is used in the strict sense under which $h^{1,0}=h^{2,0}=0$ and a smooth polarized moduli space of dimension $h^{n-1,1}$ are available. Corollary 6.19(4) explicitly invokes those Hodge-number and smooth-moduli conditions, so under broader weak or merely trivial-canonical conventions the introductory claim would be too broad.
Quoted passage
- Any principally polarized Abelian scheme over $D$ of relative dimension $g$ extends uniquely to $X$, as long as $\operatorname{dim}(X) \geq \operatorname{dim}\left(\mathscr{A}_{g}\right)+2$ (see Corollary 6.19(3)).
- Any polarized family of Calabi-Yau varieties over $D$ extends uniquely to $X$, as long as $\operatorname{dim}(X) \geq h^{n-1,1}+2$ (see Corollary 6.19(4)).
- Let $f: Y \rightarrow X$ be a smooth relative curve of genus $g \geq 2$, and let $f_{D}: Y_{D} \rightarrow D$ be its base change to $D$.
2. Hodge-theoretic cases exceed the cited theorem
- ID:
4ac4e8c9-a3cf-4846-b02e-bd0859c343a7 - Refine score:
0.44 - Original types: general
- Refine status: open
Comment
The K3-type alternative appears broader than Theorem 6.22 as stated. That theorem covers compact arithmetic quotients of Hermitian symmetric domains and Shimura varieties of PEL type; weight-one period spaces fit the latter class, but a general noncompact K3-type orthogonal quotient is not expressly covered. The K3-type branch therefore requires an additional argument or a narrower scope.
Quoted passage
- If $\mathscr{H}$ is a polarized variation of Hodge structure over $D$, induced by a period map $D \rightarrow Y$ with $Y$ quasi-projective, then $\mathscr{H}$ extends uniquely to $X$ if $\operatorname{dim}(D)>\operatorname{dim}(Y)$, and if
- $Y$ is compact, or
- $\mathscr{H}$ is of weight one, or
- $\mathscr{H}$ is of K3 type
(see Theorem 6.22). We also give versions of many of these theorems in positive characteristic. Before diving into these applications, however, we will discuss the new perspective on classical Lefschetz theorems which motivates this work.
3. Positive-characteristic hypothesis missing in Theorem 1.10
- ID:
c0c54465-4cf5-41a3-b896-35cbde5a6f23 - Refine score:
0.44 - Original types: general
- Refine status: open
Comment
The positive-characteristic alternative in Theorem 1.10 omits the hypothesis $p>\dim(X)$ imposed by Theorems 1.20, 4.29, and 5.7 and used by the supplied vanishing and Frobenius-descent arguments. Consequently, the stated positive-characteristic scope is broader than the result established later in the paper.
Quoted passage
Theorem 1.10. Let $k$ be a field and $X$ a smooth projective variety over $k$, with $\operatorname{dim}(X) \geq 3$. Let $D \subset X$ be an ample Cartier divisor, and let $Y$ be a smooth Deligne-Mumford stack over $k$. Let $f: D \rightarrow Y$ be a morphism. Suppose that either $\operatorname{char}(k)=0$ or that $k$ is perfect of characteristic $p>0$, and $X$ lifts to $W_{2}(k)$. If
4. Projectivity hypothesis is absent from Theorem 1.20
- ID:
5a94d466-4fd1-4362-af30-120f05029d2e - Refine score:
0.37 - Original types: general
- Refine status: open
Comment
Theorem 1.20 assumes only that $X$ is smooth, but its proof as Theorem 4.29 invokes projective algebraization, f-amplitude, and vanishing results requiring $X$ to be projective. An ample Cartier divisor on a nonproper variety does not by itself imply projectivity, so the stated hypotheses do not establish the announced scope.
Quoted passage
Theorem 1.20. Let $L$ be a perfect field of characteristic $p>0$. Let $X$ be a smooth $L$-variety and $D \subset X$ an ample Cartier divisor, with $\operatorname{dim}(X) \geq 3$, such that $X$ lifts to $W_{2}(L)$, and such that $\operatorname{dim}(X)<p$. Let $Y$ be a smooth $k$-variety, and let $f: D \rightarrow Y$ be a morphism.
5. Theorem 1.22 does not match Theorem 6.32
- ID:
0e43c860-d73a-462b-bb8d-937c45c21502 - Refine score:
0.32 - Original types: general
- Refine status: open
Comment
The cross-reference to Theorem 6.32 does not directly support Theorem 1.22 as stated: Theorem 6.32 requires a quasi-projective coarse moduli space and bounds $\dim(\operatorname{im} f)$, whereas Theorem 1.22 instead assumes a finite étale scheme cover and bounds $\dim Y$. Theorem 1.22 may follow from Theorem 5.1(1) together with the f-amplitude estimate, but it is not literally a specialization of Theorem 6.32 without an additional implication between these hypotheses.
Quoted passage
In Theorem 6.32, we improve this result to Theorem 1.22. Let $k$ be a field of characteristic zero. Let $X$ be a smooth projective $k$-variety with $\operatorname{dim}(X) \geq 3$, and let $D \subset X$ be an ample Cartier divisor. Let $Y$ be a smooth Deligne-Mumford stack over $k$ and let $f: D \rightarrow Y$ be a morphism. Suppose that there exists a scheme $Y^{\prime}$ and a finite surjective étale morphism $Y^{\prime} \rightarrow Y$, and that $\operatorname{dim}(Y)<\operatorname{dim}(D)-1$. Then $f$ extends uniquely to a morphism $X \rightarrow Y$.
6. Remark 1.24 conflates finite and generically finite maps
- ID:
af9c7d98-6b01-4a8d-ac5e-bbc63f4e3c35 - Refine score:
0.31 - Original types: general
- Refine status: open
Comment
Remark 1.24 appears to identify two inequivalent conditions. Torsion of the evaluation cokernel means that global one-forms generically span $\Omega_Z^1$, which under the usual hypotheses implies that the Albanese map is generically finite onto its image; it does not imply the existence of a finite or quasi-finite map to an Abelian variety. For example, the blowup of an Abelian variety of dimension at least two at a point satisfies the displayed torsion condition, but every map from it to an Abelian variety contracts the exceptional divisor. If “finite-to-one” is intended in the weaker, generically finite sense, the subsequent reference to “finite maps” remains misleading.
Quoted passage
Remark 1.24. The condition that $Z$ admits a finite-to-one map to an Abelian variety is close to nefness; it is equivalent to the condition that
$$ \operatorname{coker}\left(\Gamma\left(Z, \Omega_{Z}^{1}\right) \otimes \mathscr{O}_{Z} \rightarrow \Omega_{Z}^{1}\right) $$be torsion. The requirement that $Z$ admit an unramified map to an Abelian variety is equivalent to global generation of $\Omega_{Z}^{1}$, which of course implies nefness.
Again, Sommese and Beltrametti work over the complex numbers. Unfortunately, this theorem does not suffice for our applications once again; even if one were to extend it stacks, the targets $Z$ in many of our applications do not admit finite maps to Abelian varieties (e.g. $\mathscr{M}_{g}$ does not).
7. Perfectness of the derived pushforward needs support
- ID:
33619dea-12f7-4ac3-98eb-f242869326b3 - Refine score:
0.6 - Original types: general
- Refine status: open
Comment
The proof of Lemma 2.1 does not justify that $\mathbf{R}f_*(\mathscr{F}\otimes(\mathscr{L}^{\vee})^{\otimes n})$ is perfect under the stated hypotheses. Proper pushforward preserves perfect complexes when additional conditions such as finite Tor-dimension are available, but projectivity and bounded coherent cohomology of $f^!\mathscr O_S$ alone do not ensure this. Because the argument transfers the amplitude bound from the derived dual back to the original complex using perfection, this leaves a genuine gap unless an additional hypothesis or a different biduality argument applies.
Quoted passage
So for $n \gg 0, \mathbf{R} f_{*}\left(\mathscr{F}^{\vee} \otimes^{\mathbf{L}} \omega_{X / S} \otimes \mathscr{L}^{\otimes n}\right)$ and thus $\mathbf{R} \underline{\operatorname{Hom}}\left(\mathbf{R} f_{*}\left(\mathscr{F} \otimes\left(\mathscr{L}^{\vee}\right)^{\otimes n}\right), \mathscr{O}_{S}\right)$ have cohomology concentrated in degrees $[-n,-m+a]$. As $\mathbf{R} f_{*}\left(\mathscr{F} \otimes\left(\mathscr{L}^{\vee}\right)^{\otimes n}\right)$ is perfect, by e.g. [41, Tag 0A1E, Lemma, 35.18.1],
$$ \mathbf{R} f_{*}\left(\mathscr{F} \otimes\left(\mathscr{L}^{\vee}\right)^{\otimes n}\right) $$has cohomology concentrated in degrees $[m-a, n]$, as desired. $\square$
8. Corollary 2.6 conflates relative and global Ext
- ID:
7ef623f9-09c4-478c-a1d4-aa985581c444 - Refine score:
0.34 - Original types: general
- Refine status: open
Comment
As stated over an arbitrary base $S$, Corollary 2.5 directly gives an isomorphism of the relative cohomology sheaves $R^i(f\circ g)_*R\mathcal{H}om(\mathscr F,\mathscr G)$, not necessarily of the unadorned global Ext groups displayed in Corollary 2.6. The latter conclusion requires an additional passage to global hypercohomology, particularly because the allowed complexes may have negative cohomological degrees. The later application over $S=\operatorname{Spec}k$ is unaffected.
Quoted passage
Corollary 2.6 (Analogue of [16, $\mathrm{III}_{1}$.4.5.1]). Let $g: Y \rightarrow X$ be a proper morphism and let $f: X \rightarrow S$ be projective, with $D \subset X$ an $f$-ample Cartier divisor. Suppose that $\omega_{X / S}$ has coherent cohomology concentrated in degrees $[-n,-m]$. Let $\mathscr{F}, \mathscr{G}$ be complexes on $Y$ so that
$$ \mathbf{R} g_{*} \mathbf{R} \operatorname{Hom}(\mathscr{F}, \mathscr{G}) $$is perfect of tor-amplitude $[-a, 0]$. Let $\widehat{g}: \widehat{Y_{D}} \rightarrow \widehat{D}$ be the completion of $g$ at $D$, and let $\widehat{f}: \widehat{D} \rightarrow S$ be the structure morphism. Then the map
$$ \operatorname{Ext}^{i}(\mathscr{F}, \mathscr{G}) \rightarrow \operatorname{Ext}^{i}(\widehat{\mathscr{F}}, \widehat{\mathscr{G}}) $$is an isomorphism for $0 \leq i \leq m-a-2$. Proof. This is immediate by applying Corollary 2.5 to the complex $\mathbf{R} \operatorname{Hom}(\mathscr{F}, \mathscr{G})$. $\square$
9. Sign reversal in the Serre-vanishing choice
- ID:
0bde9f3c-5c14-4fdb-961b-78eedaa829d7 - Refine score:
0.29 - Original types: general
- Refine status: open
Comment
The inequality in the proof of Corollary 2.7 has the wrong sign. The preceding construction permits $m_2-m_1\gg0$, which is also the condition making $R\mathcal{H}om(\mathscr O_Y(-m_2),\mathscr O_Y(-m_1))\simeq\mathscr O_Y(m_2-m_1)$ sufficiently positive for relative Serre vanishing; the printed condition $m_1-m_2\gg0$ does not support the stated argument.
Quoted passage
Applying this argument to the kernel of the map above, we find a presentation
$$ \widehat{\mathscr{O}_{Y}\left(-m_{2}\right)} \otimes \widehat{f}^{*} \mathscr{O}_{X} \widehat{\left(-a_{2} D\right)^{n_{2}}} \rightarrow \widehat{\mathscr{O}_{Y}\left(-m_{1}\right)} \otimes \widehat{f}^{*} \mathscr{O}_{X} \widehat{\left(-a_{1} D\right)^{n_{1}}} \rightarrow \mathscr{F} \rightarrow 0 ; $$we may take $m_{2}-m_{1}$ arbitrarily large. Corollary 2.7. Let $k$ be a field, let $g: Y \rightarrow X$ be a quasi-projective morphism of finite-type, and let $X$ be a projective normal $k$-variety, with $D \subset$ $X$ an ample Cartier divisor. Suppose $\operatorname{dim}(X) \geq 2$. Let $\widehat{g}: \widehat{Y_{D}} \rightarrow \widehat{D}$ be the completion of $g$ at $D$. Then if $\mathscr{F}$ is a coherent sheaf on $\widehat{Y_{D}}$ with support proper over $\widehat{D}$, there exists a coherent sheaf $\mathscr{G}$ on $Y$ so that $\mathscr{F} \simeq \widehat{\mathscr{G}}$.
Proof. Without loss of generality, $g$ is projective and flat (by replacing $Y$ with a suitable projective bundle over $X$ in which it embeds). Choose a resolution
$$ \widehat{\mathscr{O}_{Y}\left(-m_{2}\right)} \otimes \widehat{g}^{*} \mathscr{O}_{X} \widehat{\left(-a_{2} D\right)^{n_{2}} \xrightarrow{p} \mathscr{O}_{Y}\left(-m_{1}\right)} \otimes \widehat{g}^{*} \mathscr{O}_{X} \widehat{\left(-a_{1} D\right)^{n_{1}}} \rightarrow \mathscr{F} \rightarrow 0 $$as above, with $m_{1}-m_{2} \gg 0$, so that
10. Corollary 3.4 invokes an unavailable hypothesis
- ID:
8157eacb-cc25-49aa-a261-ba2e9db49885 - Refine score:
0.36 - Original types: general
- Refine status: open
Comment
Corollary 3.4 invokes Corollary 3.2 without assuming that $X$ is locally $\mathbb{Q}$-factorial. The modification $X' \to X$ neither supplies that hypothesis for the intended application nor provides a descent argument. Thus the proof does not establish the corollary for an arbitrary normal $X$, although the later applications with smooth $X$ are unaffected.
Quoted passage
Corollary 3.4. Let $X$ be a normal projective $k$-variety, and let $Y$ be a quasi-projective $k$-variety. Let $D \subset X$ be an ample divisor and let $U \subset X$ be a Zariski-open subset containing $D$. Then any map $f: U \rightarrow Y$ with $\operatorname{dim}(f(D)) \leq \operatorname{dim}(D)-2$ extends uniquely to a map $X \rightarrow Y$.
Proof. Let $Y^{\prime}$ be a projective compactification of the scheme-theoretic image of $f$, and resolve the rational map $U \rightarrow Y^{\prime}$ to a regular map $f^{\prime}: X^{\prime} \rightarrow Y^{\prime}$, with $r: X^{\prime} \rightarrow X$ proper. By Corollary 3.2, it suffices to show that $\operatorname{dim}\left(Y^{\prime}\right) \leq$ $\operatorname{dim}(X)-2$; hence it suffices to show that $f(D)$ has codimension 1 in $Y^{\prime}$.
11. Proposition 3.5 does not complete the descent argument
- ID:
a69fabac-72a3-42e0-b501-3efaeeeab7db - Refine score:
0.42 - Original types: general
- Refine status: open
Comment
The final argument in Proposition 3.5 omits the step that rules out exceptional components. Because $X'$ is the normalization of the closure of the section, its map to $Y$ is finite; an exceptional rational curve would lie in a geometric fiber of $f$ and be contracted there, contradicting that finiteness. Only after this argument can purity give quasi-finiteness of $b$. The subsequent finiteness should concern the proper map $b:X'\to X$, not a map from $X'$ to itself.
Quoted passage
Let $X^{\prime} \rightarrow Y$ be the map given by $b_{Y} \circ \widetilde{s}$. There is a rational curve passing through the general point of an exceptional component of $b$ by e.g. [13, Proposition 1.43]. Thus $b_{Y} \circ \widetilde{s}$ map contracts the fibers of $b$, as $Y$ contains no rational curves, and hence $X^{\prime}$ is quasi-finite over $X$. But $f$ is proper, so $X^{\prime}$ is in fact finite over $X^{\prime}$; it is an isomorphism over the locus where $s$ is defined. Hence $X^{\prime}$ is isomorhpic to $X$ by Zariski's main theorem, providing us with a section as desired. $\square$
12. Corollary 4.2 omits the thickening hypotheses
- ID:
7cc93426-2edc-4273-a105-7587984c8dbb - Refine score:
0.3 - Original types: general
- Refine status: open
Comment
Corollary 4.2 is formally under-specified: the normal bundle $\mathscr{N}_{D/X}$ and the extension problem to $D_2$ require $D\subset X$ to be a closed lci subscheme and $D_2\subset X$ to be the subscheme defined by $\mathscr{I}_D^2$, as in Theorem 4.1.
Quoted passage
Corollary 4.2. Let $X, D, D_{2}$ be schemes over a field $k$, and let $Y$ be an arbitrary smooth $k$-scheme. Then if $s: D \rightarrow Y$ is a morphism, there is a natural obstruction class $o(s) \in \operatorname{Ext}^{1}\left(\mathscr{N}_{D / X}, s^{*} T_{Y / k}\right)$ whose vanishing is equivalent to the existence of an extension of $s$ to $D_{2}$; such extensions are a torsor for $\operatorname{Hom}\left(\mathscr{N}_{D / X}, s^{*} T_{Y / k}\right)$.
13. Le Potier’s second vanishing range is false
- ID:
7c9c2f13-b464-4b67-bc71-eed1c9da60e3 - Refine score:
0.33 - Original types: general
- Refine status: open
Comment
The second vanishing range in Theorem 4.4 is false as written. For $X=\mathbb{P}^1$, $E=\mathscr{O}(1)$, and $i=p=0$, the condition $i+p\geq n-e$ holds, but $H^0(\mathbb{P}^1,\mathscr{O}(1))\neq0$. The intended Le Potier threshold appears to be $i+p\geq n+e$.
Quoted passage
Theorem 4.4 (Le Potier). Let $E$ be an ample vector on a smooth projective variety $X$ over a field of characteristic zero, with $\operatorname{dim}(X)=n$ and $\operatorname{rk}(E)=e$. Then
$$ H^{i}\left(X, \omega_{X} \otimes \bigwedge^{a} E\right)=0 $$for $a>0, i>e-a$, and
$$ H^{i}\left(X, \Omega_{X}^{p} \otimes E\right)=0 $$for $i+p \geq n-e$.
14. Theorem 4.6 has inconsistent scope
- ID:
984d2829-9304-4381-b95b-16c950dd0398 - Refine score:
0.27 - Original types: general
- Refine status: open
Comment
The formal scope of Theorem 4.6 is inconsistent. Its opening assumption $\dim(D)\geq2$ prevents the first bullet from asserting the advertised uniqueness result for $\dim(D)=1$. In addition, the theorem statement should retain the characteristic-zero or complex setting announced immediately beforehand and required by its Nakano-vanishing argument.
Quoted passage
Indeed, there are certain results we can only obtain in characteristic zero for smooth $D$, which we state here.
Theorem 4.6. Let $X$ be a projective variety, and let $D \subset X$ be a smooth lci subscheme, with ample normal bundle, and with $\operatorname{dim}(D) \geq 2$. Let $\widehat{D}$ be the formal scheme obtained by completing $X$ at $D$. Let $Y$ be a smooth variety with Nakano semipositive cotangent bundle. Then given a morphism $f: D \rightarrow Y$,
- there is at most one extension of $f$ to a morphism $\widehat{D} \rightarrow Y$ if $\operatorname{dim}(D) \geq$ 1, and
- such an extension exists as long as $\operatorname{dim}(D) \geq 2$.
15. Base field mismatch in Theorems 1.19, 1.20, and 4.12
- ID:
49182a33-bc35-45ec-b3a4-6cfc9dfd4d87 - Refine score:
0.29 - Original types: general
- Refine status: open
Comment
Theorems 1.19, 1.20, and 4.12 have an inconsistent base field: $X$ and $D$ are defined over $L$, while $Y$ is called a smooth $k$-variety without any relationship between $k$ and $L$. Since the theorems subsequently use $F_{Y/L}$ and $Y^{(p^k)}$, they require a specified $L$-structure on $Y$ and an $L$-morphism $f:D\to Y$.
Quoted passage
Theorem 4.12. Let $X$ be variety over a field $L$ of characteristic $p$, and let $D \subset X$ be a Cartier divisor whose dualizing complex $K_{D}$ is supported in degrees $[-\operatorname{dim}(D),-r]$, and whose normal bundle is ample. Suppose $\operatorname{dim}(X) \geq$ 3. Let $\widehat{D}$ be the formal scheme obtained by completing $X$ at $D$. Let $f: D \rightarrow Y$ be a morphism, with $Y$ a smooth $k$-variety.
16. Duality degrees conflict in the proof of Theorem 4.12
- ID:
546b1d86-6aa5-4097-9343-c6a51a4d9a78 - Refine score:
0.54 - Original types: general
- Refine status: open
Comment
The Grothendieck-duality display uses a convention inconsistent with the stated normalization of $K_D$. If $K_D$ is the dualizing complex supported in degrees $[-d,-r]$, duality gives $H^i(D,A)^\vee \cong \mathbb{H}^{-i}(D,K_D\otimes A^\vee)$, not the displayed un-dualized group in degree $d-i$. The required vanishing still follows from the hypercohomology spectral sequence and the bound $\phi(f^*\Omega_Y^1\otimes\mathscr N_{D/X})<r-1$, but that argument is not expressed by the displayed identity.
Quoted passage
Recall also that $\mathscr{O}_{D}(-D)=\mathscr{N}_{D / X}^{\vee}$. But by Grothendieck duality,
$$ \begin{array}{r} H^{i}\left(D, \operatorname{Frob}_{p}^{k *}\left(f^{*} T_{Y} \otimes \mathscr{O}_{D}(-D)\right) \otimes \mathscr{O}_{D}(-s D)\right) \\ =H^{\operatorname{dim}(D)-i}\left(D, K_{D}(s D) \otimes \operatorname{Frob}_{p}^{k *}\left(f^{*} \Omega_{Y}^{1} \otimes \mathscr{N}_{D / X}\right)\right), \end{array} $$which is zero by our assumption on $k$. $\square$
17. Lemma 4.26 conflates absolute and relative Frobenius
- ID:
2643befe-721d-48a7-a887-7b1ea4831919 - Refine score:
0.39 - Original types: general
- Refine status: open
Comment
Lemma 4.26 is correct only when the factorization is interpreted in the category of schemes, or equivalently with the appropriate Frobenius twist. Over a general perfect field, absolute Frobenius is not a $k$-morphism, and the affine factor map must send $c\in k$ to $c^{1/p}$ rather than fix $k$. Thus the assignment on the variables alone does not define the implied $k$-algebra map or justify that it kills $I$. This distinction matters when Theorem 4.29 removes successive relative Frobenius factors.
Quoted passage
Lemma 4.26. Let $k$ be a perfect field of characteristic $p>0$, and let $X$ be a normal $k$-variety. Let $Y$ be an arbitrary $k$-variety. Then a morphism $f: X \rightarrow Y$ factors through $\operatorname{Frob}_{p}: Y \rightarrow Y$ if and only if the induced map $f^{*} \Omega_{Y}^{1} \rightarrow \Omega_{X}^{1}$ is zero; furthermore, this factorization is unique.
Proof. Without loss of generality, $X$ is connected and hence, by normality, integral.
18. Lemma 4.28 fails for nonreduced ample divisors
- ID:
fe79d33b-ea3b-41a4-a676-c09687981085 - Refine score:
0.88 - Original types: general
- Refine status: open
Comment
Lemma 4.28 is false for nonreduced ample Cartier divisors. For example, take $k=\mathbb{F}_p$ with $p>3$, $X=\mathbb{P}^3_k$, $D=V(z^2)$, and $Y=\mathbb{P}^2_k$. The distinct maps $g_a([x_0:x_1:x_2:z])=[x_0+a_0z:x_1+a_1z:x_2+a_2z]$ all have the same composite with Frobenius because $z^p=0$. Moreover, $g_1-g_2$ is not intrinsically defined for maps to an arbitrary scheme, and reducedness of $\Gamma(D,\mathscr{O}_D)$ does not eliminate nilpotent directions in $D$. Since Theorem 4.29 uses this lemma precisely without assuming $D$ reduced, that step needs a different argument or an additional hypothesis.
Quoted passage
Lemma 4.28. Let $k$ be a perfect field of characteristic $p>0$. Let $X$ be a smooth projective $k$-variety with $2<\operatorname{dim}(X)<p$, and let $D \subset X$ be an ample divisor. Suppose that $X$ lifts to $W_{2}(k)$. Let $Y$ be a scheme and let $f: D \rightarrow Y$ be a morphism. Then there is at most one morphism $g: D \rightarrow Y$ so that $f=g \circ \operatorname{Frob}_{p}$.
Proof. Let $\sqrt[p]{0} \subset \mathscr{O}_{D}$ be the ideal sheaf
19. Properness is missing in case (3) of Theorem 4.29
- ID:
15d9f012-0196-48c3-aa58-9cf08074a96e - Refine score:
0.44 - Original types: general
- Refine status: open
Comment
Case (3) of Theorem 4.29 lacks the properness hypothesis required to apply Proposition 3.5. A nonproper target can contain no rational curves while still admitting maps from $U$ that fail to extend across $X\setminus U$, so the stated hypotheses do not justify the extension to $X$.
Quoted passage
Suppose further that
(1) $\operatorname{dim}(Y)<\operatorname{dim}(D)$, or (2) $Y$ is projective and $\operatorname{dim}(\operatorname{im}(f))<\operatorname{dim}(D)-1$, or (3) $Y_{\bar{L}}$ contains no rational curves.
Then $f$ extends uniquely to a morphism $X \rightarrow Y$. Proof. We first prove existence of an extension. By Theorem 4.12, there exists $k \geq 0$ so that $F_{Y / L}^{k} \circ f$ extends to a morphism $\widehat{D} \rightarrow Y^{\left(p^{k}\right)}$. By Corollary 2.9, there exists a Zariski open set $U \subset X$, with $D \subset U$, so that this map extends to a morphism $U \rightarrow Y^{\left(p^{k}\right)}$. Finally, by Corollary 3.2 in case (1), Corollary 3.4 in case (2), or Proposition 3.5 in case (3) this map extends to a map $\widetilde{f}: X \rightarrow Y^{\left(p^{k}\right)}$.
20. The spreading-out obstruction argument is incomplete
- ID:
ee82fb36-552d-4a0f-8277-77bff7965266 - Refine score:
0.53 - Original types: general
- Refine status: open
Comment
The obstruction group for extending from $D_{n-1}$ to $D_n$ should involve $\mathscr I_D^{n-1}/\mathscr I_D^n$, not $\mathscr I_D^{n-1}/\mathscr I_D$. In addition, the argument must justify that the relative obstruction class specializes to the fiberwise obstruction classes—after suitable shrinking and cohomology-and-base-change—before vanishing on all closed fibers implies that the global class vanishes.
Quoted passage
We now show that there is an extension of $f$ to $\widehat{D}$ as desired. We may spread $(X, D, Y, f)$ out over an integral finite-type $\mathbb{Z}$-scheme $S$. After shrinking $S$, we may assume that for each closed point $s \in S$, the hypotheses of Theorem 4.29 are satisfied. Thus each $f_{s}$ extends uniquely to a map $X_{s} \rightarrow Y_{s}$. In particular, letting $D_{n}$ be the $n$-th infinitesimal neighborhood of $D$, there exists an extension of $f$ to $\left(D_{n}\right)_{s}$ for each $s$. But the obstruction to extending $f$ from $D_{n-1}$ to $D_{n}$ is an element of the coherent $\mathscr{O}_{S}$-module
$$ H^{1}\left(D, f^{*} T_{Y} \otimes \mathscr{I}_{D}^{n-1} / \mathscr{I}_{D}\right), $$which is zero at every closed point of $S$; hence it is zero. So the desired extension exists. $\square$
21. Theorem 5.1 appears to need projective $X$
- ID:
fa8fc2d8-b992-47fb-a649-51d9a88211f3 - Refine score:
0.39 - Original types: general
- Refine status: open
Comment
Theorem 5.1 omits projectivity of $X$, although Lemma 5.2 invokes Corollary 2.8 and Lemma 5.4 invokes Corollaries 3.2 and 3.4, all of which assume a projective base. Smoothness together with the existence of an ample Cartier divisor does not imply projectivity, so the proof does not establish the theorem at its stated generality.
Quoted passage
Theorem 5.1. Let $k$ be a field of characteristic zero. Let $X$ be a smooth $k$-variety and let $D \subset X$ be an ample Cartier divisor, with $\operatorname{dim}(X) \geq 3$. Let $\mathscr{Y}$ be a smooth Deligne-Mumford stack over $k$. Let $f: D \rightarrow \mathscr{Y}$ be a morphism such that
$$ \phi\left(\mathscr{N}_{D / X} \otimes f^{*} \Omega_{\mathscr{Y}}^{1}\right)<\operatorname{dim}(D)-1 . $$
22. Case (2) conflicts with later stated Theorems 1.10 and 6.32
- ID:
3be8178f-17af-444a-9485-e224515486c9 - Refine score:
0.38 - Original types: general
- Refine status: open
Comment
Case (2) requires the coarse space to be projective, whereas Theorems 1.10 and 6.32 claim the result for a quasi-projective coarse space (the small-image assumption gives $\dim(f(D))\leq\dim(D)-2$, but a quasi-projective coarse space need not be projective). The proof of Lemma 5.4 invokes Corollary 3.4, whose target hypothesis is quasi-projectivity, which appears to support the intended quasi-projective version. As stated, Theorem 5.1 therefore does not support those broader claims. The stated dependency is internally inconsistent and must be reconciled.
Quoted passage
Suppose further that
(1) $\operatorname{dim}(\mathscr{Y})<\operatorname{dim}(D)$, or (2) $\operatorname{dim}(f(D)) \leq \operatorname{dim}(D)-2$ and the coarse space of $\mathscr{Y}$ is projective, or (3) $\mathscr{Y}$ is proper and admits a model which is finite-type over $\mathbb{Z}$ and whose geometric fibers contain no rational curves (i.e., any map from a rational curve to this model is constant) and moreover admits a finite étale cover by a scheme.
23. Lemma 5.4 is too broad in case (2)
- ID:
6a9e9bda-c4e0-474c-8586-0ef3e6871093 - Refine score:
0.54 - Original types: general
- Refine status: open
Comment
Lemma 5.4 overstates the case-(2) conclusion: Corollary 3.4 applies only to a particular map $U\to\mathscr{Y}$ whose restriction to $D$ has image dimension at most $\dim(D)-2$, not to every object of $\mathscr{Y}(U)$. The displayed unrestricted equivalence therefore does not follow; moreover, the proof constructs extensions but does not establish the full faithfulness required for an equivalence of groupoids.
Quoted passage
Lemma 5.4. Let $X, D, \mathscr{Y}$ be as in the theorem, and let $U \subset X$ be a Zariski-open containing $D$. Then the restriction map $\mathscr{Y}(X) \rightarrow \mathscr{Y}(U)$ is an equivalence.
Proof. Let $Y$ be the coarse space of $\mathscr{Y}$, which exists by the Keel-Mori theorem [24]. Let $f: U \rightarrow \mathscr{Y}$ be a map; we may resolve the induced map $U \rightarrow Y$ to obtain a scheme $X^{\prime}$, proper over $X$ and a map $f^{\prime}: X^{\prime} \rightarrow Y$.
24. Theorem 6.5 fails without properness
- ID:
d268af38-2ce5-40cc-a0d1-366b7017fca0 - Refine score:
0.4 - Original types: general
- Refine status: open
Comment
Theorem 6.5 is false for nonproper smooth morphisms. For example, if $f:\operatorname{Tot}(\mathscr{O}_{\mathbb{P}^1}(1))\to\mathbb{P}^1$, then $\Omega^1_{Y/X}\simeq f^*\mathscr{O}_{\mathbb{P}^1}(-1)$ and the displayed evaluation map is surjective, but restriction to the zero section shows that $\Omega^1_{Y/X}$ is not nef. Here $f_*\Omega^1_{Y/X}$ is also an infinite direct sum rather than a vector bundle, so the Griffiths-positivity argument does not apply. The proper setting of Corollary 6.6 may remain valid, but Theorem 6.5 needs an appropriate properness or Hodge-theoretic hypothesis.
Quoted passage
Theorem 6.5. Let $k$ be a field of characteristic zero. Let $f: Y \rightarrow X$ be a smooth morphism of smooth $k$-varieties with $\Omega_{X / Y}^{1}$ relatively globally generated (i.e., the map
$$ f^{*} f_{*} \Omega_{Y / X}^{1} \rightarrow \Omega_{Y / X}^{1} $$is surjective). Then $\Omega_{Y / X}^{1}$ is nef. Proof. As the quotient of a nef vector bundle is nef, it suffices to show that $f^{*} f_{*} \Omega_{Y / X}^{1}$ is nef. As nefness is preserved by pullbacks, it is enough to show that $f_{*} \Omega_{Y / X}^{1}$ is nef. But this is a consequence of Griffiths positivity; see e.g. [29, Theorem 5] or [18, Corollary 7.8] for the dual result. $\square$
25. Relative-duality degree in Lemma 6.9
- ID:
c4c9de67-751e-46ef-9ba2-5a0c48c91478 - Refine score:
0.52 - Original types: general
- Refine status: open
Comment
The first relative-duality display in Lemma 6.9 has the wrong cohomological degree. For relative dimension $n$, relative Serre duality gives $(R^1f_*T_{Y/X})^\vee \simeq R^{n-1}f_*(\Omega^1_{Y/X}\otimes\omega_{Y/X})$, not an $R^1f_*$ term on the right. The displayed equality is therefore valid only when $n=2$, while the lemma and its applications include arbitrary relative dimension.
Quoted passage
Proof. We have
$$ \left(\mathbf{R}^{1} f_{*} T_{Y / X}^{1}\right)^{\vee}=\mathbf{R}^{1} f_{*}\left(\Omega_{Y / X}^{1} \otimes \omega_{Y / X}\right) . $$As each geometric fiber of $f$ has globally generated cotangent bundle, $\Omega_{Y / X}^{1}$ is isomorphic to $f^{*} f_{*} \Omega_{X / Y}^{1}$, arguing as in Corollary 6.6.
26. Duals are reversed in Lemma 6.12
- ID:
87eae2a5-4512-4ddb-9ba5-4ff0a0e72793 - Refine score:
0.45 - Original types: general
- Refine status: open
Comment
The duals are reversed in Lemma 6.12. If $W=(f_*\Omega^1_{Y/X})^\vee\otimes R^1f_*\mathscr O_Y\simeq R^1a_*T_{A/X}$, Lemma 6.9 establishes that $W^\vee$ is nef, not $W$. Likewise, dualizing the assumed injection $R^1f_*T_{Y/X}\hookrightarrow W$ yields a surjection $W^\vee\twoheadrightarrow(R^1f_*T_{Y/X})^\vee$. Thus the moduli cotangent bundle is obtained as a quotient of the nef bundle $W^\vee$, whereas the proof incorrectly uses $W$.
Quoted passage
Then $\left(f_{*} \Omega_{Y / X}^{1}\right)^{\vee} \otimes\left(\mathbf{R}^{1} f_{*} \mathscr{O}_{Y / X}\right)$ is isomorphic to $\mathbf{R}^{1} a_{*} T_{A / X}$ and hence is nef by Lemma 6.9 (one may also see this directly using Griffiths positivity). But $f^{*} \Omega_{\mathscr{M}}^{1}$ is a quotient of $\left(\mathbf{R}^{1} f_{*} T_{Y / X}\right)^{\vee}$ and hence a quotient of $\left(f_{*} \Omega_{Y / X}^{1}\right)^{\vee} \otimes$ $\left(\mathbf{R}^{1} f_{*} \mathscr{O}_{Y / X}\right)$ by assumption and hence is nef as well.
27. Undefined bundle maps in Theorem 6.16
- ID:
ad52faff-e6a4-4514-ae7d-d0a629054520 - Refine score:
0.63 - Original types: general
- Refine status: open
Comment
The displayed chain in Theorem 6.16 is not well typed. The classifying map is $g:X\to\mathscr M$, so the first term should involve pullback along $g$; the Kuga–Satake family has a structure morphism $a:A\to X$, so $T_{A/Y}$ and $R^1f_*T_{A/Y}$ are not defined from the stated data. As written, the chain does not establish the claimed injection from the hyperkähler deformation bundle into a Kuga–Satake bundle with nef dual.
Quoted passage
We may consider the natural map
$$ \begin{aligned} f^{*} T_{\mathscr{M}}^{1} & \rightarrow \mathbf{R}^{1} f_{*} T_{Y / X} \rightarrow \mathbf{R}^{1} f_{*} T_{A / Y} \\ & \rightarrow\left(f_{*} \Omega_{A / X}^{1}\right)^{\vee} \otimes\left(\mathbf{R}^{1} f_{*} \mathscr{O}_{A / X}\right) \rightarrow\left(f_{*} \Omega_{Y / X}^{2}\right)^{\vee} \otimes\left(\mathbf{R}^{1} f_{*} \Omega_{Y / X}^{1}\right) . \end{aligned} $$This map is injective by the local Torelli theorem for hyperkählers [21], Section 5].
28. Theorem 6.21 does not cover the surface case as proved
- ID:
a2b24267-f752-4755-9adb-d901c1bf72ac - Refine score:
0.35 - Original types: general
- Refine status: open
Comment
The displayed proof does not derive the surface-case surjectivity from Theorem 5.1 or Theorem 4.12, since both assume $\operatorname{dim}(X)\geq 3$. Although Theorem 6.21 is separately attributed to [19], recovering its first bullet by the paper’s methods requires a distinct dimension-two full-faithfulness argument.
Quoted passage
Theorem 6.21 (Found in [19, Théorème 3.10]). Let $k$ be a field and let $X$ be a smooth projective $k$-variety. Let $D \subset X$ be an ample divisor. Then the natural map (obtained after choosing a base-point) $\pi_{1}(D) \rightarrow \pi_{1}(X)$ is
- surjective if $\operatorname{dim}(X) \geq 2$, and
- an isomorphism if $\operatorname{dim}(X) \geq 3$.
Proof. This is immediate from Theorem 5.1 if $k$ is of characteristic zero by applying the theorem in the case that $Y=B G$, for $G$ a finite étale group scheme. In characteristic $p>0$ we may deduce the result from Theorem 4.12, but we omit the proof. $\square$
29. Proposition 6.25 incorrectly asserts unramifiedness
- ID:
676ec50c-c0c2-45ec-8927-ef7828e37a64 - Refine score:
0.31 - Original types: general
- Refine status: open
Comment
Normalization of a rational curve need not be unramified in the ambient smooth variety, so the asserted surjection onto $\Omega_{\mathbb{P}^{1}}^{1}$ is not justified. The conclusion still follows by using the image of the generically nonzero differential, which is a negative-degree line-bundle quotient of $\iota^*\Omega_X^1$.
Quoted passage
Proposition 6.25. Let $X$ be a smooth variety with nef cotangent bundle. Then $X$ contains no rational curves.
Proof. Suppose to the contrary that there is a non-constant morphism $f$ : $\mathbb{P}^{1} \rightarrow X$. Then the image of $f$ is a rational curve, $C$; taking its normalization gives an unramified map $\iota: \mathbb{P}^{1} \rightarrow X$. Thus there is a surjection
$$ \iota^{*} \Omega_{X}^{1} \rightarrow \Omega_{\mathbb{P}^{1}}^{1} \rightarrow 0 . $$But $\Omega_{\mathbb{P}^{1}}^{1}$ has negative degree, contradicting the nefness of $\Omega_{X}^{1}$. $\square$
30. Theorem 6.29 omits curves of genus above one
- ID:
f6103fd6-3f44-41e6-847a-e42cf3fe90a2 - Refine score:
0.29 - Original types: general
- Refine status: open
Comment
The proof of Theorem 6.29(1) treats only genus $1$, while the statement covers all $g\geq 1$. For $g>1$, $\deg(\Omega_Y^1)=2g-2>0$, so the cotangent line bundle is ample and hence arithmetically nef in positive characteristic; Proposition 6.28 then gives the conclusion. This case is presently omitted from the proof.
Quoted passage
Theorem 6.29. Let $Y$ be a smooth projective variety over a field $k$ such that one of the following holds:
(1) $Y$ is a curve of genus at least 1. (2) $Y$ has trivial cotangent bundle. (3) There exists a smooth map $f: Y \rightarrow X$ with $\Omega_{X}^{1}, \Omega_{Y / X}^{1} f$-semipositive. (4) There exists an étale morphism $g: Y \rightarrow Y^{\prime}$ with $\Omega_{Y^{\prime}}^{1} f$-semipositive. (5) There exists a finite étale morphism $g: Y^{\prime} \rightarrow Y$ with $\Omega_{Y^{\prime}}^{1}$ f-semipositive and such that $\operatorname{char}(k)$ does not divide $\operatorname{deg}(g)$. (6) $Y$ is a divisor in smooth variety $Y^{\prime}$ so that $\Omega_{Y^{\prime}}^{1}$ is $f$-semipositive and $\mathscr{N}_{Y / Y^{\prime}}^{\vee}$ is arithmetically nef.
Then $\Omega_{Y}^{1}$ is $f$-semipositive. Proof. In every case it suffices to work in positive characteristic.
(1) A curve of genus 1 has arithmetically nef cotangent bundle, so the result is immediate from Proposition 6.28.
31. Twist indexing in the proof of Theorem 6.35
- ID:
b397bf7b-9393-4b84-a20a-0507a31e20b7 - Refine score:
0.29 - Original types: general
- Refine status: open
Comment
The displayed restriction sequence has the wrong kernel twist for $D\in|\mathscr{L}^{\otimes n}|$: with twists measured by the fixed ample bundle, the kernel is $(s^*\Omega_{A/X}^1)^\vee\otimes\mathscr{L}^{-(n'+n)}$, not the $-(n'+1)$ twist shown. The subsequent vanishing argument still goes through because $n'+n\geq n$, so this is a local indexing error rather than a failure of Theorem 6.35.
Quoted passage
Let $D \in\left|\mathscr{L}^{\otimes n}\right|$ and let $r: D \rightarrow A$ be a section to $f_{D}$; we wish to show that $r$ extends uniquely to $X$. Now by Serre duality and the short exact sequence
$$ \left.0 \rightarrow\left(s^{*} \Omega_{A / X}^{1}\right)^{\vee}\left(-n^{\prime}-1\right) \rightarrow\left(s^{*} \Omega_{A / X}^{1}\right)^{\vee}\left(-n^{\prime}\right) \rightarrow\left(s^{*} \Omega_{A / X}^{1}\right)^{\vee}\left(-n^{\prime}\right)\right|_{D} \rightarrow 0 $$we have that
$$ H^{0}\left(D,\left.\left(s^{*} \Omega_{A / X}^{1}\right)^{\vee}\left(-n^{\prime}\right)\right|_{D}\right)=H^{1}\left(D,\left.\left(s^{*} \Omega_{A / X}^{1}\right)^{\vee}\left(-n^{\prime}\right)\right|_{D}\right)=0 $$for all $n^{\prime} \geq n$.
32. Characteristic scope of Remark 6.36
- ID:
4a3c0ab9-7cee-46e6-a17b-12b5c53d3bdc - Refine score:
0.24 - Original types: general
- Refine status: open
Comment
Remark 6.36 is supported by Corollary 6.6 only when $k$ has characteristic zero. Because Theorem 6.35 is stated over an arbitrary field, the remark otherwise appears to assert an unproved arbitrary-characteristic $n=1$ strengthening.
Quoted passage
But by Corollary 2.9, this map automatically extends to a section on some open neighborhood $U$ of $D$. As Abelian varieties contain no rational curves, such a section extends to all of $X$ by Proposition 3.5. Such an extension is unique by Corollary 2.10. $\square$
Remark 6.36. By Corollary 6.6, we may take $n=1$ above if
$$ \text { rel. } \operatorname{dim} .(f)<\operatorname{dim}(X)-1 . $$
Scope
- Paper:
03 Published and Submitted Work/Published/P20_Litt_Non_Abelian_Lefschetz.pdf - Refine report:
.refine/results/Published/P20_Litt_Non_Abelian_Lefschetz.review.json - Rubric:
rubric/comment_triage_rubric.md - Version assessed: local published PDF, Journal of Algebraic Geometry 27 (2018), 593–646
- Detailed Refine comments assessed: 32
- Assessment date: 2026-07-31
The local PDF is authoritative. All 32 passages and their relevant dependencies were checked there; formula-heavy printed pages 615, 619, 624, 634–635, and 643 were also rendered and visually inspected.
Summary
| # | Short title | Validity | Category | Standardness | Impact | Challenge | Repair | Disposition | Priority | Confidence |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Scope of “Calabi–Yau” | V3 | C5 | E-NA | I1 | Q0 | R1 | D1 | P3 | HIGH |
| 2 | Noncompact K3-type quotients | V4 | C9 | E-NA | I2 | Q0 | R1 | D3 | P2 | HIGH |
| 3 | Missing \(p>\dim X\) | V4 | C5 | E4 | I4 | Q5 | R4 | D4 | P0 | HIGH |
| 4 | Projectivity in Theorem 1.20 | V4 | C5 | E4 | I4 | Q5 | R4 | D4 | P0 | HIGH |
| 5 | Theorem 1.22 cross-reference | V4 | C7 | E3 | I2 | Q2 | R2 | D3 | P2 | MEDIUM |
| 6 | Generically finite versus finite | V4 | C5 | E-NA | I2 | Q0 | R1 | D3 | P2 | HIGH |
| 7 | Perfect pushforward in Lemma 2.1 | V4 | C6 | E3 | I2 | Q2 | R2 | D3 | P2 | MEDIUM |
| 8 | Relative versus global Ext | V4 | C5 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 9 | Serre-vanishing sign | V4 | C1 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 10 | \(\mathbb Q\)-factoriality in Corollary 3.4 | V4 | C5 | E3 | I2 | Q2 | R2 | D3 | P2 | HIGH |
| 11 | Descent in Proposition 3.5 | V4 | C6 | E3 | I2 | Q2 | R2 | D3 | P2 | HIGH |
| 12 | Thickening hypotheses | V4 | C5 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 13 | Le Potier threshold | V4 | C6 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 14 | Scope of Theorem 4.6 | V4 | C5 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 15 | Base-field mismatch | V4 | C4 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 16 | Duality degrees in Theorem 4.12 | V4 | C6 | E3 | I2 | Q2 | R2 | D3 | P2 | HIGH |
| 17 | Relative Frobenius twist | V4 | C4 | E3 | I2 | Q2 | R2 | D3 | P2 | HIGH |
| 18 | Nonreduced divisors in Lemma 4.28 | V4 | C6 | E4 | I4 | Q5 | R4 | D4 | P0 | HIGH |
| 19 | Properness in Theorem 4.29(3) | V4 | C5 | E-NA | I4 | Q5 | R4 | D4 | P0 | HIGH |
| 20 | Spreading-out obstruction | V4 | C6 | E3 | I2 | Q2 | R2 | D3 | P2 | HIGH |
| 21 | Projectivity in Theorem 5.1 | V4 | C5 | E-NA | I2 | Q2 | R2 | D3 | P2 | HIGH |
| 22 | Projective versus quasi-projective coarse space | V4 | C5 | E-NA | I2 | Q2 | R2 | D3 | P2 | HIGH |
| 23 | Overbroad equivalence in Lemma 5.4 | V4 | C6 | E3 | I2 | Q2 | R2 | D3 | P2 | HIGH |
| 24 | Properness in Theorem 6.5 | V4 | C5 | E-NA | I2 | Q2 | R2 | D3 | P2 | HIGH |
| 25 | Relative-duality degree in Lemma 6.9 | V4 | C1 | E-NA | I2 | Q2 | R2 | D3 | P2 | HIGH |
| 26 | Reversed duals in Lemma 6.12 | V4 | C6 | E-NA | I2 | Q2 | R2 | D3 | P2 | HIGH |
| 27 | Bundle maps in Theorem 6.16 | V4 | C4 | E3 | I2 | Q2 | R2 | D3 | P2 | HIGH |
| 28 | Surface case of Theorem 6.21 | V4 | C6 | E3 | I2 | Q2 | R2 | D3 | P2 | HIGH |
| 29 | Normalization need not be unramified | V4 | C6 | E2 | I2 | Q2 | R2 | D3 | P2 | HIGH |
| 30 | Genus \(>1\) case | V3 | C3 | E1 | I0 | Q1 | R0 | D0 | P4 | HIGH |
| 31 | Twist in Theorem 6.35 | V4 | C1 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 32 | Characteristic of Remark 6.36 | V4 | C5 | E-NA | I2 | Q0 | R1 | D3 | P2 | HIGH |
Comments 3, 4, 18, and 19 require scope-changing corrections and are classified I4/Q5/R4/D4. For Comment 3, add \(p>\dim X\) to the positive-characteristic branch of Theorem 1.10. Comments 4 and 19 have direct counterexamples to headline statements. Comment 18 has a direct counterexample to Lemma 4.28 and a complete safe repair—assume \(D\) reduced—but that repair narrows the paper's dependent existence theorems. All four require author/formal-corrigendum review.
Detailed assessments
1. “Calabi–Yau” needs the strict convention
ID: 55e7a22a-0e78-409c-8b94-c8d52512acd0 (PDF p. 595). Corollary 6.19(4) assumes \(h^{1,0}=h^{2,0}=0\) and a smooth polarized moduli space. These are available under the strict Calabi–Yau convention, so the introduction is not simply false, but it is ambiguous under trivial-canonical/weak conventions. Add the Hodge and moduli hypotheses or say “strict Calabi–Yau.” V3/C5/I1/R1/D1, Q0, high confidence.
2. The K3-type bullet exceeds Theorem 6.22
ID: 4ac4e8c9-a3cf-4846-b02e-bd0859c343a7 (PDF p. 595). Theorem 6.22 covers compact arithmetic Hermitian quotients and PEL Shimura varieties, and expressly says the required nefness is unknown for general noncompact Hermitian quotients. A general K3-type orthogonal quotient is usually noncompact. Restrict the bullet to compact K3-type quotients or to the hyperkähler moduli situation treated by Theorem 6.16. V4/C9/I2/R1/D3, Q0.
3. Theorem 1.10 omits \(p>\dim X\)
ID: c0c54465-4cf5-41a3-b896-35cbde5a6f23 (PDF pp. 598–599). Theorems 4.11, 4.29, and 5.7 all require \(p>\dim X\); the supplied Deligne–Illusie/Arapura argument uses it. The selected correction is to add \(p>\dim X\) to the positive-characteristic branch of Theorem 1.10 and align the corresponding introductory and summary statements. This exactly matches every supplied dependency and removes the unsupported small-characteristic branch. Because it narrows a headline theorem, the final classification is V4/C5/E4/I4/Q5/R4/D4, P0, high confidence; author/formal-corrigendum review is required.
4. Theorem 1.20 and Theorem 4.29 need projective \(X\)
ID: 5a94d466-4fd1-4362-af30-120f05029d2e (PDF pp. 602–603, 625). Their proofs use projective algebraization (Corollary 2.9), the projective extension results Corollaries 3.2 and 3.4, Serre duality and Theorem 4.11, and the projective hypotheses in Lemmas 4.27–4.28 and Theorem 4.21. The formal extension theorem, Theorem 4.12, is local and does not supply the missing global steps. An ample line bundle on a nonproper variety does not imply projectivity, and the stated nonproper theorem remains false even when the target is proper. Over an algebraically closed field \(k\) of characteristic \(p>3\), choose a smooth elliptic curve
and take
The principal divisor \(D\) has \(\mathscr O_X(D)\simeq\mathscr O_X\), which is ample on the affine scheme \(X\). Let \(Y=E\) and let \(f:D\to E\) be projection followed by the open immersion \(E\setminus\{\infty\}\hookrightarrow E\). Here \(Y\) is proper and contains no rational curves, \(\dim Y<\dim D\), \(N_{D/X}\otimes f^*\Omega_Y^1\simeq\mathscr O_D\), and \(\phi(\mathscr O_D)=0<1\) because \(D\) is affine. But \(f\) cannot extend to \(X\): every morphism \(\mathbb A^3\to E\) is constant, since its restriction to every affine line extends to \(\mathbb P^1\) and \(E\) has no rational curves, whereas \(f\) is nonconstant. Thus properness of \(Y\) does not repair the missing projectivity of \(X\); this counterexample satisfies both cases (1) and (3) of Theorem 4.29 and both alternatives of Theorem 1.20.
Thus both Theorems 1.20 and 4.29 need “projective” (equivalently here, proper plus the stated ample divisor) added to \(X\). This restores Lemma 4.27, Lemma 4.28, Theorem 4.11, the Serre-duality step, and the algebraization arguments, but materially narrows a headline theorem. V4/C5/E4/I4/Q5/R4/D4, P0, high confidence as to falsity and the projective repair; author/journal review is required.
5. Theorem 1.22 is not literally Theorem 6.32
ID: 0e43c860-d73a-462b-bb8d-937c45c21502 (PDF pp. 604, 642). The hypotheses differ: finite étale scheme cover versus quasi-projective coarse space, and \(\dim Y\) versus \(\dim\operatorname{im}f\). A repair is to remove the literal cross-reference and derive the stated finite-cover variant from Theorem 5.1(1), Lemma 6.31, and finite étale descent, with those steps written. V4/C7/E3/I2/Q2/R2/D3; medium confidence because the stack f-amplitude descent should be stated explicitly.
6. Remark 1.24 confuses finite and generically finite
ID: af9c7d98-6b01-4a8d-ac5e-bbc63f4e3c35 (PDF pp. 604–605). Torsion of the evaluation cokernel says global one-forms generically span, hence gives a generically finite Albanese map under the usual hypotheses, not a finite map. Blowing up an abelian variety at a point is a counterexample to the latter implication. Replace “finite-to-one/finite” by “generically finite onto its image,” or state the stronger condition separately. V4/C5/I2/R1/D3, Q0.
7. Lemma 2.1 does not establish a perfect pushforward
ID: 33619dea-12f7-4ac3-98eb-f242869326b3 (PDF p. 606). The final bidual amplitude step needs \(Rf_*(\mathscr F\otimes\mathscr L^{-n})\) perfect. Projectivity and bounded coherent \(f^!\mathscr O_S\) do not alone imply preservation of perfect complexes. Add that \(f\) is proper perfect/has finite Tor-dimension (or that the displayed pushforward is perfect). The later base-field applications are safe. V4/C6/E3/I2/Q2/R2/D3; medium confidence.
8. Corollary 2.6 proves relative Ext
ID: 7ef623f9-09c4-478c-a1d4-aa985581c444 (PDF p. 608). Corollary 2.5 identifies the relative cohomology sheaves of the derived pushforward. Global Ext needs a hypercohomology argument and care with negative degrees. State the result for relative \(\mathcal Ext^i_S\), or impose \(S=\operatorname{Spec}k\) (the later use). V4/C5/I2/R2/D3, Q0.
9. Corollary 2.7 reverses the positivity sign
ID: 0bde9f3c-5c14-4fdb-961b-78eedaa829d7 (PDF p. 609). The construction permits \(m_2-m_1\gg0\), and the Hom bundle is \(\mathscr O_Y(m_2-m_1)\). Replace \(m_1-m_2\gg0\) by \(m_2-m_1\gg0\). V4/C1/I2/R2/D3, Q0.
10. Corollary 3.4 lacks local \(\mathbb Q\)-factoriality
ID: 8157eacb-cc25-49aa-a261-ba2e9db49885 (PDF pp. 611–612). The proof invokes Corollary 3.2, whose source must be locally \(\mathbb Q\)-factorial; the chosen modification does not supply or descend that hypothesis. Add it to Corollary 3.4, or replace the modification by a verified \(\mathbb Q\)-factorial one and descend. Main applications have smooth \(X\). V4/C5/E3/I2/Q2/R2/D3.
11. Proposition 3.5 reverses the descent inference
ID: a69fabac-72a3-42e0-b501-3efaeeeab7db (PDF p. 612). The normalized closure \(X'\to Y\) is finite. An exceptional rational curve of \(b:X'\to X\) lies in a geometric fiber of \(Y\to X\); absence of rational curves makes its image a point, contradicting finiteness. Hence there is no exceptional divisor; purity gives quasi-finiteness of \(b\), and properness makes \(b\) finite. Correct the printed “finite over \(X'\)” to “finite over \(X\).” V4/C6/E3/I2/Q2/R2/D3.
12. Corollary 4.2 omits the lci square-zero setup
ID: 7cc93426-2edc-4273-a105-7587984c8dbb (PDF p. 613). Retain from Theorem 4.1 that \(D\hookrightarrow X\) is closed lci and \(D_2\) is defined by \(\mathscr I_D^2\); otherwise the normal bundle and obstruction statement are not defined as used. V4/C5/I2/R2/D3, Q0.
13. The second Le Potier range is false
ID: 7c9c2f13-b464-4b67-bc71-eed1c9da60e3 (PDF p. 615). For \(X=\mathbb P^1,E=\mathscr O(1),i=p=0\), the printed \(i+p\ge n-e\) predicts false vanishing. Replace it by the standard \(i+p\ge n+e\). The subsequent argument uses the other stated Le Potier vanishing. V4/C6/I2/R2/D3, Q0.
14. Theorem 4.6 has contradictory and unstated scope
ID: 984d2829-9304-4381-b95b-16c950dd0398 (PDF pp. 615–616). The opening \(\dim D\ge2\) makes the uniqueness bullet for \(\dim D=1\) vacuous, and Nakano positivity requires the complex/characteristic-zero setting announced immediately before. Remove the opening lower bound, place the separate bounds in the bullets, and say \(k=\mathbb C\) (or the precise characteristic-zero analytic setting). V4/C5/I2/R2/D3, Q0.
15. The positive-characteristic theorems mix \(k\) and \(L\)
ID: 49182a33-bc35-45ec-b3a4-6cfc9dfd4d87 (PDF pp. 602–603, 617, 625). The relative Frobenius \(F_{Y/L}\), twists \(Y^{(p^k)}\), and \(f:D\to Y\) require \(Y\) to be a smooth \(L\)-variety and \(f\) an \(L\)-morphism. Replace every stray \(k\) by \(L\), or specify an extension \(L\to k\) and use the corresponding relative constructions. V4/C4/I2/R2/D3, Q0.
16. Theorem 4.12's duality display has the wrong degree
ID: 546b1d86-6aa5-4097-9343-c6a51a4d9a78 (PDF p. 619). With \(K_D\) in degrees \([-d,-r]\),
not the printed un-dualized \(H^{d-i}\). In the hypercohomology spectral sequence, a term contributing to degree \(-i\) has \(a=-i-b\ge r-i\); for \(i=0,1\), the hypothesis \(\phi<r-1\) kills it. This is a complete local proof repair. V4/C6/E3/I2/Q2/R2/D3.
17. Lemma 4.26 must use relative Frobenius
ID: 2643befe-721d-48a7-a887-7b1ea4831919 (PDF p. 623). Absolute Frobenius is not generally a \(k\)-morphism. State factorization through \(F_{Y/k}:Y\to Y^{(p)}\), or explicitly work in schemes rather than \(k\)-schemes; on rings, the factor sends \(c\) to \(c^{1/p}\). With that scalar rule, reducedness shows the chosen roots kill \(I\). V4/C4/E3/I2/Q2/R2/D3.
18. Lemma 4.28 is false for nonreduced divisors
ID: fe79d33b-ea3b-41a4-a676-c09687981085 (PDF pp. 624–626). For \(X=\mathbb P^3_{\mathbb F_p}\), \(D=V(z^2)\), and \(Y=\mathbb P^2\), the distinct maps
have equal composites with Frobenius when \(p>3\). The expression \(g_1^\#-g_2^\#\) is an additive map into the nilradical, but it is not in general an \(f^{-1}\mathscr O_Y\)-linear map, so the printed subtraction does not prove uniqueness.
The complete safe repair is to assume that \(D\) is reduced. Frobenius \(F_D:D\to D\) is then a universal homeomorphism whose map on structure sheaves is injective, hence an epimorphism of schemes. In the proof of Theorem 4.29, after writing \(\bar f=F_Y\circ h\), restriction to \(D\) and universal commutativity of Frobenius give
Epimorphy of \(F_D\) therefore gives
which completes the printed induction on \(k\).
Reducedness must be added to Lemma 4.28 and to the existence statements depending on this Frobenius-descent step: Theorems 1.20, 4.29, 4.22, 4.31, and 5.1, together with the corresponding existence clauses and applications in Theorems 1.8, 1.10, 1.11, 6.1, 6.2, and 6.32. The uniqueness-only results (Theorems 4.21 and 5.7 and the “at most one” clauses) are unaffected, as is Theorem 6.35, which has a separate direct deformation argument. In characteristic zero, a reduced divisor is geometrically reduced; after taking a flat model and shrinking the spreading-out base, the closed fibers remain geometrically reduced, so the corrected positive-characteristic theorem applies.
This is a complete statement-and-proof repair, but it changes the scope of headline existence theorems and does not recover the advertised arbitrary nonreduced case. The strongest more general replacement would require vanishing of
for every nilpotent layer and Frobenius predecessor \(u\); the paper's stated f-amplitude hypothesis does not establish these vanishings. V4/C6/E4/I4/Q5/R4/D4, P0, high confidence; author/formal-corrigendum review required.
19. Properness is essential in Theorem 4.29(3)
ID: 15d9f012-0196-48c3-aa58-9cf08074a96e (PDF p. 625). Proposition 3.5 requires a proper target over \(X\). More strongly, take a liftable abelian variety \(A\) of dimension at least three in characteristic \(p>\dim A\), an ample divisor \(D\) avoiding \(0\), and \(Y=A\setminus\{0\}\). The inclusion \(D\to Y\) satisfies the no-rational- curves and amplitude conditions but cannot extend to \(A\to Y\): composing with \(Y\hookrightarrow A\) would give a map agreeing with the identity on the ample divisor, hence the identity, which hits \(0\).
This disproves case (3), and the same omission occurs in Theorem 1.20(2). Adding properness repairs the proof but changes a main theorem's scope. Independently confirmed V4/C5/E-NA/I4/Q5/R4/D4, P0, high confidence; author/formal-corrigendum review is required.
20. The spreading-out obstruction is misstated
ID: ee82fb36-552d-4a0f-8277-77bff7965266 (PDF pp. 626–627). The obstruction from \(D_{n-1}\) to \(D_n\) uses \(\mathscr I_D^{n-1}/\mathscr I_D^n\), not \(\mathscr I_D^{n-1}/\mathscr I_D\). After shrinking the spreading base, cohomology-and-base-change identifies the relative obstruction with each fiber obstruction; a coherent section vanishing at all closed points is zero. This supplies the missing argument. V4/C6/E3/I2/Q2/R2/D3.
21. Theorem 5.1 needs projective \(X\)
ID: fa8fc2d8-b992-47fb-a649-51d9a88211f3 (PDF pp. 627–630). Lemmas 5.2 and 5.4 invoke results whose base is projective. Add “projective” to Theorem 5.1. The headline Theorem 1.10 already has it, so all advertised applications survive. V4/C5/I2/Q2/R2/D3.
22. Theorem 5.1(2) should say quasi-projective
ID: 3be8178f-17af-444a-9485-e224515486c9 (PDF pp. 628–630, 642). The proof uses Corollary 3.4, which asks for a quasi-projective target, while Theorems 1.10 and 6.32 claim quasi-projective coarse space. Replace “projective” in case (2) by “quasi-projective”; the proof and later statements then agree. V4/C5/I2/Q2/R2/D3.
23. Lemma 5.4 states an unrestricted equivalence
ID: 6a9e9bda-c4e0-474c-8586-0ef3e6871093 (PDF pp. 629–630). Case (2) extends only the particular object whose restriction to \(D\) has small image. Rewrite essential surjectivity objectwise under that condition, and prove full faithfulness separately using the formal uniqueness/diagonal argument (Corollary 2.10 and purity). This is enough for Theorem 5.1. V4/C6/E3/I2/Q2/R2/D3.
24. Theorem 6.5 is false for nonproper \(f\)
ID: d268af38-2ce5-40cc-a0d1-366b7017fca0 (PDF pp. 633–634). For \(Y=\operatorname{Tot}(\mathscr O_{\mathbb P^1}(1))\to\mathbb P^1\), relative cotangent is the pullback of \(\mathscr O(-1)\) and is relatively globally generated, but not nef. The pushforward is not a finite-rank Hodge bundle. Add smooth properness (and the standard Hodge/base-change hypotheses). Corollary 6.6 already has properness. V4/C5/I2/Q2/R2/D3.
25. Lemma 6.9 uses \(R^1\) instead of \(R^{n-1}\)
ID: c4c9de67-751e-46ef-9ba2-5a0c48c91478 (PDF p. 634). Relative duality gives
After \(\Omega^1_{Y/X}=f^*f_*\Omega^1_{Y/X}\), the second factor is \(R^{n-1}f_*\omega\simeq(R^1f_*\mathscr O_Y)^\vee\). Both are Hodge bundles with the required nefness. V4/C1/I2/Q2/R2/D3.
26. Lemma 6.12 reverses every dual
ID: 87eae2a5-4512-4ddb-9ba5-4ff0a0e72793 (PDF p. 635). If \(W=(f_*\Omega^1)^\vee\otimes R^1f_*\mathscr O\), Lemma 6.9 makes \(W^\vee\) nef. Dualizing \(R^1f_*T\hookrightarrow W\) gives \(W^\vee\twoheadrightarrow(R^1f_*T)^\vee\), and the pulled-back moduli cotangent is a quotient of this nef bundle. Replace \(f^*\Omega_M\) by the classifying pullback \(g^*\Omega_M\). V4/C6/I2/Q2/R2/D3.
27. Theorem 6.16's displayed chain is not typed
ID: ad52faff-e6a4-4514-ae7d-d0a629054520 (PDF pp. 635–636). The corrected beginning is
where \(a:A\to X\) is Kuga–Satake. Continue with \((a_*\Omega^1_{A/X})^\vee\otimes R^1a_*\mathscr O_A\). The Kuga–Satake Hodge map followed by local Torelli makes this composite injective; its target has nef dual by corrected Lemma 6.9, so \(g^*\Omega_{\mathscr M}\) is nef. V4/C4/E3/I2/Q2/R2/D3.
28. The displayed proof of Theorem 6.21 misses surfaces
ID: a2b24267-f752-4755-9adb-d901c1bf72ac (PDF pp. 637–638). Theorem 5.1 and Theorem 4.12 require dimension at least three and therefore do not prove surjectivity when \(\dim X=2\). The theorem is independently cited from SGA; in this paper, use the dimension-two full-faithfulness result (Theorem 5.7 applied to finite étale torsors) or cite the external theorem for the first bullet. V4/C6/E3/I2/Q2/R2/D3.
29. Normalization of a rational curve need not be unramified
ID: 676ec50c-c0c2-45ec-8927-ef7828e37a64 (PDF p. 639). The normalization map is generically immersive, not necessarily unramified. The image of \(\iota^*\Omega_X^1\to\Omega_{\mathbb P^1}^1\) is a line bundle \(\Omega_{\mathbb P^1}^1(-R)\) of negative degree, hence a quotient of \(\iota^*\Omega_X^1\), contradicting nefness. This also handles ramification. V4/C6/E2/I2/Q2/R2/D3.
30. The genus \(>1\) case is safely implicit
ID: f6103fd6-3f44-41e6-847a-e42cf3fe90a2 (PDF pp. 640–641). For \(g>1\), \(\deg\Omega_Y^1=2g-2>0\), so it is ample and arithmetically nef; Proposition 6.28 immediately implies f-semipositivity. This one-line case is easier than the written genus-one case and safe for the intended reader. V3/C3/E1/I0/Q1/R0/D0; no erratum is needed.
31. Theorem 6.35 has the wrong kernel twist
ID: b397bf7b-9393-4b84-a20a-0507a31e20b7 (PDF p. 643). For \(D\in|\mathscr L^n|\), restricting \(E\otimes\mathscr L^{-n'}\) has kernel \(E\otimes\mathscr L^{-(n'+n)}\), not exponent \(-(n'+1)\). The vanishing remains valid because \(n'+n\ge n\). V4/C1/I2/R2/D3, Q0.
32. Remark 6.36 is only proved in characteristic zero
ID: 4a3c0ab9-7cee-46e6-a17b-12b5c53d3bdc (PDF p. 643). Corollary 6.6 assumes characteristic zero, while Theorem 6.35 is arbitrary characteristic. Begin the remark with “If \(\operatorname{char}k=0\).” V4/C5/I2/R1/D3, Q0.
P21 Manifolds containing an ample $\mathbb{P}^{\mathbf{1}}$-bundle0 detailed comments · no adopted correction
No adopted erratum
The detailed Refine report contains no anchored comments, and the audit adopted no public correction for this paper.
Report metadata
| Field | Value |
|---|---|
| Category | Published |
| Processing status | completed |
| Detailed comments | 0 |
| Domain | stem/mathematics |
| Completed | 2026-07-29T18:03:09.091860+00:00 |
| Refine document ID | 37bfb38e-d556-4970-8876-923cfdb76050 |
Refine summary
This paper investigates Sommese's conjecture on the classification of smooth projective varieties containing a projective bundle as an ample divisor. The author proves the conjecture when the base variety has Picard rank 1 or is not uniruled, and otherwise reduces it to a conjectural characterization of projective spaces involving ample vector bundles.
Overall feedback
Formal extension induction in Lemma 4
Lemma 4 successfully identifies the obstruction group for extending the morphism from one infinitesimal neighborhood to the next. However, the induction linking the obstruction class sequence to the global extension is currently left implicit.
The argument must explicitly run this induction. It is necessary to state that the vanishing of each obstruction class is sufficient to choose the next extension, that failure at some stage produces the asserted nonzero morphism, and that an infinite compatible sequence precisely defines a morphism from the formal completion. Invoking the smoothness of the target and the square-zero nature of each thickening will fully justify these implications.
Direct image properties and Conjecture 2
A pivotal POSITIVITY step in Lemma 4 asserts from [9, Theorem 1.2] that $p_{*}(\omega_{Y/Z} \otimes \mathscr{O}_Y(nY))$ is either zero or ample. Because Conjecture 2 strictly requires an ample vector bundle, the local freeness and the precise hypotheses of [9] need to be verified. It is essential to confirm that [9] applies to this smooth projective $\mathbb{P}^1$-bundle and yields pure ampleness rather than a weaker positivity property.
Fiberwise, this direct image yields $H^0(\mathbb{P}^1, \mathscr{O}(na-2))$, where $a$ is the positive fiber degree of $\mathscr{O}_Y(Y)$. Applying cohomology and base change here will establish whether the direct image is zero or locally free of constant positive rank.
The subsheaf formulation in Corollary 6
Corollary 6 relies on the statement that $T_Z$ "contains an ample subsheaf," based on the existence of the morphism $\mathscr{E} \to T_Z$. A nonzero bundle morphism can have variable rank, meaning its image may be a non-locally-free torsion-free sheaf. The statement regarding an ample subsheaf therefore requires further structural justification.
For the non-uniruled branch, this can be addressed by restricting the argument to a sufficiently general complete-intersection curve, taking the saturated image, and utilizing the fact that it is a positive-degree quotient of the ample bundle $\mathscr{E}|_C$. Dualizing then supplies the required contradiction with Miyaoka's generic nefness of $\Omega_Z$. For the Picard-rank-one branch, the subsheaf formulation can simply be bypassed by directly invoking [2, Corollary 4.3] in the nonzero-Hom form described in Remark 3.
Hypothesis matching in Lemma 5 and Corollary 6
Lemma 5 acts as the singular step converting the formal extension from Lemma 4 into the global morphism necessary for Sommese's classification. The proof currently summarizes two non-abelian Lefschetz citations, but the mechanism requires exact hypothesis and conclusion matching. The text must explicitly demonstrate that the first result algebraizes the formal morphism to a morphism on a Zariski neighborhood of $Y$, and the second result makes the resulting rational map $X \dashrightarrow Z$ regular under the condition $\dim Z < \dim Y$. Additionally, the phrase "this rational map to $Y$" specifies the wrong target and should be corrected to $Z$.
The proof must also account for uniqueness. In Corollary 6, it is necessary to explicitly check that for a $\mathbb{P}^1$-bundle, $\dim Y = \dim Z + 1$ and $\dim X = \dim Z + 2$, establishing that all underlying hypotheses of Lemma 5 rigorously apply.
Detailed comments
Refine returned no detailed comments for this paper.
Scope
- Paper:
03 Published and Submitted Work/Published/P21_Litt_Manifolds_Ample_P1_Bundle.pdf - Refine report:
.refine/results/Published/P21_Litt_Manifolds_Ample_P1_Bundle.review.json - Rubric:
rubric/comment_triage_rubric.md - Version assessed: local published PDF, manuscripta mathematica 152 (2017), 533-537
- Detailed Refine comments assessed: 0
- Assessment date: 2026-07-30
The report's feedback.detailed.comments array is empty. Accordingly, there are no anchored Refine allegations to classify and no assessment rows to place in the accompanying CSV.
The report also contains unanchored prose in feedback.overall. Consistently with the project rubric's unit-of-review rule, that prose was retained as context but was not converted into comment rows: it does not supply detailed comment IDs, scores, or individually anchored allegations. This zero-comment result records the contents of the Refine detailed-comment report; it is not an independent certification that the paper has no possible errors.
Summary
| Detailed comments | Assessment rows | Errata candidates | Escalations |
|---|---|---|---|
| 0 | 0 | 0 | 0 |
No comment-level classification or severity challenge was triggered.
P22 Zeta Functions of Curves with No Rational Points7 detailed comments · 5 numbered corrections 2 I15 I2
These errata refer to Daniel Litt, “Zeta Functions of Curves with No Rational Points,” Michigan Mathematical Journal 64, no. 2 (2015), pp. 383--395, doi:10.1307/mmj/1434731929. Page references below are to that published version.
Page 386, Remark 12. The comparison with Kapranov's remark says “is rational,” although the claim under discussion is that the indicated expression is a polynomial. Replace the sentence beginning “The remark states” by:
The remark states that
\[ (1-\mathbb L^nt^n)(1-t^n)Z_X(t) \]is a polynomial, where $n>0$ is minimal such that $\operatorname{Pic}^n(X)(k)\neq\varnothing$; in the example, $\operatorname{Pic}^1(X)=\operatorname{Spec}(\mathbb R)$, so the remark suggests that $(1-\mathbb Lt)(1-t)Z_X(t)$ is a polynomial.
This correction concerns only the comparison with Remark 1.3.5(a) of [6]; Theorem 8 and its proof are unchanged.
Page 386, paragraph following Remark 12. The assertion that the Abel--Jacobi morphism is a Severi--Brauer scheme requires the stable-range hypothesis $n>2g-2$. Replace the second sentence of the paragraph, beginning “Of course,” by:
For $n>2g-2$ (and assuming that $C$ is geometrically connected), after a finite extension of the base field the Abel--Jacobi morphism is a projective-space bundle over $\operatorname{Pic}^n(C)$; hence, in this range,
\[ \operatorname{Sym}^n(C)\longrightarrow\operatorname{Pic}^n(C) \]is a Severi--Brauer scheme over $\operatorname{Pic}^n(C)$.
Section 5 already imposes $n>2g-2$ before using this description, so the proof of Theorem 8 is unchanged.
Page 389, paragraph following Corollary 22. The projectivization and endomorphism-algebra constructions require a twisted vector bundle, rather than an arbitrary twisted quasi-coherent sheaf. Replace the three sentences beginning “Similarly, given an $\alpha$-twisted sheaf” through “consider $\operatorname{End}(\mathcal E)$” by:
Similarly, given an $\alpha$-twisted vector bundle $\mathcal E$ of positive locally constant rank over a scheme $X$, we may obtain a Severi--Brauer scheme with Brauer class $\alpha$ by considering $\mathbb P(\mathcal E)$, which gives \'{e}tale descent data for a scheme over $X$. Since $\mathbb P(\mathcal E)$ is anticanonically polarized over $X$, these descent data are effective, and we obtain a Severi--Brauer scheme over $X$. To obtain an Azumaya algebra with Brauer class $\alpha$, consider $\operatorname{End}(\mathcal E)$.
The constructions used in Corollary 23, Theorem 24, Proposition 28, and Section 5 already involve twisted vector bundles, so those results are unchanged.
Pages 389--390, final paragraph of Section 3, and pages 392--393, Section 5. A $\operatorname{PGL}$-valued \v{C}ech $1$-cocycle generally lifts locally to a $\operatorname{GL}$-valued $1$-cochain, not to a $\operatorname{GL}$-valued $1$-cocycle. Its scalar coboundary records the Brauer class.
On pages 389--390, replace the final sentence of the paragraph by:
It is not hard to see that every Severi--Brauer variety or Azumaya algebra is obtained in this fashion; indeed, after refining the cover if necessary, choose local $\operatorname{GL}_n$-lifts of the $\operatorname{PGL}_n$-valued \v{C}ech $1$-cocycle defining the Severi--Brauer variety or Azumaya algebra. These lifts form a \v{C}ech $1$-cochain, and their scalar coboundary is a $\mathbb G_m$-valued \v{C}ech $2$-cocycle representing $\alpha$.
On pages 392--393, replace the sentence beginning “Choosing an arbitrary lift” by:
After refining the cover $\operatorname{Pic}^n(C)_K\to\operatorname{Pic}^n(C)$ if necessary, choose local lifts of this $\operatorname{PGL}(p_{K*}\mathcal L_n)$-valued $1$-cocycle to $\operatorname{GL}(p_{K*}\mathcal L_n)$. These lifts form a \v{C}ech $1$-cochain whose scalar coboundary is a $\mathbb G_m$-valued \v{C}ech $2$-cocycle representing a class
\[ \alpha\in H^2\!\left(\operatorname{Pic}^n(C),\mathbb G_m\right). \]Accordingly, $p_{K*}\mathcal L_n$ with this twisted descent datum is an $\alpha$-twisted vector bundle $\mathcal F_n$ on $\operatorname{Pic}^n(C)$, and
\[ \operatorname{Sym}^n(C)\simeq \mathbb P_{\operatorname{Pic}^n(C)}(\mathcal F_n). \]This supplies the twisted descent datum used in Section 5. Theorem 24 and the recurrence in the proof of Theorem 8 are unchanged.
Page 394, Corollary 30 and its proof. Over an imperfect field, the normalization of a reduced curve can be regular without being geometrically regular, so its projective model need not satisfy the smoothness hypothesis of Theorem 8. Replace the statement of Corollary 30 by:
Corollary 30. Let $C$ be a curve over $k$, let $C^\nu$ be the normalization of $C_{\mathrm{red}}$, and suppose that every irreducible component of $C^\nu$ is geometrically regular and geometrically irreducible over $k$. Then there exists a polynomial
\[ p(t)\in1+tK_0(\operatorname{Var}_k)[t] \]such that $p(t)Z_C(t)$ is a polynomial with constant term $1$.
In the proof, replace the opening through the displayed scissor relation by:
We reduce to the case of a smooth projective curve. Since $[C]=[C_{\mathrm{red}}]$, we may first assume that $C$ is reduced. Let $\widetilde C$ be the disjoint union of the smooth projective models of the irreducible components of $C^\nu$. Normalization and compactification change the curve only along zero-dimensional subschemes, so there exist zero-dimensional $k$-schemes $X$ and $Y$ such that
\[ [C]=[\widetilde C]+[X]-[Y]. \]Finally, replace the sentence beginning “But $\widetilde C$ is a disjoint union” by:
By the hypotheses, $\widetilde C$ is a disjoint union of smooth, projective, geometrically connected curves $C_i$. Each $C_i$ satisfies the conditions of Theorem 8, and
\[ Z_{\widetilde C}(t)=\prod_i Z_{C_i}(t), \]so we are done.
This changes only the scope of Corollary 30. Theorem 8 is unchanged, and no later theorem depends on the corollary.
Report metadata
| Field | Value |
|---|---|
| Category | Published |
| Processing status | completed |
| Detailed comments | 7 |
| Domain | stem/mathematics |
| Completed | 2026-07-29T18:04:20.906528+00:00 |
| Refine document ID | b4a09c04-d093-4cec-b635-89c76e1f91a5 |
Refine summary
This paper shows that the motivic zeta functions of smooth, geometrically connected curves without rational points are rational functions. It achieves this by studying the class of a Severi-Brauer scheme over a general base in the Grothendieck ring of varieties.
Overall feedback
Here are some observations regarding the geometric and structural arguments in the paper.
Projectivization conventions
The text utilizes different projectivization conventions at various stages of the argument, which complicates the global coherence of the proofs. In Section 3, $P(E)$ is defined as the scheme of hyperplanes, utilizing the rank-one quotient convention. Conversely, Theorem 24 operates on the line convention, where an injection $E_1 \to E_2$ is treated as inducing a closed immersion $P(E_1) \to P(E_2)$. This variance appears again in Section 5, where multiplication by the section defining $D$ is used to obtain $P(F_m) \to P(F_{m+n})$, implicitly relying on effective divisors corresponding to lines of sections.
Adopting a single, uniform convention throughout the paper is necessary. Once chosen, the consequences must be explicitly tracked by rechecking Theorem 24, Lemma 25, and Corollary 23, alongside the Abel–Jacobi identification, normal-bundle ranks, Brauer-class signs, and Lefschetz powers.
Twisted descent and local freeness
Sections 3 and 5 state that a PGL-valued 1-cocycle can be lifted to a GL-valued 1-cocycle. In general, lifting such a cocycle yields a GL-valued cochain whose scalar coboundary represents the Brauer obstruction, whereas an actual GL cocycle would trivialize that obstruction. Because the twisted bundles $F_m$ and $F_{m+n}$ are chosen independently, multiplication by the section of $O_C(D)$ does not automatically descend to the asserted morphism $b_D^m$, nor does it guarantee translation-compatible twisting classes. Constructing this requires compatible descent data or passage through the Picard gerbe.
Additionally, the deduction that $R_m = \operatorname{coker}(b_D^m)$ is locally free of rank $n$ requires explicit justification. Showing this demands an application of cohomology-and-base-change and the uniform vanishing of $R^1p_*L_m$ for $m>2g-2$. Without establishing these properties for $R_m$, Theorem 24 and Proposition 27 cannot be safely applied to obtain the required recurrence.
Stratified reduction over arbitrary bases
The proof of Theorem 24 specifies that "without loss of generality, $S$ is integral and affine." However, the theorem itself is stated for every finite-type scheme and asserts an equality in the unlocalized Grothendieck ring. Establishing the formula over an arbitrary base requires supplying a finite affine stratification, splitting the restricted exact sequence on each individual stratum, explicitly accounting for the reductions of nonreduced strata, and finally assembling the formula using scissor relations.
Similarly, the argument for Proposition 28 requires explicit justification for several steps. The generic simple twisted bundle, its spreading out, the resulting direct-sum decomposition after shrinking, and the independence of $P$ under common refinements all need to be formalized with the appropriate stratifications, as this relative theorem forms the principal independent technical contribution of the work.
Genus zero generating series
In Theorem 8, the correction term is summed over the index $m=2g-1$ to $2g+n-2$. For $g=0$, the index begins at $m=-1$, which introduces undefined objects $P_{-1}$ and $t^{-1}$ into a formal power series expression. This directly affects the point-free genus-zero case highlighted in Example 11 and Remark 12. The argument must instead select a complete set of nonnegative representatives modulo $n$, which will require recomputing the finite initial polynomial and the associated index shifts.
Smooth projective models in Corollary 30
The argument in Corollary 30 invokes a smooth projective model beyond the stated hypotheses. Over an imperfect field $k$, the normalization of a curve is not guaranteed to be smooth over $k$. Furthermore, geometric irreducibility is insufficient to rule out geometric nonreducedness or inseparable phenomena. Consequently, asserting that the normalization possesses a smooth projective model whose components satisfy Theorem 8 does not hold under the current assumptions. The corollary requires either additional hypotheses guaranteeing smooth normalized components, or a distinct argument extended to regular, non-smooth curves.
Detailed comments
1. Remark 12 switches polynomial to rational
- ID:
4d42f684-efc7-48ea-a13e-4932ba2ad9ae - Refine score:
0.29 - Original types: general
- Refine status: open
Comment
Remark 12 changes the asserted conclusion from polynomiality to rationality. With $n=1$, the stated claim would make $(1-\mathbb{L}t)(1-t)Z_X(t)$ a polynomial; mere rationality already follows from Theorem 8, so the subsequent skepticism can coherently concern only polynomiality.
Quoted passage
The remark states that $\left(1-\mathbb{L}^{n} t^{n}\right)\left(1-t^{n}\right) Z_{X}(t)$ is a polynomial, where $n>0$ is minimal such that $\operatorname{Pic}^{n}(X)(k) \neq \emptyset$; in the example, $\operatorname{Pic}^{1}(X)=\operatorname{Spec}(\mathbb{R})$, so the remark suggests that $(1-\mathbb{L} t)(1-t) Z_{X}(t)$ is rational. We do not know a proof of this fact and do not believe it to be true (though we have no proof that it is false).
2. Stable-range condition missing in Severi–Brauer claim
- ID:
88acb30c-41c9-490a-bd91-2ab5c5595dff - Refine score:
0.3 - Original types: general
- Refine status: open
Comment
The conclusion that the Abel–Jacobi map is a Severi–Brauer scheme is valid only in the stable range $n>2g-2$. Outside that range, the dimensions of the fibers may vary and the map need not be surjective, defects that no finite extension of the base field can remove.
Quoted passage
The issue identified in Example 11 is that $\operatorname{Sym}^{n}(C) \rightarrow \operatorname{Pic}^{n}(C)$ may not be a Zariski fiber bundle. Of course (if $C$ is geometrically connected), after a finite extension of the base field, we recover the usual situation of a projective space bundle over the $\operatorname{Pic}^{n}(C)$, so in general $\operatorname{Sym}^{n}(C) \rightarrow \operatorname{Pic}^{n}(C)$ is a Severi-Brauer scheme over $\operatorname{Pic}^{n}(C)$. Thus, we will proceed by studying the class $[V]$ of a Severi-Brauer $S$-scheme $V / S$ in $K_{0}\left(\operatorname{Var}_{k}\right)$.
3. Cocycle/cochain conflict in Remark 14
- ID:
bfa7fa1c-3ff4-4098-909d-a1f88f8bf748 - Refine score:
0.23 - Original types: general
- Refine status: open
Comment
In Remark 14, $\beta$ must be a Čech $1$-cochain, not a $1$-cocycle: a $1$-cocycle satisfies $d\beta=1$, whereas equality of the cohomology classes of $\lambda$ and $\lambda'$ implies that $\lambda^{-1}\lambda'$ is a coboundary. The displayed equivalence is otherwise correct.
Quoted passage
Remark 14. A priori, the definition of $\mathrm{QCoh}(X, \alpha)$ depends on the choice of cocycle $\lambda$ representing $\alpha \in H^{2}\left(X, \mathbb{G}_{m}\right)$. However, if $\lambda$ and $\lambda^{\prime}$ are two cocycles representing $\alpha$, then the categories of twisted sheaves they define are (noncanonically) equivalent [2, Lemma 1.2.8]. Namely, refine the covers on which $\lambda$ and $\lambda^{\prime}$ are defined and choose a 1 -cocycle $\beta$ with $d \beta=\lambda^{-1} \lambda^{\prime}$. Then the functor
$$ (\mathcal{E}, \phi) \mapsto(\mathcal{E}, \beta \phi) $$
4. Projectivization requires a twisted vector bundle
- ID:
e619e3cf-f88e-49b7-beb6-dbd0825ec224 - Refine score:
0.3 - Original types: general
- Refine status: open
Comment
The assertion about $\mathbb{P}(\mathcal{E})$ is valid only when $\mathcal{E}$ is an $\alpha$-twisted vector bundle of positive locally constant rank. For a general object of $\operatorname{QCoh}(X,\alpha)$, the projectivization need not be étale-locally a projective-space bundle and therefore need not be a Severi–Brauer scheme.
Quoted passage
Similarly, given an $\alpha$-twisted sheaf $\mathcal{E}$ over a scheme $X$, we may obtain a Severi-Brauer variety with Brauer class $\alpha$ by considering $\mathbb{P}(\mathcal{E})$, which is étale descent data for a scheme over $X$. Since $\mathbb{P}(\mathcal{E})$ is anticanonically polarized over $X$, these descent data are effective, and we obtain a Severi-Brauer variety over $X$.
5. The PGL cocycle does not lift to a GL cocycle in Sections 3 and 5
- ID:
1ba37ac3-97f8-4cb5-8309-1080a3494729 - Refine score:
0.34 - Original types: general
- Refine status: open
Comment
A $\mathrm{PGL}_n$-cocycle generally lifts only to a $\mathrm{GL}_n$-valued Čech 1-cochain, not to a $\mathrm{GL}_n$-cocycle. The scalar failure of the lifted matrices to satisfy the cocycle condition is the Čech 2-cocycle representing the Brauer class; an actual $\mathrm{GL}_n$-cocycle lift would instead give an ordinary vector bundle and a split Brauer class. This same issue appears in Section 5, where the lift of the $\operatorname{PGL}$-valued cocycle for the descent data of $\operatorname{Sym}^{n}(C)$ is similarly misidentified as a $\operatorname{GL}$-valued cocycle.
Quoted passage
It is not hard to see that every Severi-Brauer variety or Azumaya algebra is obtained in this fashion; indeed, take the $\mathrm{PGL}_{n}$-cocycle defining the
Severi-Brauer variety or Azumaya algebra and lift it to an arbitrary cocycle for $\mathrm{GL}_{n}$. (To do so, we may have to refine the cover on which the cocycle is defined.)
6. Map direction in the proof of Theorem 24
- ID:
9df8f779-ce61-413a-ab5b-fe48eed5fccb - Refine score:
0.28 - Original types: general
- Refine status: open
Comment
Under the paper’s convention that $\mathbb{P}(\mathcal{E})$ parametrizes rank-one quotients, the inclusion $\mathcal{E}_1\hookrightarrow\mathcal{E}_2$ does not induce the displayed closed embedding. After the exact sequence has been split, the required embedding is instead induced by the resulting projection $\mathcal{E}_2\twoheadrightarrow\mathcal{E}_1$. The subsequent complement calculation remains valid with this correction.
Quoted passage
The morphism $\mathcal{E}_{1} \rightarrow \mathcal{E}_{2}$ induces a closed embedding $\mathbb{P}\left(\mathcal{E}_{1}\right) \hookrightarrow \mathbb{P}\left(\mathcal{E}_{2}\right)$, so
$$ \left[\mathbb{P}\left(\mathcal{E}_{2}\right)\right]=\left[\mathbb{P}\left(\mathcal{E}_{1}\right)\right]+[U], $$where $U:=\mathbb{P}\left(\mathcal{E}_{2}\right) \backslash \mathbb{P}\left(\mathcal{E}_{1}\right)$. We wish to identify $U$ with the total space of a vector bundle over $\mathbb{P}\left(\mathcal{E}_{3}\right)$.
7. Corollary 30 needs geometric reducedness
- ID:
a4a64a0e-6345-40fc-a8b6-64154ee393b6 - Refine score:
0.48 - Original types: general
- Refine status: open
Comment
The final application of Theorem 8 is not justified over an arbitrary imperfect field: a normal, hence regular, geometrically irreducible curve need not be smooth over $k$, because geometric irreducibility does not ensure geometric regularity or even geometric reducedness. The normalized components therefore require a hypothesis ensuring smoothness, such as geometric regularity or perfectness of $k$, or a separate treatment of the inseparable case.
Quoted passage
Proof. We reduce to the case where $C$ is smooth and projective. Indeed, we may assume that $C$ is reduced as $[C]=\left[C_{\text {red }}\right]$; let $\tilde{C}$ be the smooth projective model of $C$. Then $[C]=[\tilde{C}]+[X]-[Y]$, where $X$ and $Y$ are zero-dimensional schemes. In particular,
$$ Z_{C}(t) Z_{Y}(t)=Z_{\tilde{C}}(t) Z_{X}(t) $$by Remark 3. We leave to the reader to show that there exist polynomials $p_{X}(t), p_{Y}(t) \in 1+t K_{0}\left(\operatorname{Var}_{k}\right)[t]$ such that
$$ p_{X}(t) Z_{X}(t), p_{Y}(t) Z_{Y}(t) $$are polynomials with constant term one; thus, to prove the theorem for $C$, it suffices to prove it for $\tilde{C}$. But $\tilde{C}$ is a disjoint union of components $C_{i}$ satisfying the conditions of Theorem 8, and
Scope
- Paper:
03 Published and Submitted Work/Published/P22_Litt_Zeta_Functions_Curves.pdf - Refine report:
.refine/results/Published/P22_Litt_Zeta_Functions_Curves.review.json - Rubric:
rubric/comment_triage_rubric.md - Version assessed: local published PDF, Michigan Mathematical Journal 64 (2015), 383-395
- Detailed Refine comments assessed: 7
- Assessment date: 2026-07-30
The local published PDF is authoritative. PDF pages 4-5, 7-10, and 12 were rendered and visually inspected to verify the projectivization notation, Čech-descent terminology, displayed formulas, and the hypotheses of Corollary
- The unanchored
feedback.overallprose was used as context but was not converted into additional assessment rows.
Summary
| # | Short title | Validity | Category | Standardness | Impact | Challenge | Repair | Disposition | Priority | Confidence |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | “Rational” should be “polynomial” | V4 | C1 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 2 | Missing stable-range qualifier | V4 | C5 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 3 | Čech cochain called a cocycle | V4 | C1 | E-NA | I1 | Q0 | R1 | D1 | P3 | HIGH |
| 4 | Projectivization needs local freeness | V4 | C5 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 5 | GL lifts are cochains, not cocycles | V4 | C1 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 6 | Theorem 24 cites the wrong bundle map | V4 | C1 | E-NA | I1 | Q1 | R1 | D1 | P3 | HIGH |
| 7 | Corollary 30 lacks a smoothness hypothesis | V4 | C5 | E-NA | I2 | Q2 | R2 | D3 | P2 | HIGH |
There are five local errata candidates and two optional precision edits. No issue remains at I3 or higher. In particular, Comment 7 changes the stated scope of Corollary 30 over imperfect fields but does not affect Theorem 8 or the paper's main result.
1. Remark 12 says “rational” where its logic requires “polynomial”
Comment ID: 4d42f684-efc7-48ea-a13e-4932ba2ad9ae Location: PDF p. 4, Remark 12.
The preceding sentence reports a claim that
is a polynomial. For the real conic in Example 11, the asserted minimum is \(n=1\). The next sentence therefore has to say that \((1-\mathbb Lt)(1-t)Z_X(t)\) is polynomial, not merely rational. Theorem 8 already proves rationality, so the following sentence's skepticism can only concern polynomiality.
- Classification:
V4/C1/I2 - Repair: replace “is rational” by “is a polynomial”
- Dependency trace: the sentence discusses the strength of Kapranov's remark; neither Theorem 8 nor any later proof uses it
- Disposition:
R2/D3;P2/HIGH
2. The Severi-Brauer description needs the stable range
Comment ID: 88acb30c-41c9-490a-bd91-2ab5c5595dff Location: PDF p. 4, paragraph following Remark 12; compare Section 5 on PDF p. 10.
The Abel-Jacobi morphism \(\operatorname{Sym}^n(C)\to\operatorname{Pic}^n(C)\) is a projective-space bundle after a splitting field only when \(n>2g-2\). Outside that range it can fail to be surjective and its fibers need not have constant dimension, so it is not generally a Severi-Brauer scheme.
The proof of Theorem 8 is safe: Section 5 explicitly imposes \(n>2g-2\) before using this Severi-Brauer description.
- Classification:
V4/C5/I2 - Repair: begin the sentence “For \(n>2g-2\), after a finite extension ...” and restrict the resulting Severi-Brauer assertion to that range
- Dependency trace: all later uses already satisfy the missing range
- Disposition:
R2/D3;P2/HIGH
3. Remark 14 needs a Čech 1-cochain
Comment ID: bfa7fa1c-3ff4-4098-909d-a1f88f8bf748 Location: PDF p. 5, Remark 14.
If \(\lambda\) and \(\lambda'\) represent the same class, then \(\lambda^{-1}\lambda'\) is a coboundary. Thus one chooses a Čech 1-cochain \(\beta\) satisfying
A 1-cocycle has trivial coboundary, so the printed noun is wrong. The equation itself and the functor \((\mathcal E,\phi)\mapsto(\mathcal E,\beta\phi)\) unambiguously give the correct construction.
- Classification:
V4/C1/I1 - Repair: replace “1-cocycle \(\beta\)” by “1-cochain \(\beta\)”
- Dependency trace: no mathematical construction changes
- Disposition:
R1/D1;P3/HIGH
4. A general twisted quasi-coherent sheaf need not give a Severi-Brauer scheme
Comment ID: e619e3cf-f88e-49b7-beb6-dbd0825ec224 Location: PDF p. 7, paragraph following Corollary 22.
The paper defined an \(\alpha\)-twisted sheaf as a twisted quasi-coherent sheaf, reserving “twisted vector bundle” for a locally free object. For a general twisted quasi-coherent sheaf, \(\mathbb P(\mathcal E)\) need not be étale-locally a projective-space bundle; likewise \(\operatorname{End}(\mathcal E)\) need not be Azumaya. The construction is valid for an \(\alpha\)-twisted vector bundle of positive locally constant rank.
Every subsequent application in Corollary 23, Theorem 24, Proposition 28, and Section 5 uses twisted vector bundles, so this is a local scope error.
- Classification:
V4/C5/I2 - Repair: replace both occurrences of “\(\alpha\)-twisted sheaf” in this construction by “\(\alpha\)-twisted vector bundle of positive locally constant rank”
- Dependency trace: later hypotheses already supply local freeness
- Disposition:
R2/D3;P2/HIGH
5. The PGL cocycle lifts to a GL cochain
Comment ID: 1ba37ac3-97f8-4cb5-8309-1080a3494729 Location: PDF pp. 7-8, end of Section 3, and PDF p. 10, Section 5.
A \(\operatorname{PGL}_n\)-valued Čech 1-cocycle can be lifted locally to a \(\operatorname{GL}_n\)-valued 1-cochain. The scalar failure of that cochain to satisfy the cocycle condition is exactly the Čech 2-cocycle defining the Brauer twist. If it lifted to an actual GL cocycle, it would descend to an ordinary vector bundle and the corresponding Brauer class would split.
Both passages immediately use the lifted data as twisted descent data, so the intended construction is recoverable; the repeated word “cocycle” is nevertheless mathematically meaning-bearing.
- Classification:
V4/C1/I2 - Repair: in both passages, replace “lift ... to a 1-cocycle valued in \(\operatorname{GL}\)” by “choose local GL lifts, forming a 1-cochain whose scalar coboundary represents \(\alpha\)”
- Dependency trace: this wording supplies exactly the twisted vector bundle used later; it does not change Theorem 24 or the recurrence in Theorem 8
- Disposition:
R2/D3;P2/HIGH
6. Theorem 24's embedding comes from the split projection
Comment ID: 9df8f779-ce61-413a-ab5b-fe48eed5fccb Location: PDF p. 9, proof of Theorem 24.
Under the paper's quotient/hyperplane convention for \(\mathbb P(\mathcal E)\), the injection \(\mathcal E_1\hookrightarrow\mathcal E_2\) does not itself induce the displayed closed embedding. The proof has just split the short exact sequence. The resulting projection \(\mathcal E_2\twoheadrightarrow\mathcal E_1\) does induce \(\mathbb P(\mathcal E_1)\hookrightarrow\mathbb P(\mathcal E_2)\), after which the stated complement calculation is unchanged.
- Classification:
V4/C1/I1 - Standard-step check:
Q1; the chosen splitting explicitly supplies the contravariant bundle map and all ranks in Lemma 25 remain the same - Repair: replace “the morphism \(\mathcal E_1\to\mathcal E_2\) induces” by “the projection \(\mathcal E_2\twoheadrightarrow\mathcal E_1\) supplied by the chosen splitting induces”
- Dependency trace: the theorem's equality and all later applications are unchanged
- Disposition:
R1/D1;P3/HIGH
7. Corollary 30 does not obtain smoothness over every imperfect field
Comment ID: a4a64a0e-6345-40fc-a8b6-64154ee393b6 Location: PDF p. 12, Corollary 30 and its proof.
The normalization of a reduced curve is regular, but over an imperfect field a regular finite-type curve need not be smooth over the base. Moreover, geometric irreducibility alone is a topological condition and does not exclude nilpotents or inseparable singular behavior after base change. Therefore the normalized components need not satisfy Theorem 8's smoothness hypothesis.
Severity challenge and local repair
The exact missing implication is:
normalized component regular and geometrically irreducible \(\Longrightarrow\) smooth and geometrically connected over \(k\).
It fails because smoothness requires geometric regularity, not only regularity over \(k\) and geometric irreducibility. A complete bounded repair is to assume that each normalized component is geometrically regular (and geometrically irreducible), or more simply to assume that \(k\) is perfect in addition to the printed geometric-irreducibility hypothesis. Then each normalized component is smooth and geometrically connected, so Theorem 8 applies componentwise and the zero-dimensional scissor argument proceeds unchanged.
This repair does not alter Theorem 8, which already starts with a smooth projective geometrically connected curve. Corollary 30 is the only stated result whose scope changes, and no later theorem depends on it.
- Severity status:
Q2 - Classification:
V4/C5/I2 - Repair/disposition:
R2/D3 - Priority/confidence:
P2/HIGH
P23 SYMMETRIC POWERS DO NOT STABILIZE9 detailed comments · 7 numbered corrections 2 I17 I2
These errata refer to the version published in Proceedings of the American Mathematical Society 142 (2014), no. 12, 4079--4094, doi:10.1090/S0002-9939-2014-12155-1. Page references below are to that version.
Page 4083, Remark 8.
The degree components of the Picard functor of a smooth projective geometrically integral curve are representable even when the curve has no rational point. The obstruction to the projective-bundle calculation is instead the possible absence of a universal Poincar\'e line bundle. Replace the three sentences beginning “For curves with no rational point” and ending “if this issue can be rectified” by the following.
For curves with no rational point, this projective-bundle argument need not apply. The schemes $\operatorname{Pic}^n(X)$ are still representable, but a universal Poincar\'e line bundle on $X\times\operatorname{Pic}^n(X)$ need not exist. For $n>2g-2$, the Abel--Jacobi morphism
\[ \operatorname{Sym}^n(X)\longrightarrow\operatorname{Pic}^n(X) \]may therefore be a nontrivial Severi--Brauer scheme rather than the projectivization of a vector bundle, so the projective-bundle identity in $K_0(\operatorname{Var}_k)$ used in Kapranov's argument does not follow directly.
No theorem in the paper uses Remark 8, and Section 5 separately assumes that the curve has a rational point.
Page 4084, Theorem 9.
Bittner's presentation requires the ambient variety in each blow-up relation to be smooth. In the sentence following the displayed relation, replace “for $X$ proper” by “for $X$ smooth and proper,” so that the sentence reads:
for $X$ smooth and proper, $Y$ a smooth closed subvariety of $X$, and $E$ the exceptional divisor of the blowup $\operatorname{Bl}_Y(X)$.
This is the hypothesis in Bittner's cited Theorem 3.1. Every subsequent use of the presentation and of the duality map is on smooth proper varieties, so no later result changes.
Page 4084, Conjecture 14.
Denef and Loeser [2, Section 3.3] do not support the conjecture that $\mathbb L$ is not a zero divisor. They state that injectivity of
\[ K_0(\operatorname{Var}_k) \longrightarrow K_0(\operatorname{Var}_k)[\mathbb L^{-1}] \]is unknown and that their later discussion relies on the guess that this map is not injective. That guess would imply that a nonzero class is annihilated by a power of $\mathbb L$. Delete “[2, 3.3],” from the heading of Conjecture 14. The corrected heading and statement are:
Conjecture 14 (Cancellation of the Lefschetz motive [13, remarks after Assertion 1]). $\mathbb L$ is not a zero divisor in $K_0(\operatorname{Var}_k)$.
The other citations following the conjecture remain unchanged, as do all results stated conditionally on Conjecture 14.
Page 4087, final paragraph of the proof of Theorem 19.
A fiber of the restriction of $\pi_n$ to $\pi_m^{-1}(y)$ is contained in, but need not equal, a fiber of $\pi_n$ on $U$. Thus the printed argument gives a lower bound for the dimension of the image, not an equality. Replace the paragraph beginning “Choosing $x\in\operatorname{Sym}^n(X)$” through the end of the proof by the following.
Choose $x\in\operatorname{Sym}^n(X)$ in the image of $\pi_n$ and outside the subvariety $W$ supplied by Theorem 20. Choose $y$ as above, and put
\[ F=\pi_m^{-1}(y), \qquad Y=\overline{\pi_n(F)}\subset\operatorname{Sym}^n(X). \]Then $x\in Y\setminus W$. The variety $F$ is a dense open subset of $\mathbb A^{2n-2m+l}$, so $Y$ is unirational. Every fiber of $\pi_n|_F\colon F\to Y$ is contained in a fiber of $\pi_n\colon U\to\operatorname{Sym}^n(X)$ and hence has dimension at most $l$. The fiber-dimension theorem therefore gives
\[ \dim Y\geq \dim F-l=2n-2m. \]Since $Y\setminus W$ is a nonempty open subset of $Y$, it has the same dimension as $Y$; as $Y$ is unirational, its points correspond to rationally equivalent zero-cycles on $X$. If $n>2m$, then
\[ \dim(Y\setminus W)\geq2n-2m>n, \]contradicting Theorem 20.
The corrected lower bound is precisely what the contradiction requires. Theorem 19 and Corollaries 21 and 23 are unchanged.
Page 4088, Remarks 24 and 25.
The $p$-adic point-counting argument proves nonconvergence and hence the failure of False Claim 4. It does not prove the failure of False Claim 5: equality modulo $\mathbb L$ would imply only equality of point counts modulo $q$, which is compatible with failure of $p$-adic convergence. Replace Remark 24 by the following.
Remark 24. We sketch here a proof that False Claim 4 also fails for $k=\mathbb F_q$. Let
\[ \psi_q\colon K_0(\operatorname{Var}_k)\longrightarrow\mathbb Z, \qquad [X]\longmapsto \#X(\mathbb F_q), \]be the point-counting homomorphism. It extends to a continuous homomorphism $\widehat\psi_q\colon R\to\mathbb Z_p$, so it is enough to find an $X$ for which $\widehat\psi_q([\operatorname{Sym}^n(X)])$ does not converge in $\mathbb Z_p$.
This happens if the zeta function
\[ \zeta_X(t)= \sum_{n=0}^{\infty} \psi_q([\operatorname{Sym}^n(X)])t^n, \]which is rational by the Weil conjectures, has a pole at a unit $y\in\mathcal O_{\mathbb C_p}^{\times}$ with $y\neq1$. There are many such abelian surfaces, by Honda--Tate theory; more simply, the product of two ordinary elliptic curves suffices.
In Remark 25, replace the first sentence by:
More generally, if $X$ is a smooth projective variety over $k=\mathbb F_q$, with nonvanishing $h^0(\Omega_X^{2n})$ for some $n>0$, and the $2n$-th Newton polygon of the zeta function of $X$ equals its $2n$-th Hodge polygon (for example, if $X$ is an ordinary abelian variety), then the same reasoning shows that False Claim 4 is false.
Thus these remarks make no assertion about False Claim 5 or MSSP over finite fields. The characteristic-zero results, including Corollary 23, are unaffected.
Page 4089, coefficient computation preceding Lemma 27.
The displayed identities involving $\mathbb P^{g-2}$ and the use of Lemma 27 do not cover genus zero, while the genus-one identity requires a convention for $\mathbb P^{-1}$. After the sentence ending “let us compute its coefficients,” insert:
If $g=0$, then the rational point identifies $X$ with $\mathbb P^1$, and
\[ Z_X^{\mathrm{mot}}(t) =\frac{1}{(1-t)(1-\mathbb L t)}. \]Hence
\[ (1-t)(1-\mathbb L t)Z_X^{\mathrm{mot}}(t)=1, \]whose Newton polygon and Hodge polygon are both the trivial polygon. For the remainder of the argument through Corollary 33, assume $g\geq1$.
After the sentence “Here we take $[\operatorname{Sym}^n(X)]=0$ for $n<0$,” insert:
When $g=1$, we also use the convention $[\mathbb P^{-1}]=0$.
With these additions, the three displayed coefficient identities and Lemma 27 are used only for $g\geq1$; the direct computation supplies the genus-zero cases of Corollaries 30 and 33. Both corollaries remain valid for every genus.
Page 4092, Proposition 35.
The printed statement does not bind $n$ before its first occurrence and then uses the same symbol for the pluricanonical exponent and the symmetric-power degree. Replace Proposition 35 by:
Proposition 35. Let $n>0$ be an integer, and let $S$ be smooth and projective, with $\dim(S)>1$. Suppose that either
\[ \dim(S)\text{ is even and }h^0(S,\omega_S^n)\neq0, \qquad\text{or}\qquad h^0(S,\omega_S^{2n})\neq0. \]Then, for every integer $m\geq0$, if $\operatorname{Sym}^n(S)$ is stably birational to $\operatorname{Sym}^m(S)$, one has $m=n$.
Thus both plurigenus hypotheses refer to the fixed positive integer $n$, and the phrase “for some $n$” is deleted. In the following application, a nonzero pluricanonical section has nonzero positive powers, so the stated failures of False Claim 5 and the conditional failures of MSSP for surfaces are unchanged.
Report metadata
| Field | Value |
|---|---|
| Category | Published |
| Processing status | completed |
| Detailed comments | 9 |
| Domain | stem/mathematics |
| Completed | 2026-07-29T18:16:20.330131+00:00 |
| Refine document ID | 19a2f314-60ff-4664-8d9e-1f5b8ee3b6a3 |
Refine summary
This paper discusses the stabilization of symmetric products of smooth projective varieties in the Grothendieck ring of varieties. The author shows that for smooth projective surfaces with a non-zero geometric genus, these products do not stabilize, providing a conditional counterexample to a conjecture by Vakil and Wood.
Overall feedback
Constructing the constant-cycle subvariety in Theorem 19
The proof constructs $Y=\pi_n(\pi_m^{-1}(y))$ and states its dimension is exactly $2n-2m$. Readers will likely observe that since fibers of the restricted map $\pi_n|_{\pi_m^{-1}(y)}$ need not have dimension $l$, only the lower bound $\dim Y\ge 2n-2m$ follows immediately. Furthermore, $Y$ begins as a constructible image. Concluding that unirationality forces all its points to be rationally equivalent requires explicit navigation of the morphism from the open subset of affine space and the properness of $\operatorname{Sym}^n(X)$. The argument requires extracting a full-dimensional locally closed subset that expressly meets the complement of Mumford's exceptional divisor and verifying the rational equivalence of its cycles before invoking Theorem 20.
Bridging completion and discrete quotients in Corollary 23
To conditionally refute MSSP, the argument relies on a hypothetical convergence of $[\operatorname{Sym}^n(X)]/\mathbb{L}^{2n}$ in $\hat{K}$ to force eventual equality in the discrete quotient $F^0/F^{-1}$. At present, the exposition passages directly from nonstabilizing stable-birational classes to the failure of MSSP without laying down this topological bridge. Providing the direction of the inverse limit defining $\hat{K}$ and cementing why a convergent sequence in $F^0$ is eventually constant upon projection to $F^0/F^{-1}$ will fully prime the text for Proposition 15 or Corollary 17.
Definitional boundaries for the motivic order
In Corollary 30, $v_{\mathbb{L}}(x)$ is defined as the greatest integer $n$ such that $x\in(\mathbb{L}^n)$. Because Remark 3 deliberately allows for a nonzero element to reside within every such ideal, and zero coefficients create an identical problem, the stated valuation and lower convex hull may remain undefined within the precise nonseparated setting the document allows. Realigning the definition to utilize a supremum in the extended nonnegative integers, and delineating how coefficients of infinite order are treated, ensures Corollary 33 functions securely as a coefficient-divisibility theorem without relying on implicit valuation-domain conventions.
Finite-field arguments modulo $\mathbb{L}$ limits
The discussion spanning Remarks 24–25 notes that nonconvergence of point counts in $\mathbb{Z}_p$ disproves convergence in the $\mathbb{L}$-adic completion, thereby resolving False Claim 4. It does not, however, disprove eventual equality modulo $\mathbb{L}$, which yields only eventual congruence of point counts modulo $q$, rather than $p$-adic convergence. A pole at a unit $y\ne1$ is insufficient to compel this without controlling its reduction. As a result, the claimed failure of False Claim 5—and the consequential conditional MSSP deduction—requires an active nonstabilization-modulo-$q$ argument to remain viable.
Formulation and extensions in Proposition 35
Proposition 35 carries substantial mathematical extensions, but its current formulation introduces material ambiguities. The variable $n$ is assigned to both the pluricanonical exponent in $h^0(S,\omega_S^n)$ or $h^0(S,\omega_S^{2n})$ and the degree of $\operatorname{Sym}^n(S)$, leaving the quantifiers and underlying conclusion difficult to parse. Because this proposition relies on a personal communication while introducing counterexamples such as Enriques surfaces and surfaces with $p_g=0$, stating it formally with distinct variables and supplying a proof or precise citation will secure its role in revising the final heuristic.
Detailed comments
1. Abstract overstates the Newton-polygon result
- ID:
74d5355a-080f-4d1c-a890-a922fee0172f - Refine score:
0.29 - Original types: general
- Refine status: open
Comment
The abstract's Newton-polygon claim is imprecise: Corollary 33 proves the equality for the polynomial $(1-t)(1-\mathbb{L}t)Z_X^{\mathrm{mot}}(t)$, not directly for the full motivic zeta function $Z_X^{\mathrm{mot}}(t)$. Because the paper defines Newton polygons for polynomials and the removed factors encode nontrivial poles, the stated result should retain this normalization.
Quoted passage
Finally, we discuss conjectural Hodge-theoretic obstructions to the stabilization of symmetric products. We provide evidence for these obstructions by showing that the Newton polygon of the motivic zeta function associated to a curve equals the Hodge polygon of the curve.
2. Residue convention in the finite-field analogy
- ID:
ac8fabb9-6a54-4e6f-9fb7-6213c515aaa7 - Refine score:
0.23 - Original types: general
- Refine status: open
Comment
The displayed value is $\lim_{t\to 1}(1-t)\zeta_X(t)$, whereas the standard residue defined using the local parameter $t-1$ is its negative. Thus $\operatorname{res}_{t=1}$ is correct here only under the unstated convention that it means the coefficient of $(1-t)^{-1}$.
Quoted passage
If $X$ is a smooth proper curve, the limit on the left specializes under $\psi_{q}$ to the "analytic class number formula" for the zeta functions appearing in the Weil conjectures. We have that
$$ \operatorname{res}_{t=1} \zeta_{X}(t)=\frac{\# \operatorname{Jac}(X)\left(\mathbb{F}_{q}\right)}{1-q}, $$and likewise
$$ \left.(1-t) Z_{X}^{m o t}(t)\right|_{t=1}=\frac{[\operatorname{Jac}(X)]}{1-\mathbb{L}} . $$
3. Picard representability claim in Remark 8
- ID:
78381d51-e490-4856-b9a9-bf00dcbcc4c8 - Refine score:
0.32 - Original types: general
- Refine status: open
Comment
Remark 8 appears to conflate representability of the relative Picard functor with existence of a universal line bundle. For a smooth projective geometrically integral curve, the Picard scheme and its degree components are representable without a rational point; the relevant possible obstruction to the projective-bundle calculation is instead the absence of a universal Poincaré bundle, which can make the Abel family a twisted projective bundle rather than the projectivization of a vector bundle.
Quoted passage
Kapranov shows that this is true for curves with a rational point [8, (1.3.5)(a)], where the hypothesis of the existence of a rational point is left implicit. For curves with no rational point the argument does not work. The issue is that the usual Picard functor is not representable in this case, and so $\operatorname{Sym}^{n}(X)$ is not a projective space bundle over $\operatorname{Pic}^{n}(X)$, which is an obstruction to Kapranov's argument. It is unclear to the author if this issue can be rectified.
4. Smoothness hypothesis missing in Theorem 9
- ID:
17b0cb1e-715e-410c-9779-4e420f800fed - Refine score:
0.27 - Original types: general
- Refine status: open
Comment
Theorem 9 omits smoothness of the ambient variety $X$. In the stated presentation, $X$ must be smooth and proper, with $Y\subset X$ smooth and closed; otherwise $X$ and $\operatorname{Bl}_Y(X)$ need not belong to the declared smooth-proper generating set.
Quoted passage
Theorem 9 (Bittner [1, Theorem 3.1]). $K_{0}\left(\operatorname{Var}_{k}\right)$, for $k$ algebraically closed and of characteristic zero, is generated by the classes of smooth proper $k$-varieties, subject only to the following relations:
$$ \left[\mathrm{Bl}_{Y}(X)\right]-[E]=[X]-[Y] $$for $X$ proper, $Y$ a smooth closed subvariety of $X$, and $E$ the exceptional divisor of the blowup $\operatorname{Bl}_{Y}(X)$.
5. Denef–Loeser (2004) express the opposite expectation about the Lefschetz class
- ID:
39b41027-96ee-492c-9c8b-eab9e3e9f3cc - Refine score:
0.78 - Original types: external_references
- Refine status: open
Comment
Denef and Loeser do not conjecture in §3.3 that the Lefschetz class is not a zero divisor. They say that injectivity of the localization map obtained by inverting the Lefschetz class is unknown and add that their discussion relies on the guess that this map is not injective. Noninjectivity would imply that a nonzero element is annihilated by a power of the Lefschetz class and hence that the Lefschetz class is a zero divisor. The citation therefore points in the opposite direction from the attributed conjecture. See https://arxiv.org/pdf/math/0212202.
Quoted passage
Conjecture 14 (Cancellation of the Lefschetz motive [2, 3.3], [13, remarks after Assertion 1]). $\mathbb{L}$ is not a zero divisor in $K_{0}\left(\operatorname{Var}_{k}\right)$.
6. Fiber dimension in the proof of Theorem 19
- ID:
c0dc1a9a-bccc-4468-be5c-7af2a20ac618 - Refine score:
0.33 - Original types: general
- Refine status: open
Comment
The equality $\dim Y=2n-2m$ is not justified: the fibers of $\pi_n|_{\pi_m^{-1}(y)}$ are intersections with the $l$-dimensional fibers of $\pi_n$ and can have dimension smaller than $l$. The argument establishes the sufficient bound $\dim Y\geq 2n-2m$, so the contradiction and theorem remain valid.
Quoted passage
Choosing $x \in \operatorname{Sym}^{n}(X)$ lying away from the subvariety $W$ of $\operatorname{Sym}^{n}(X)$ coming from Theorem 20, we choose $y$ as above and let $Y=\pi_{n}\left(\pi_{m}^{-1}(y)\right)$. As $\pi_{m}^{-1}(y)$ is an open subset of affine space, $Y$ is unirational; furthermore $Y$ has dimension $2 n-2 m$, as the non-empty fibers of $\pi_{n}$ have dimension $l$. As $Y$ is unirational, points in it correspond to rationally equivalent 0-cycles.
7. Finite-field argument does not reach Claim 5
- ID:
a0a0169e-83a7-4d66-acdc-72b4f4b304c9 - Refine score:
0.53 - Original types: general
- Refine status: open
Comment
The point-counting argument in Remark 24 establishes failure of False Claim 4, but not False Claim 5. Stabilization modulo $\mathbb L$ implies only eventual constancy of the point counts modulo $q$, which is compatible with nonconvergence in $\mathbb Z_p$. Thus the stated pole argument does not by itself establish the finite-field failure of Claim 5 or the conditional MSSP consequence derived from it.
Quoted passage
Remark 24. We sketch here a proof that False Claims 4 and 5 also fail for $k=\mathbb{F}_{q}$ a finite field; this also falsifies MSSP conditional on resolution of singularities, weak factorization of birational maps, and either of Conjectures 13 or [14, via the methods of Corollary 23. Let $\psi_{q}: K_{0}\left(\operatorname{Var}_{k}\right) \rightarrow \mathbb{Z}$ be the homomorphism
$$ \psi_{q}:[X] \mapsto \# X\left(\mathbb{F}_{q}\right) . $$Then $\psi_{q}$ extends to a continuous homomorphism $\widehat{\psi_{q}}: R \rightarrow \mathbb{Z}_{p}$. So it suffices to find an $X$ such that $\widehat{\psi_{q}}\left(\left[\operatorname{Sym}^{n}(X)\right]\right)$ does not converge in $\mathbb{Z}_{p}$.
8. Genus-zero case is undefined in the coefficient argument
- ID:
4114c1f6-95ff-4d07-9624-75e763015603 - Refine score:
0.27 - Original types: general
- Refine status: open
Comment
The coefficient argument does not formally cover the allowed case $g=0$: the displayed identities and Lemma 27 use negative symmetric powers and negative-index projective spaces, which are not defined by the earlier coefficient convention. At $g=1$, the third identity likewise uses $\mathbb{P}^{-1}$ without stating the convention $[\mathbb{P}^{-1}]=0$. The genus-zero Newton/Hodge-polygon conclusion is true by a direct computation, but it is not established by this part of the proof as written.
Quoted passage
Also,
$$ \begin{gathered} {\left[\operatorname{Sym}^{2 g}(X)\right]=[\operatorname{Jac}(X)]\left[\mathbb{P}^{g}\right]} \\ {\left[\operatorname{Sym}^{2 g-1}(X)\right]=[\operatorname{Jac}(X)]\left[\mathbb{P}^{g-1}\right]} \\ {\left[\operatorname{Sym}^{2 g-2}(X)\right]=[\operatorname{Jac}(X)]\left[\mathbb{P}^{g-2}\right]+\mathbb{L}^{g-1}} \end{gathered} $$where the last equality follows from the fact that $\omega_{X}$ is the unique degree $2 g-2$ line bundle $\mathcal{L}$ with $h^{0}(X, \mathcal{L})=g$ (from Serre Duality and Riemann-Roch), and all other line bundles $\mathcal{L}^{\prime}$ of degree $2 g-2$ satisfy $h^{0}\left(X, \mathcal{L}^{\prime}\right)=g-1$.
9. Proposition 35 reuses its quantified exponent
- ID:
4599ee62-730b-4a52-849d-351553e46aa1 - Refine score:
0.21 - Original types: general
- Refine status: open
Comment
There seems to be an issue with the quantification of $n$ in Proposition 35: Suppose that either $\dim(S)$ is even and $h^{0}(S, \omega_{S}^{n})$ is non-zero, or that $h^{0}(S, \omega_{S}^{2 n})$ is non-zero for some $n$. The variable $n$ appears first without a clear quantifier, and then as an existential variable before being reused as a specific fixed degree in the conclusion. Clarifying the phrasing (e.g., 'Suppose $n$ is a positive integer such that...') would resolve this ambiguity, even though the intended mathematical conclusion regarding non-stabilization is unaffected since non-vanishing plurigenera inherently produce infinitely many such degrees.
Quoted passage
Proposition 35. Let $S$ be smooth and projective, with $\operatorname{dim}(S)>1$. Suppose that either $\operatorname{dim}(S)$ is even and $h^{0}\left(S, \omega_{S}^{n}\right)$ is non-zero, or that $h^{0}\left(S, \omega_{S}^{2 n}\right)$ is non-zero for some $n$. Then if $\operatorname{Sym}^{n}(S)$ is stably birational to $\operatorname{Sym}^{m}(S), m=n$.
Scope
- Paper:
03 Published and Submitted Work/Published/P23_Litt_Symmetric_Powers.pdf - Refine report:
.refine/results/Published/P23_Litt_Symmetric_Powers.review.json - Rubric:
rubric/comment_triage_rubric.md - Version assessed: local published PDF, Proceedings of the American Mathematical Society 142 (2014), 4079-4094
- Detailed Refine comments assessed: 9
- Assessment date: 2026-07-30
The local published PDF is authoritative. PDF pages 1, 5, 6, 9-11, and 14 were rendered and visually inspected. The cited wording, formulas, hypotheses, and quantifiers are present in the published typesetting.
Summary
| # | Short title | Validity | Category | Standardness | Impact | Challenge | Repair | Disposition | Priority | Confidence |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Newton-polygon claim in the abstract | V4 | C9 | E-NA | I1 | Q0 | R1 | D1 | P3 | HIGH |
| 2 | Residue sign convention | V4 | C4 | E-NA | I1 | Q0 | R1 | D1 | P3 | HIGH |
| 3 | Picard representability in Remark 8 | V4 | C6 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 4 | Smooth ambient variety in Theorem 9 | V4 | C5 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 5 | Denef-Loeser attribution | V4 | C7 | E-NA | I2 | Q0 | R1 | D3 | P2 | HIGH |
| 6 | Dimension of the image in Theorem 19 | V4 | C6 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 7 | Finite-field Claim 5 overreach | V4 | C6 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 8 | Genus-zero and genus-one edge cases | V3 | C5 | E-NA | I2 | Q0 | R2 | D3 | P2 | HIGH |
| 9 | Reused exponent in Proposition 35 | V4 | C5 | E-NA | I2 | Q0 | R2 | D3 | P2 | MEDIUM |
None of the nine findings is I3 or higher. Comments 6 and 8 require actual mathematical corrections, but the repairs are bounded and leave every advertised main theorem unchanged. Comment 7 removes unsupported finite-field extensions from a remark rather than changing the paper's characteristic-zero results.
1. The abstract suppresses the polynomial normalization
Comment ID: 74d5355a-080f-4d1c-a890-a922fee0172f Location: PDF p. 1, abstract; compare Corollary 33 on PDF p. 13.
The abstract says that the Newton polygon of the motivic zeta function equals the Hodge polygon. Section 5 defines the \(L\)-adic Newton polygon only for a polynomial, and Corollary 33 proves the equality for
the degree-\(2g\) polynomial obtained after removing the two standard denominator factors. The abstract is understandable as shorthand, but it is strictly broader than the proved and defined statement.
- Classification:
V4/C9withabstract_overstatement;normalization/E-NA/I1 - Severity challenge:
Q0; this is claim calibration, not a proof defect - Repair: insert the normalized polynomial in the abstract
- Dependency trace: Corollary 33 and all later reasoning already use the normalized polynomial, so no mathematical result changes
- Disposition:
R1/D1;P3/HIGH
2. Equation (2) uses the opposite local parameter from the standard residue
Comment ID: ac8fabb9-6a54-4e6f-9fb7-6213c515aaa7 Location: PDF p. 5 (printed p. 4083), equation (2).
For a smooth proper curve over \(\mathbb F_q\),
The coefficient of \((t-1)^{-1}\), however, is the negative of this number. Thus the displayed value is correct for a coefficient of \((1-t)^{-1}\), but not for the standard \(t-1\) residue convention. The motivic comparison in equation (3) makes the intended normalization clear.
- Classification:
V4/C4withresidue_convention;sign_convention/E-NA/I1 - Severity challenge:
Q0; no mathematical dependency uses the sign - Repair: replace \(\operatorname{res}_{t=1}\) by \(\lim_{t\to1}(1-t)\), or explicitly declare the \((1-t)^{-1}\) convention
- Dependency trace: the analytic analogy and motivic formula are unchanged
- Disposition:
R1/D1;P3/HIGH
3. Remark 8 misidentifies the obstruction for curves without rational points
Comment ID: 78381d51-e490-4856-b9a9-bf00dcbcc4c8 Location: PDF p. 5 (printed p. 4083), Remark 8.
For a smooth projective geometrically integral curve, the relative Picard functor and its degree components are representable even without a rational point. What may fail is the existence of a universal Poincare line bundle on \(X\times\operatorname{Pic}^n(X)\). Consequently, the Abel family can be a Brauer-twisted projective-space fibration rather than the projectivization of a vector bundle, and the simple Grothendieck-ring calculation need not follow. The remark's caution about Kapranov's argument can remain, but its stated cause is false.
- Classification:
V4/C6withpicard_scheme;universal_bundle;brauer_obstruction/E-NA/I2 - Severity challenge:
Q0; the defect is confined to a contextual remark - Repair: replace the nonrepresentability assertion by the universal-bundle obstruction and describe the Abel family as potentially twisted
- Dependency trace: no theorem in the paper uses Remark 8; the rational-point hypothesis in Section 5 remains explicit
- Disposition:
R2/D3;P2/HIGH
4. The blow-up relations in Theorem 9 require \(X\) to be smooth
Comment ID: 17b0cb1e-715e-410c-9779-4e420f800fed Location: PDF p. 6 (printed p. 4084), Theorem 9.
Bittner's presentation uses classes of smooth proper varieties and blow-up relations with \(X\) smooth proper and \(Y\subset X\) smooth closed. The printed version says only that \(X\) is proper. With singular \(X\), neither \(X\) nor its blow-up is necessarily in the stated smooth-proper generating family.
- Classification:
V4/C5withmissing_hypothesis;citation_statement/E-NA/I2 - Severity challenge:
Q0; inserting one hypothesis restores the cited theorem - Repair: replace "for \(X\) proper" by "for \(X\) smooth and proper"
- Dependency trace: all subsequent uses of the presentation and duality map are on smooth proper varieties, so the correction changes no downstream result
- Disposition:
R2/D3;P2/HIGH
5. Denef-Loeser express the opposite expectation about \(\mathbb L\)
Comment ID: 39b41027-96ee-492c-9c8b-eab9e3e9f3cc Location: PDF p. 6 (printed p. 4084), Conjecture 14.
Denef-Loeser, Section 3.3 says that injectivity of
is unknown, and that their later discussion relies on the guess that the map is not injective. Noninjectivity means that a nonzero class is killed by some power of \(\mathbb L\), hence that \(\mathbb L\) is a zero divisor. Their expectation therefore points opposite to Conjecture 14 as attributed. The conjecture may still be presented as a standard open question using the paper's other sources.
- Classification:
V4/C7withincorrect_attribution;external_reference_check/E-NA/I2 - Severity challenge:
Q0; the conjecture is used conditionally, and only the attribution is wrong - Repair: remove Denef-Loeser from the supporting citations or cite them as a contrasting expectation
- Dependency trace: Propositions 16 and Corollary 23 remain explicitly conditional on Conjecture 14 and are unaffected
- Disposition:
R1/D3;P2/HIGH
6. The restricted projection gives a dimension lower bound, not equality
Comment ID: c0dc1a9a-bccc-4468-be5c-7af2a20ac618 Location: PDF p. 9 (printed p. 4087), proof of Theorem 19.
Let \(F=\pi_m^{-1}(y)\). Then \(\dim F=2n-2m+l\). Although every nonempty fiber of \(\pi_n:U\to\operatorname{Sym}^n(X)\) has dimension \(l\), a fiber of the restriction \(F\to\pi_n(F)\) is an intersection with such a fiber and can have dimension less than \(l\). The fiber-dimension theorem gives
This lower bound is exactly what the contradiction needs: when \(n>2m\), the unirational image outside \(W\) has dimension greater than \(n\).
- Classification:
V4/C6withfiber_dimension;inequality_direction/E-NA/I2 - Severity challenge:
Q0; the corrected inequality is immediate and stronger than the bound needed downstream - Repair: replace "\(\dim Y=2n-2m\)" by "\(\dim Y\ge2n-2m\)", taking the closure of the constructible image when naming \(Y\)
- Dependency trace: Theorem 19, Corollaries 21 and 23, and the main nonstabilization results remain unchanged
- Disposition:
R2/D3;P2/HIGH
7. Remark 24 proves nonconvergence but not failure modulo \(\mathbb L\)
Comment ID: a0a0169e-83a7-4d66-acdc-72b4f4b304c9 Location: PDF p. 10 (printed p. 4088), Remark 24.
A pole at a \(p\)-adic unit other than \(1\) can show that \(\#\operatorname{Sym}^n(X)(\mathbb F_q)\) does not converge in \(\mathbb Z_p\). By continuity of point counting, that disproves False Claim 4. False Claim 5 is different: eventual equality modulo \(\mathbb L\) implies only eventual constancy of these counts modulo \(q\). A sequence can be constant modulo \(q\) and still fail to converge \(p\)-adically. The pole argument given therefore does not prove False Claim 5, and the characteristic-zero route from Claim 5 to conditional failure of MSSP cannot simply be imported.
- Classification:
V4/C6withinsufficient_argument;finite_field;topology_vs_congruence/E-NA/I2 - Severity challenge:
Q0; the affected assertions occur only in Remarks 24-25 - Repair: restrict Remarks 24-25 to failure of False Claim 4, unless a separate invariant or congruence argument is supplied for Claim 5 and MSSP
- Dependency trace: all characteristic-zero theorems and Corollary 23 remain valid; only the asserted finite-field extensions of Claim 5 and MSSP are lost
- Disposition:
R2/D3;P2/HIGH
8. The coefficient proof needs separate low-genus conventions
Comment ID: 4114c1f6-95ff-4d07-9624-75e763015603 Location: PDF p. 11 (printed p. 4089), coefficient computation and Lemma 27.
The paper explicitly declares \([\operatorname{Sym}^n(X)]=0\) for \(n<0\), so the comment overstates that part. It is nevertheless correct that \(\mathbb P^{-1}\) and \(\mathbb P^{-2}\) are undefined and that Lemma 27 cannot compare \(\operatorname{Sym}^{-1}(X)\) with \(\operatorname{Sym}^0(X)\) when \(g=0\). For \(g=1\), the displayed third identity works after declaring \([\mathbb P^{-1}]=0\). For \(g=0\), \(X\simeq\mathbb P^1\) and directly
so the Newton and Hodge polygons are both the trivial polygon.
- Classification:
V3/C5withedge_case;undefined_negative_index;comment_overstatement/E-NA/I2 - Severity challenge:
Q0; the direct \(g=0\) calculation and the \([\mathbb P^{-1}]=0\) convention completely repair the low-genus cases - Repair: handle \(g=0\) before the coefficient argument, state \([\mathbb P^{-1}]=0\) for \(g=1\), and run Lemma 27 only for \(g\ge1\)
- Dependency trace: Corollaries 30 and 33 remain true for every genus
- Disposition:
R2/D3;P2/HIGH
9. Proposition 35 binds \(n\) ambiguously
Comment ID: 4599ee62-730b-4a52-849d-351553e46aa1 Location: PDF p. 14 (printed p. 4092), Proposition 35.
The exponent \(n\) first appears in \(h^0(S,\omega_S^n)\), is explicitly quantified only after the second disjunct in "for some \(n\)", and is then reused as the symmetric-power degree in the conclusion. The likely intended statement fixes a positive integer \(n\) at the outset and assumes the relevant plurigenus is nonzero for that same \(n\), but the exact scope should be checked against the result communicated by Melanie Wood.
- Classification:
V4/C5withambiguous_quantifier;variable_capture/E-NA/I2 - Severity challenge:
Q0; the issue is the formal scope of one proposition, not a defect in the paper's proved results - Repair: begin "Let \(n>0\), and suppose that either ..." and remove "for some \(n\)", if that matches the communicated theorem
- Dependency trace: the following paragraph uses only the existence of infinitely many nonzero plurigenus degrees; a nonzero pluricanonical section yields nonzero sections in all positive multiples
- Disposition:
R2/D3;P2/MEDIUM - Residual check: the author or Melanie Wood should confirm the intended quantifier before an erratum gives exact replacement wording
Action queue
- Add public errata entries for comments 3-9.
- Treat comments 1 and 2 as optional editorial clarifications.
- Confirm the exact intended quantifier in Proposition 35 before publishing its replacement statement.