Pith. sign in

REVIEW 2 major objections 5 minor 300 references

Regular proper quasi-F^∞-split varieties with trivial canonical bundle and no nontrivial global functions are not geometrically uniruled and have geometrically canonical singularities.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-01 10:18 UTC pith:IM2DSBWH

load-bearing objection A real advance in the quasi-F-split program with a coherent proof skeleton, but the central transfer lemma depends on the author's own unpublished preprints, so treat it as conditional rather than solid. the 2 major comments →

arxiv 2607.20272 v1 pith:IM2DSBWH submitted 2026-07-22 math.AG

On the global and local geometry of quasi-F-split varieties with trivial canonical bundle

Jefferson Baudin This is my paper
classification math.AG MSC 14G1714E3014J32
keywords quasi-F-splitquasi-F^∞-splitWitt vector cohomologytrivial canonical bundleunirulednesscanonical singularitieslog canonical singularitiespositive characteristic
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper studies schemes of finite type over a field that is finitely generated over a perfect field of characteristic p, whose canonical bundle is numerically trivial and which admit a quasi-F-split structure — a weakening of Frobenius splitting phrased through Witt vector cohomology. It proves that regular proper quasi-F^∞-split varieties with trivial canonical bundle, over a field with no global sections except constants, are not geometrically uniruled and have geometrically canonical singularities. It also produces, in characteristic 2, a smooth projective surface with trivial canonical bundle that is quasi-F-split but not quasi-F^∞-split, showing that the two notions genuinely differ for surfaces and answering a question raised in the literature. A further theorem states that normal quasi-F-split varieties with trivial canonical bundle are geometrically normal. Finally, under a resolution-of-singularities hypothesis on a prime-to-p index cover, quasi-F^e-pure normal varieties with m p^e K_X Cartier (m coprime to p) are log canonical.

Core claim

The central claim is that the quasi-F^∞-split condition transfers nonzero sections of the canonical sheaf from a variety to every generically finite cover (Proposition 3.8 in the paper), and a proper variety with a nonzero section of its canonical sheaf cannot admit a birational map onto a product with P^1. This yields the non-uniruledness and geometric canonicity of regular proper K-trivial quasi-F^∞-split varieties. The paper also constructs an explicit bielliptic surface in characteristic 2 — a diagonal quotient of an ordinary and a supersingular elliptic curve — that is quasi-F-split but not quasi-F^2-split, demonstrating that quasi-F-splitting alone does not imply quasi-F^∞-splitting. I

What carries the argument

The central object is the sheaf q^∞S^0(X,ω_X), defined as the intersection over e>0 and n>0 of the images of the e-th iterated Frobenius trace map on the n-truncated Witt-vector canonical sheaf W_nω_X; nonvanishing of this module is the quasi-F^∞-split condition. The proof's engine is a transfer lemma: for a generically finite separable morphism Y→X with X regular, there is a canonical injection q^eS^0_n(X,ω_X) ↪ q^eS^0_n(Y,ω_Y), obtained by identifying W_nω_X with the top piece of the de Rham-Witt complex and pulling back differential forms. This injection is what forces every cover to carry a nonzero canonical section, contradicting the possibility of a birational product with P^1, since P

Load-bearing premise

The central bet is that the Witt-vector canonical sheaf W_nω_X is compatibly identified with the top de Rham-Witt differentials, with matching Cartier and trace maps under generically finite separable pullback, a compatibility imported from a classical theorem and an earlier remark of the author rather than verified here; if that compatibility fails, the transfer lemma backing non-uniruledness collapses, and Theorem D also needs log resolutions whose existence is open in posi

What would settle it

Take a separable double cover Y→X of a smooth proper K-trivial surface over a non-perfect field, and compute the induced map q^1S^0_1(X,ω_X) → q^1S^0_1(Y,ω_Y) using the de Rham-Witt identification of the Witt canonical sheaf. If a nonzero element on X pulls back to zero on Y, the transfer lemma fails and Theorem A has no proof. Alternatively, a regular proper quasi-F^∞-split K-trivial variety that admits a dominant rational map from a rational variety would directly contradict the non-uniruledness conclusion.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Regular proper quasi-F^∞-split varieties with trivial canonical bundle are not geometrically uniruled and have geometrically canonical singularities.
  • The quasi-F-split world is strictly larger than the quasi-F^∞-split world for K-trivial surfaces: the characteristic-2 bielliptic quotient is quasi-F-split but not quasi-F^2-split; the paper conjectures that for each fixed dimension d there is a prime p_d such that all smooth proper quasi-F-split K-trivial varieties of dimension d over an algebraically closed field are quasi-F^∞-split when p > p_d
  • Normal quasi-F-split K-trivial varieties are geometrically normal; the proof works through a purely inseparable cover argument and does not need resolution of singularities.
  • If a prime-to-p index cover of a quasi-F^e-pure normal variety admits a log resolution, then the variety is log canonical whenever m p^e K_X is Cartier for m coprime to p.
  • The method yields as a byproduct that any quotient of a regular quasi-F-split K-trivial variety by a 1-foliation has a nonzero global canonical section (Remark 3.9 in the paper).

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the transfer lemma works for all F-finite fields once the relevant de Rham-Witt compatibility is established, the finitely-generated-over-perfect-field hypothesis (which the paper flags as unnatural) could be removed; the paper explicitly states this is the only reason for that assumption.
  • The p=2 exception involves a supersingular elliptic curve factor, suggesting a general principle: quasi-F-split K-trivial varieties that fail to be quasi-F^∞-split should contain a supersingular factor in some canonical decomposition. Testing this on K3 surfaces with finite Artin-Mazur height, which are quasi-F-split, would be a natural next step.
  • The crystalline criterion of Lemma 4.3 — if the top and top-plus-one crystalline cohomology groups are p-torsion-free then the variety is quasi-F^∞-split — could serve as a practical test for Calabi-Yau varieties in positive characteristic, where torsion-freeness of crystalline cohomology is often computable.
  • Since non-uniruledness now holds for quasi-F^∞-split K-trivial varieties without any resolution assumption, the remaining gap between quasi-F-split and quasi-F^∞-split may be measurable by Artin-Mazur height; a conjecture in the paper asks whether this gap disappears above a dimension-dependent prime.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

2 major / 5 minor

Summary. The paper studies the geometry and singularities of quasi-F-split varieties with trivial canonical bundle over a field K finitely generated over a perfect field k of characteristic p. Theorem A asserts that regular proper quasi-F^∞-split varieties X with K_X≡0 and H^0(X,O_X)=K are not geometrically uniruled and have geometrically canonical singularities. Theorem B constructs a smooth projective quasi-F-split surface with trivial canonical bundle that is not quasi-F^∞-split in characteristic 2, answering negatively a question of Kawakami–Takamatsu–Tanaka–Witaszek–Yobuko–Yoshikawa. Theorem C shows that normal quasi-F-split K-trivial varieties are geometrically normal. Theorem D proves that quasi-F^e-pure normal varieties with m p^e K_X Cartier (m coprime to p) are log canonical, assuming a resolution of singularities for a prime-to-p index cover. The main mechanism is to propagate nonvanishing of q^∞S^0(X,ω_X) through generically finite covers and use H^0(ω)≠0 to obstruct uniruledness.

Significance. If the main theorems are correct, the paper represents a substantial advance. Theorem A generalizes and simplifies earlier results of Patakfalvi–Zdanowicz, and the key idea of propagating q^∞S^0 through both inseparable and separable covers is elegant. Theorem B gives the first counterexample to a natural question about quasi-F-splitting versus quasi-F^∞-splitting. The paper is well-structured and the exposition is generally clear. However, two load-bearing technical points—the compatibility of Witt canonical sheaves with de Rham-Witt differentials in Lemma 3.6 and a coefficient estimate in Proposition 5.1—are not sufficiently justified in the manuscript. Several key results are cited from the author's own unpublished preprints, which makes verification difficult.

major comments (2)
  1. [Lemma 3.6 / Section 3] The proof of Lemma 3.6 asserts that a regular variety X over K can be spread out to a smooth scheme over the perfect field k with generic fiber X. This would imply X is smooth over K, which is not part of the assumptions and is generally false for regular varieties over imperfect fields. Moreover, the isomorphism W_nω_X ≅ W_nΩ^r_X and its compatibility with the inclusions W_{n-1}ω_X→W_nω_X and the Cartier operators C^e_n are cited to [Eke84, Thm 4.1] and the unpublished [Bau25, Rem. 2.2.2, Lemma 5.1.9] without verification. Since Lemma 3.6 is the bridge used in Proposition 3.8 to propagate q∞S^0 to every generically finite cover, Theorem A depends on this. The author should supply a proof or a published reference, or explicitly restrict to the smooth case and explain how the regular case follows.
  2. [Proposition 5.1 / Section 5] In the proof of Proposition 5.1, the statement 'a_i ∈ Q_{<1} for all i∈I by construction' is not justified. From K_Y+π^{-1}_*Δ+Σ a_iE_i+⌊E^-⌋ ∼_Q π^*(K_X+Δ), one obtains Σ a_iE_i = E^+ - (E^- - ⌊E^-⌋), so a_i = e^+_i - frac(e^-_i), which can exceed 1 when the positive discrepancies e^+_i are large. The subsequent proof of the claim relies on a_i<1 to ensure J≠∅. This vanishing statement is used to obtain Corollary 5.2 and ultimately Theorem 5.4, so this gap is load-bearing. The argument needs to be repaired, or the definitions of E^+ and E^- must be clarified.
minor comments (5)
  1. [Lemma 4.3] In the first paragraph of the proof, 'we only need to show that H^d(X,WOX)⊗Q=0' should read '≠0'.
  2. [Theorem 5.4] The statement begins 'Let X be an-quasi-F^e-pure' — this should be 'n-quasi-F^e-pure'.
  3. [Theorem 4.4 proof] The phrase 'a K3 surface of an abelian surface' should be 'a K3 surface or an abelian surface'.
  4. [Lemma 3.6] The symbol X is used both for the original variety over K and for the smooth model over k; this is confusing and should be changed (e.g., use X̃ for the model).
  5. [Proposition 3.8] The final step 'In particular, X is not geometrically uniruled' would benefit from a sentence explaining why H^0(P^1×W,ω)=0 contradicts q∞S^0(Y,ω_Y)≠0, since this is the core of the argument.

Circularity Check

1 steps flagged

Lemma 3.6 is the load-bearing self-cited bridge: the trace compatibility of W_nω_X ≅ W_nΩ^r_X is imported from the author's unpublished [Bau25], so Theorem A's proof leans on an unverified self-citation even though the derivation is not definitionally circular.

specific steps
  1. self citation load bearing [Section 3, Lemma 3.6 (proof); used in Proposition 3.8 and Theorem A]
    "we obtain by [Eke84, Theorem 4.1] that there is a natural isomorphism W_nΩ^r_X ∼= W_nω_X, and it is explained in [Bau25, Remark 2.2.2] why this isomorphism preserves all relevant trace maps (i.e. W_{n−1}ω_X ,→ W_nω_X and C_n^e : F^e_* W_nω_X → W_nω_X)... preserves the Cartier operator and the inclusions W_{n−1}ω ,→ W_nω by [Bau25, Lemma 5.1.9] and localizing."

    Lemma 3.6 is the bridge that lets Proposition 3.8 push q∞S0(X,ω_X) ≠ 0 to every generically finite separable cover, and Proposition 3.8 is the entire mechanism behind Theorem A. Its content is the compatibility of W_nω_X ≅ W_nΩ^r_X with the inclusions and Cartier maps C^e_n that define q^eS0_n in Definition 2.1/Lemma 3.2. That compatibility is not proved in this paper; it is delegated to the author's own unpublished preprint [Bau25] (Remark 2.2.2, Lemma 5.1.9), with no external or machine-checked verification. Remark 3.7 explicitly identifies this imported compatibility as the reason for the perfect-base-field hypothesis. The injection q∞S0(X,ω_X) → q∞S0(Y,ω_Y), and hence Theorem A, therefore rests on the self-cited [Bau25] rather than on an in-paper proof.

full rationale

The paper's global derivation is a genuine deduction, not a fit: q∞S0 is a defined cohomological invariant, and Theorem A's non-uniruledness follows by transferring it along generically finite covers and using H^0(P^1,ω) = 0. There is no fitted parameter, no prediction equal to an input, and no renaming of an empirical pattern. The only substantial circularity-related weakness is the load-bearing self-citation in Lemma 3.6: the compatibility of Ekedahl's isomorphism W_nω_X ≅ W_nΩ^r_X with trace/inclusion maps is imported from the author's own unpublished [Bau25]. Since Proposition 3.8 and Theorem A collapse if that compatibility is false or narrower than claimed, this is more than cosmetic. A second self-citation, [BKR25, Lemma 3.8] in Corollary 3.10, is similar but less severe because the same statement is essentially proved in this paper as Corollary 5.2. Remark 3.7 also flags the imported compatibility as the reason for the global hypothesis on K; Theorem D's reliance on log resolutions is an open-problem hypothesis, not a circularity. Thus the central claim still has independent mathematical content, and the score is 4 rather than higher.

Axiom & Free-Parameter Ledger

0 free parameters · 6 axioms · 0 invented entities

No fitted parameters and no new postulated objects. The 'prime-to-p index cover' and cyclic covers are standard constructions. The ledger is dominated by domain assumptions inherited from the quasi-F-split program and by the explicit resolution hypothesis.

axioms (6)
  • domain assumption Definition and descent properties of quasi-F-splittings under quasi-étale covers ([TWY24, Prop 3.24])
    Used in Corollary 3.5 and Theorem 5.4 to pass between X and its cyclic/prime-to-p cover.
  • domain assumption W_nω_X is isomorphic to W_nΩ^r_X with compatibility of the trace/Cartier operator ([Eke84, Thm 4.1]; [Bau25, Rem 2.2.2])
    Central to Lemma 3.6, which supplies the injection q∞S0(X,ω)→q∞S0(Y,ω) that powers Theorem A; the compatibility part is from the author's own prior work.
  • domain assumption Finite generation of H^d(X,W O_X) over W(k_0) for quasi-F-split X ([Yob23, Thm 2.7.(2)])
    Used in Lemma 4.1(c), Corollary 4.2 and Lemma 4.3 to control Hodge-Witt cohomology lengths.
  • domain assumption Decomposition H^d_dR(X/k) ≅ ⊕_{a+b=d} H^b(X,Ω^a_X) for quasi-F-split X ([Pet25, Thm 1.1])
    Used in Lemma 4.3 to identify H^d_dR with Hodge cohomology via the Mazur-Ogus property.
  • domain assumption Existence of log resolutions for the prime-to-p index cover in Theorem D and for the cyclic cover in Corollary 3.10
    Stated hypothesis; resolution of singularities is not known in general in positive characteristic, so these are conditional results.
  • domain assumption Completeness of the classification of smooth projective K-trivial surfaces in positive characteristic (implicit; relying on [Lan79], [Ill79], [Ser58])
    Theorem B enumerates K3, abelian, Enriques and bielliptic surfaces; the exhaustiveness of the list is imported from the literature.

pith-pipeline@v1.3.0-alltime-deepseek · 16185 in / 24405 out tokens · 204696 ms · 2026-08-01T10:18:44.381703+00:00 · methodology

0 comments
read the original abstract

We solve certain questions related to the geometry and singularities of quasi-$F$-split varieties with trivial canonical bundle. First, we prove that regular quasi-$F^{\infty}$-split varieties are not geometrically uniruled (this generalizes and significantly simplifies the earlier results of Patakfalvi and Zdanowicz) and have geometrically canonical singularities. Second, we show that there exist quasi-$F$-split surfaces with trivial canonical bundle which are not quasi-$F^{\infty}$-split, answering negatively a question raised by Kawakami, Takamatsu, Tanaka, Witaszek, Yobuko and Yoshikawa. Third, we show that normal quasi-$F$-split varieties with trivial canonical bundle are geometrically normal (this extends a result of Kawakami, Takamatsu and Yoshikawa), and finally we prove that quasi-$F^e$-pure normal varieties $X$ such that $mp^eK_X$ is Cartier for $m$ coprime to $p$ are log canonical, under a resolution of singularities hypothesis.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

300 extracted references · 96 canonical work pages

  1. [1]

    , TITLE =

    Grothendieck, A. , TITLE =. Inst. Hautes \'. 1961 , PAGES =

    1961

  2. [2]

    and Raynaud, M

    Illusie, L. and Raynaud, M. , TITLE =. Inst. Hautes \'. 1983 , PAGES =

    1983

  3. [3]

    , TITLE =

    Ekedahl, T. , TITLE =. 1986 , PAGES =

    1986

  4. [4]

    , TITLE =

    Grothendieck, A. , TITLE =. Inst. Hautes \'. 1960 , PAGES =

    1960

  5. [5]

    , TITLE =

    Grothendieck, A. , TITLE =. Inst. Hautes \'. 1965 , PAGES =

    1965

  6. [6]

    , TITLE =

    Grothendieck, A. , TITLE =. Inst. Hautes \'. 1966 , PAGES =

    1966

  7. [7]

    Fantechi, B. and G\". Fundamental algebraic geometry , SERIES =. 2005 , PAGES =. doi:10.1090/surv/123 , URL =

    doi:10.1090/surv/123 2005

  8. [8]

    1971 , PAGES =

    Th\'. 1971 , PAGES =

    1971

  9. [9]

    , TITLE =

    Bourbaki, N. , TITLE =. 2007 , PAGES =

    2007

  10. [10]

    Algebraic geometry

    G\". Algebraic geometry. [2020] 2020 , PAGES =. doi:10.1007/978-3-658-30733-2 , URL =

    doi:10.1007/978-3-658-30733-2 2020

  11. [11]

    Katz, N. M. , TITLE =. Modular functions of one variable,. 1973 , MRCLASS =

    1973

  12. [12]

    and Lazarsfeld, R

    Green, M. and Lazarsfeld, R. , TITLE =. Invent. Math. , FJOURNAL =. 1987 , NUMBER =. doi:10.1007/BF01388711 , URL =

    work page doi:10.1007/bf01388711 1987

  13. [13]

    and Lazarsfeld, R

    Green, M. and Lazarsfeld, R. , TITLE =. J. Amer. Math. Soc. , FJOURNAL =. 1991 , NUMBER =. doi:10.2307/2939255 , URL =

    work page doi:10.2307/2939255 1991

  14. [14]

    , TITLE =

    Lazarsfeld, R. , TITLE =. 2004 , PAGES =. doi:10.1007/978-3-642-18808-4 , URL =

    doi:10.1007/978-3-642-18808-4 2004

  15. [15]

    , TITLE =

    Matsumura, H. , TITLE =. 1989 , PAGES =

    1989

  16. [16]

    , TITLE =

    Hartshorne, R. , TITLE =. 1977 , PAGES =

    1977

  17. [17]

    , TITLE =

    Hartshorne, R. , TITLE =. 2010 , PAGES =. doi:10.1007/978-1-4419-1596-2 , URL =

    doi:10.1007/978-1-4419-1596-2 2010

  18. [18]

    , TITLE =

    Pries, R. , TITLE =. Computational arithmetic geometry , SERIES =. 2008 , MRCLASS =. doi:10.1090/conm/463/09051 , URL =

    work page doi:10.1090/conm/463/09051 2008

  19. [19]

    , TITLE =

    Simpson, C. , TITLE =. Ann. Sci. \'. 1993 , NUMBER =

    1993

  20. [20]

    and Popa, M

    Pareschi, G. and Popa, M. , TITLE =. Amer. J. Math. , FJOURNAL =. 2011 , NUMBER =. doi:10.1353/ajm.2011.0000 , URL =

    arXiv 2011

  21. [21]

    and Lazarsfeld, R

    Ein, L. and Lazarsfeld, R. , TITLE =. J. Amer. Math. Soc. , FJOURNAL =. 1997 , NUMBER =. doi:10.1090/S0894-0347-97-00223-3 , URL =

    work page doi:10.1090/s0894-0347-97-00223-3 1997

  22. [22]

    Hacon, C. D. and Kov\'. Classification of higher dimensional algebraic varieties , SERIES =. 2010 , PAGES =. doi:10.1007/978-3-0346-0290-7 , URL =

    doi:10.1007/978-3-0346-0290-7 2010

  23. [23]

    Rational curves on algebraic varieties , SERIES =

    Koll\'. Rational curves on algebraic varieties , SERIES =. 1996 , PAGES =. doi:10.1007/978-3-662-03276-3 , URL =

    doi:10.1007/978-3-662-03276-3 1996

  24. [24]

    Birational geometry of algebraic varieties , SERIES =

    Koll. Birational geometry of algebraic varieties , SERIES =. 1998 , PAGES =

    1998

  25. [25]

    , TITLE =

    Matsuki, K. , TITLE =. 2002 , PAGES =. doi:10.1007/978-1-4757-5602-9 , URL =

    doi:10.1007/978-1-4757-5602-9 2002

  26. [26]

    , TITLE =

    Debarre, O. , TITLE =. 2001 , PAGES =. doi:10.1007/978-1-4757-5406-3 , URL =

    doi:10.1007/978-1-4757-5406-3 2001

  27. [27]

    , TITLE =

    Fulton, W. , TITLE =. 1998 , PAGES =. doi:10.1007/978-1-4612-1700-8 , URL =

    doi:10.1007/978-1-4612-1700-8 1998

  28. [28]

    and Cascini, P

    Birkar, C. and Cascini, P. and Hacon, C. D. and McKernan, J. , TITLE =. J. Amer. Math. Soc. , FJOURNAL =. 2010 , NUMBER =. doi:10.1090/S0894-0347-09-00649-3 , URL =

    doi:10.1090/s0894-0347-09-00649-3 2010

  29. [29]

    , TITLE =

    Patakfalvi, Zs. , TITLE =. J. Pure Appl. Algebra , FJOURNAL =. 2015 , PAGES =

    2015

  30. [30]

    and Patakfalvi, Zs

    Codogni, G. and Patakfalvi, Zs. , TITLE =. Invent. Math. , FJOURNAL =. 2021 , NUMBER =. doi:10.1007/s00222-020-00999-y , URL =

    work page doi:10.1007/s00222-020-00999-y 2021

  31. [31]

    and Blickle, M

    Bhatt, B. and Blickle, M. and Lyubeznik, G. and Singh, A. and Zhang, W. , TITLE =. J. Reine Angew. Math. , FJOURNAL =. 2025 , PAGES =. doi:10.1515/crelle-2025-0028 , URL =

    work page doi:10.1515/crelle-2025-0028 2025

  32. [32]

    Petrov, A. , YEAR=. Decomposition of the de

  33. [33]

    and Takagi, S

    Sato, K. and Takagi, S. and Yoshikawa, S. , YEAR=. Quasi-

  34. [34]

    and Takagi, S

    Kawakami, T. and Takagi, S. and Yoshikawa, S. , YEAR=. Quasi-

  35. [35]

    , TITLE =

    Nakkajima, Y. , TITLE =. J. Reine Angew. Math. , FJOURNAL =. 2022 , PAGES =. doi:10.1515/crelle-2022-0010 , URL =

    work page doi:10.1515/crelle-2022-0010 2022

  36. [36]

    , TITLE =

    Serre, J.-P. , TITLE =. Amer. J. Math. , FJOURNAL =. 1958 , PAGES =. doi:10.2307/2372780 , URL =

    work page doi:10.2307/2372780 1958

  37. [37]

    , TITLE =

    Pareschi, G. , TITLE =. Current developments in algebraic geometry , SERIES =. 2012 , MRCLASS =

    2012

  38. [38]

    2024 , note =

    On vanishing of higher direct images of the structure sheaf , AUTHOR=. 2024 , note =

    2024

  39. [39]

    Ren, F. and R. Duality for. 2024 , note =

    2024

  40. [40]

    , TITLE =

    Kato, K. , TITLE =. Compositio Math. , FJOURNAL =. 1987 , NUMBER =

    1987

  41. [41]

    and Levine, M

    Geisser, T. and Levine, M. , TITLE =. Invent. Math. , FJOURNAL =. 2000 , NUMBER =. doi:10.1007/s002220050014 , URL =

    doi:10.1007/s002220050014 2000

  42. [42]

    , TITLE =

    Matsuue, H. , TITLE =. Math. Z. , FJOURNAL =. 2017 , NUMBER =. doi:10.1007/s00209-016-1755-1 , URL =

    work page doi:10.1007/s00209-016-1755-1 2017

  43. [43]

    and Bernasconi, F

    Arvidsson, E. and Bernasconi, F. and Lacini, J. , TITLE =. Compos. Math. , FJOURNAL =. 2022 , NUMBER =. doi:10.1112/S0010437X22007394 , URL =

    doi:10.1112/s0010437x22007394 2022

  44. [44]

    and Koll\'

    Bernasconi, F. and Koll\'. Vanishing theorems for three-folds in characteristic. Int. Math. Res. Not. IMRN , FJOURNAL =. 2023 , NUMBER =. doi:10.1093/imrn/rnab316 , URL =

    doi:10.1093/imrn/rnab316 2023

  45. [45]

    and Ho, W

    Bhatt, B. and Ho, W. and Patakfalvi, Zs. and Schnell, C. , TITLE =. Compos. Math. , FJOURNAL =. 2013 , NUMBER =. doi:10.1112/S0010437X13007288 , URL =

    doi:10.1112/s0010437x13007288 2013

  46. [46]

    2019 , note =

    Families of divisors , AUTHOR=. 2019 , note =

    2019

  47. [47]

    and Xu, C

    Patakfalvi, Zs. and Xu, C. , TITLE =. Algebr. Geom. , FJOURNAL =. 2017 , NUMBER =. doi:10.14231/AG-2017-002 , URL =

    work page doi:10.14231/ag-2017-002 2017

  48. [48]

    , TITLE =

    Koll\'ar, J. , TITLE =. Available at

  49. [49]

    Singularities of pairs , BOOKTITLE =

    Koll. Singularities of pairs , BOOKTITLE =. 1997 , MRCLASS =

    1997

  50. [50]

    Projectivity of the moduli space of stable log-varieties and subadditivity of log-

    Kov\'. Projectivity of the moduli space of stable log-varieties and subadditivity of log-. J. Amer. Math. Soc. , FJOURNAL =. 2017 , NUMBER =. doi:10.1090/jams/871 , URL =

    doi:10.1090/jams/871 2017

  51. [51]

    , TITLE =

    Tanaka, H. , TITLE =. Nagoya Math. J. , FJOURNAL =. 2014 , PAGES =. doi:10.1215/00277630-2801646 , URL =

    doi:10.1215/00277630-2801646 2014

  52. [52]

    , TITLE =

    Tanaka, H. , TITLE =. Ann. Inst. Fourier (Grenoble) , FJOURNAL =. 2018 , NUMBER =

    2018

  53. [53]

    , TITLE =

    Tanaka, H. , TITLE =. Ann. Sc. Norm. Super. Pisa Cl. Sci. (5) , FJOURNAL =. 2020 , NUMBER =

    2020

  54. [54]

    , TITLE =

    Birkar, C. , TITLE =. Ann. Sci. \'. 2016 , NUMBER =. doi:10.24033/asens.2279 , URL =

    doi:10.24033/asens.2279 2016

  55. [55]

    Hacon, C. D. and Witaszek, J. , TITLE =. Duke Math. J. , FJOURNAL =. 2022 , NUMBER =. doi:10.1215/00127094-2022-0024 , URL =

    work page doi:10.1215/00127094-2022-0024 2022

  56. [56]

    , TITLE =

    Witaszek, J. , TITLE =. Ann. of Math. (2) , FJOURNAL =. 2022 , NUMBER =. doi:10.4007/annals.2022.195.2.4 , URL =

    work page doi:10.4007/annals.2022.195.2.4 2022

  57. [57]

    and Tanaka, H

    Nakamura, Y. and Tanaka, H. , TITLE =. Compos. Math. , FJOURNAL =. 2020 , NUMBER =. doi:10.1112/s0010437x1900770x , URL =

    doi:10.1112/s0010437x1900770x 2020

  58. [58]

    Available at

    Birational geometry of log surfaces , AUTHOR =. Available at

  59. [59]

    and Zdanowicz, M

    Patakfalvi, Zs. and Zdanowicz, M. , TITLE =. Ann. Sci. \'. 2026 , NUMBER =

    2026

  60. [60]

    and Waldron, J

    Patakfalvi, Zs. and Waldron, J. , TITLE =. Amer. J. Math. , FJOURNAL =. 2022 , NUMBER =. doi:10.1353/ajm.2022.0009 , URL =

    arXiv 2022

  61. [61]

    Quotients by finite equivalence relations , BOOKTITLE =

    Koll\'. Quotients by finite equivalence relations , BOOKTITLE =. 2012 , MRCLASS =

    2012

  62. [62]

    , TITLE =

    Keel, S. , TITLE =. Ann. of Math. (2) , FJOURNAL =. 1999 , NUMBER =. doi:10.2307/121025 , URL =

    doi:10.2307/121025 1999

  63. [63]

    and Nakamura, Y

    Hashizume, K. and Nakamura, Y. and Tanaka, H. , TITLE =. Math. Res. Lett. , FJOURNAL =. 2020 , NUMBER =

    2020

  64. [64]

    , TITLE =

    Hartshorne, R. , TITLE =. Illinois J. Math. , FJOURNAL =. 2007 , NUMBER =

    2007

  65. [65]

    and Hacon, C

    Das, O. and Hacon, C. D. , TITLE =. Math. Z. , FJOURNAL =. 2016 , NUMBER =. doi:10.1007/s00209-016-1655-4 , URL =

    work page doi:10.1007/s00209-016-1655-4 2016

  66. [66]

    2017 , note =

    On the projectivity of the moduli space of stable surfaces in char >5 , AUTHOR =. 2017 , note =

    2017

  67. [67]

    , TITLE =

    Waldron, J. , TITLE =. Nagoya Math. J. , FJOURNAL =. 2018 , PAGES =. doi:10.1017/nmj.2017.2 , URL =

    work page doi:10.1017/nmj.2017.2 2018

  68. [68]

    , TITLE =

    Fujino, O. , TITLE =. Publ. Res. Inst. Math. Sci. , FJOURNAL =. 2012 , NUMBER =. doi:10.2977/PRIMS/71 , URL =

    doi:10.2977/prims/71 2012

  69. [69]

    Hacon, C. D. and McKernan, J. and Xu, C. , TITLE =. J. Eur. Math. Soc. (JEMS) , FJOURNAL =. 2018 , NUMBER =. doi:10.4171/JEMS/778 , URL =

    work page doi:10.4171/jems/778 2018

  70. [70]

    Hacon, C. D. and Xu, C. , TITLE =. Invent. Math. , FJOURNAL =. 2013 , NUMBER =. doi:10.1007/s00222-012-0409-0 , URL =

    doi:10.1007/s00222-012-0409-0 2013

  71. [71]

    Hacon, C. D. and Xu, C. , TITLE =. Minimal models and extremal rays (. 2016 , MRCLASS =. doi:10.2969/aspm/07010361 , URL =

    arXiv 2016

  72. [72]

    and Gongyo, Y

    Fujino, O. and Gongyo, Y. , TITLE =. Compos. Math. , FJOURNAL =. 2014 , NUMBER =. doi:10.1112/S0010437X13007495 , URL =

    work page doi:10.1112/s0010437x13007495 2014

  73. [73]

    , TITLE =

    Bernasconi, F. , TITLE =. Eur. J. Math. , FJOURNAL =. 2019 , NUMBER =. doi:10.1007/s40879-018-00310-7 , URL =

    work page doi:10.1007/s40879-018-00310-7 2019

  74. [74]

    and Tanaka, H

    Cascini, P. and Tanaka, H. , TITLE =. Amer. J. Math. , FJOURNAL =. 2019 , NUMBER =. doi:10.1353/ajm.2019.0025 , URL =

    arXiv 2019

  75. [75]

    Hacon, C. D. and Xu, C. , TITLE =. J. Amer. Math. Soc. , FJOURNAL =. 2015 , NUMBER =. doi:10.1090/S0894-0347-2014-00809-2 , URL =

    doi:10.1090/s0894-0347-2014-00809-2 2015

  76. [76]

    Hacon, C. D. and Witaszek, J. , TITLE =. Peking Math. J. , FJOURNAL =. 2022 , NUMBER =. doi:10.1007/s42543-021-00037-7 , URL =

    doi:10.1007/s42543-021-00037-7 2022

  77. [77]

    Two examples of surfaces with normal crossing singularities , JOURNAL =

    Koll\'. Two examples of surfaces with normal crossing singularities , JOURNAL =. 2011 , NUMBER =. doi:10.1007/s11425-010-4161-x , URL =

    work page doi:10.1007/s11425-010-4161-x 2011

  78. [78]

    , journal=

    Serre, J.-P. , journal=. Morphismes universels et vari

  79. [79]

    , TITLE =

    Tanaka, H. , TITLE =. Math. Z. , FJOURNAL =. 2020 , NUMBER =. doi:10.1007/s00209-019-02345-2 , URL =

    work page doi:10.1007/s00209-019-02345-2 2020

  80. [80]

    , TITLE =

    Tanaka, H. , TITLE =. Osaka J. Math. , FJOURNAL =. 2016 , NUMBER =

    2016

Showing first 80 references.