REVIEW 5 major objections 5 minor 1 cited by
Higher-rank instantons sheaves on Fano threefolds
T0 review · 5 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash
Pith's one-line read The paper establishes the minimal charge $k_0$ for slope-stable instanton bundles on Fano threefolds and constructs such bundles at and above it.
desk verdict A real non-emptiness theorem for higher-rank instantons on Fano threefolds; the main ideas are sound and the missing-reference gap is fixable. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The engine is an inductive extension-and-deformation process. Start from an unobstructed $\mu$-stable $(n,k)$-instanton bundle $E$; because $\chi(E,F_0)<0$ for $k$ above the threshold, there is a nontrivial extension $0\to F_0\to E'\to E\to 0$ (with $F_0=\mathcal{O}$ for even index). Lemma 2.12 pins down the unique torsion-free slope-$\mu(E')$ quotient of $E'$, Lemma 2.13 transfers unobstructedness, and a dimension count inside the moduli space of Gieseker-semistable or simple sheaves shows that $E'$ deforms to a $\mu$-stable bundle with the same rank and charge. The base case is supplied by rank-2 't Hooft bundles, built via Serre correspondence from locally complete intersection curves: disjoint lines for $i_X=3,4$, families of unobstructed elliptic curves for $i_X=2$, and elliptic curves or rank-zero instanton transformations for $i_X=1$.
What would settle it
Exhibit a slope-stable $(n,k)$-instanton sheaf with $k<k_0^n$ on a smooth Fano threefold of Picard rank one with $i_X\ge 2$, or with $i_X=1$ and genus $g\ge 4$ outside the singular-quadric exception; the theorem asserts none exist. The companion check is to verify the base: construct or rule out an unobstructed slope-stable rank-2 instanton bundle in each charge class used in the induction.
Extended reading notes
Core claim
The central claim is Main Theorem 1: for a smooth Fano threefold $X$ of Picard rank one and index $i_X$, writing the Chern character of an $(n,k)$-instanton as $\gamma(n,k)=n\,\mathrm{ch}(F_0)-k\,\mathrm{ch}(\mathcal{O}_l(q_X-1))$, the minimal charge is $k_0^n=\lceil n/2\rceil$ for $i_X=4$, $k_0^n=\lceil n/3\rceil$ for $i_X=3$, $k_0^n=n$ for $i_X=2$, and for $i_X=1$ it is $k_0^1=0$, $k_0^n=1$ for odd genus, and $k_0^n=n$ for even genus. For all $n\ge 2-r_X$ and $k\ge k_0^n$ the paper constructs unobstructed $\mu$-stable instanton bundles, with generic splitting $0^n$ for even index and $(0^n,-1^n)$ for odd index, and proves that for $n\ge 2$ no slope-stable $(n,k)$-instanton sheaves exist below $k_0^n$, except possibly for index one with genus $\le 3$, or genus $4$ with $X$ lying in a singular quadric. The construction is inductive: rank-2 't Hooft bundles (obtained by Serre correspondence from disjoint lines or unobstructed elliptic curves) form the base, and each higher-rank instanton arises from a non-split extension by $\mathcal{O}$ (even index) or by the minimal instanton $F_0$ (odd index), which is then deformed to a $\mu$-stable bundle.
Load-bearing premise
The whole construction rests on the base case: for the needed charge ranges there must exist unobstructed slope-stable rank-2 instanton bundles, a fact imported from earlier work; if such a rank-2 bundle fails to exist on some boundary Fano threefold of index one, the higher-rank induction has no starting point.
Editorial extensions
If this is right
- For $\mathbb{P}^3$, the moduli space $\mathrm{MI}_{\mathbb{P}^3}(n,k)$ is nonempty for all $n\ge 2$ and $k\ge \lceil n/2\rceil$, with a generically smooth component of dimension $1-n^2+4nk$.
- For Del Pezzo threefolds of index 2, $\mathrm{MI}_X(n,k)$ is nonempty for $k\ge n$, with a generically smooth component of dimension $2kn-n^2+1$.
- For the quadric threefold, $\mathrm{MI}_X(n,k)$ is nonempty for $k\ge \lceil n/3\rceil$, with a generically smooth component of dimension $1-n^2+6nk$; for index-one threefolds the corresponding dimensions are $1-n^2+2nk$ (even genus) and $1+2nk$ (odd genus).
- A general member of the constructed component has stable restriction to a general anticanonical K3 surface, and the restriction map cuts out a Lagrangian subvariety of the K3 moduli space.
- For Fano threefolds with $H^3(X)=0$, every Gieseker-semistable $(n,k)$-instanton is the homology of a monad $F_{-1}^{\oplus k}\to W\otimes F_0\to F_1^{\oplus k}$, giving an explicit linear-algebra description.
Reading between the lines
- The index-one boundary cases with genus $\le 3$ or the singular-quadric genus 4 appear to be failures of the rank-2 base-case existence results, not of the inductive step; if a single unobstructed rank-2 instanton in each needed charge class were found there, the same construction should go through verbatim.
- The Lagrangian subvarieties in the K3 moduli spaces are natural places to transfer enumerative invariants from the instanton components, provided the restriction map is birational onto its image; this is a concrete property one could test next.
- The monadic description for $H^3(X)=0$ is likely to extend to every Fano threefold whose derived category has a full exceptional collection, with the middle term dimension $w$ read off the same cohomology table, so checking the next such threefolds would be a direct extension of Proposition 7.1.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper develops a systematic existence theory for higher-rank instanton sheaves on smooth Fano threefolds of Picard rank one. It defines (n,k)-instantons, proves lower bounds for the charge of slope-stable instantons, and constructs unobstructed μ-stable instanton bundles for all k above a rank-dependent threshold k_0^n, by induction from rank-two 't Hooft bundles. The cases i_X=4,3,2 and i_X=1 are treated separately, with the index-one case conditioned on genus and on very-ampleness of −K_X. The paper also proves stability of restrictions to general anticanonical K3 sections, discusses instantons on Fano threefolds with curvilinear Kuznetsov components, and gives a monadic description when H^3(X)=0.
Significance. If the gaps identified below are repaired, this would be a substantial contribution: it gives, for a large class of Fano threefolds, essentially sharp necessary and sufficient numerical conditions for the existence of slope-stable higher-rank instanton bundles, together with unobstructedness and generic splitting statements. The inductive construction from rank-two 't Hooft bundles is natural and, where carried out in detail, is convincing. The applications to K3 restrictions and to moduli spaces of sheaves on the anticanonical section are also valuable and connect with the symplectic geometry of moduli spaces.
major comments (5)
- [§4.1, Theorem 4.4, Step 3] The proof invokes the vanishing Ext^2(S,F_n^k)=0 by saying 'since Ext^2(S,E_n^k)=0, see the proof of Lemma ??', but no such lemma appears in the manuscript. This vanishing is load-bearing: it is exactly one of the two Ext^2 vanishings required to apply Lemma 2.13 and conclude that F_{n+1}^k is unobstructed. Moreover, passing from 0→O(−1)→E_n^k→F_n^k→0 to Ext^2(S,F_n^k)=0 also requires a vanishing of Ext^3(S,O(−1)), which is not stated. Without a correct proof of Ext^2(S,F_n^k)=0, the induction for the quadric threefold case of Main Theorem 1 is unsupported.
- [Main Theorem 1 vs. §4.2, Theorem 4.14] The first paragraph of Main Theorem 1 asserts existence of unobstructed μ-stable (n,k)-instanton bundles for all n≥2−r_X and k≥k_0^n with no exception for i_X=1. However, Theorem 4.14, which is the only source for the existence statement in the index-one case, is stated only for genus g≥4 and assumes that X is not contained in a singular quadric when g=4. In particular, for g=3 or for g=4 with X contained in a singular quadric, the theorem does not prove the asserted existence; for g=4 singular quadric and n=1, k_0^1=0, while Proposition 4.10 only gives existence for k≥1. Either the statement of Main Theorem 1 must be restricted, or the exceptional cases must be proved.
- [§3.1, Theorem 3.2, Step 5] The dimension computation intended to show that a general deformation of F_{n+1}^k has no global sections contains an unexplained and apparently incorrect equality: the text claims ext^1(F_{n+1}^k,F_{n+1}^k)−ext^1(F_n^k,F_n^k)−ext^1(F_n^k,O)−1 = χ(O,F_n^k) = χ(E_n^k) > 0. But for a (n,k)-instanton on P^3 one has χ(E_n^k)=n−2k, which is not positive when k≥⌈n/2⌉. The displayed equalities also appear to mix ext and χ terms without justification. This step is essential for the P^3 case of Main Theorem 1, so it must be corrected and rewritten.
- [§3.2, Theorem 3.8] The proof of Theorem 3.8 consists of the single sentence 'The steps of the proof are equivalent to the ones we used for Theorem 3.2.' Since Theorem 3.8 is load-bearing for the index-two case of Main Theorem 1, and since the base of the induction is different here (using the smoothing result Proposition 3.5 for elliptic curves), the proof needs to be written out, or at least the analogous Step 5 dimension count and the use of Lemma 2.13 must be spelled out.
- [§4.1, Theorem 4.4, Step 5] The inequality used to show that the component of M_Q(v) through [F_{n+1}^k] is not contained in Z_{v'} is justified by the formula ext^1(F_{n+1}^k,F_{n+1}^k)−ext^1(F_n^k,F_n^k)−ext^1(F_{2n}^k,S)+1 = 3k−n > 0. The object F_{2n}^k is not defined, and the equality is not derived. Since this computation is exactly what rules out the closed locus of sheaves admitting a quotient of type F_n^k, the argument is incomplete as written.
minor comments (5)
- [Abstract and Introduction] The abstract contains several typos ('intantons', 'such as and generic splitting'); these should be corrected.
- [§4.1, Theorem 4.4, Step 3] The reference 'Lemma ??' is a broken cross-reference and must be replaced by a precise statement and proof.
- [§4.2, Lemma 4.12(a)] The sentence 'allows to conclude that Ext^2(F0,E)' is missing the equality '=0' and should be completed.
- [§4.2, Theorem 4.14, Step 3] The condition 'v0/v1 = −1/2' should presumably be 'v1/v0 = −1/2', since slope is c1/rank.
- [§4.2, Proposition 4.10] The notation 'H 3+k, 1' for the Hilbert scheme of elliptic curves is not defined in the proof; it should be introduced explicitly.
Circularity Check
No significant circularity: the higher-rank induction is independent once the rank-2 base bundles are granted; the dangling 'Lemma ??' is an internal gap, not a circle.
full rationale
Score 0. The central derivation is not circular. The higher-rank existence theorems (3.2, 3.8, 4.4, 4.14) run by induction from rank-2 't Hooft bundles: the induction step produces a strictly mu-semistable extension, proves unobstructedness via Lemma 2.13, and deforms to a mu-stable bundle by dimension counts. None of these steps assumes the existence of the mu-stable (n+1,k)-bundle being proved. The lower-bound statements are obtained from Riemann-Roch and stability (Lemmas 2.10, 4.1, 4.12), not from the existence claim. The only notable dependence is the rank-2 base: Theorem 4.8 imports existence from [Fae14], [BF11] and [CFK24]. The first two are authored or co-authored by Faenzi, so there is self-citation, but [CFK24] is external and the rank-2 results serve as a base, not as a restatement of the higher-rank theorem; this is legitimate inductive support, not a circular reduction. I also flag a genuine internal gap that is not circular: in Theorem 4.4, Step 3, the vanishing Ext^2(S,F_n^k)=0 is justified only by 'since Ext^2(S,E_n^k)=0, see the proof of Lemma ??', a lemma that does not exist in the manuscript; the closest Lemma 4.1 concerns Gieseker-stable instantons, not the mu-stable inductively constructed ones, and the passage from 0 -> O(-1) -> E_n^k -> F_n^k -> 0 to Ext^2(S,F_n^k) additionally requires Ext^3(S,O(-1))=0, which is not recorded. This threatens the quadric induction in the range k >= ceil(n/3), but it is a missing proof or internal gap, not circularity: it does not identify the conclusion with an input, nor rename a fit as a prediction.
Assumptions & free parameters
assumptions (6)
- standard math A smooth Fano threefold of Picard rank one has index i_X in {1,2,3,4} with generator H_X (Kobayashi-Ochiai).
- standard math Maruyama boundedness, Hoppe's criterion, and moduli space facts from [HL10], including finiteness of destabilizing Hilbert polynomials and closedness of Quot loci.
- standard math Serre correspondence between rank 2 bundles and l.c.i. curves, with unobstructedness of the curve implying unobstructedness of the bundle.
- domain assumption Known classification and existence of rank 2 instanton bundles on prime Fano threefolds, given in Theorem 4.8 from [BF11], [CFK24], and [Fae14].
- domain assumption Smoothing theorem [HH85] for nodal elliptic curves and the existence of smooth unobstructed elliptic curves on Del Pezzo threefolds.
- domain assumption For i_X=1, the moduli space of minimal instantons is a fine moduli space, as in [BF11, Proposition 3.4].
Cite this review
Pith. "Pith review of Higher-rank instantons sheaves on Fano threefolds." pith.science (2026). https://pith.science/paper/HBCRCEUD
@misc{pith2026250413505,
author = {Pith},
title = {Pith review of: Higher-rank instantons sheaves on Fano threefolds},
year = {2026},
howpublished = {\url{https://pith.science/paper/HBCRCEUD}},
note = {Machine review of arXiv:2504.13505}
}
read the original abstract
We define instanton sheaves of higher rank on smooth Fano threefolds X of Picard rank one and show that their topological classification depends on two integers, namely the rank n (or the half of it, if the Fano index of X is odd) and the charge k. We elucidate the value of the minimal charge k0 of slope-stable n-instanton bundles (except for Fano threefolds of index 1 and genus 3 or 4), as an integer depending only on the genus of X and on n and we prove the existence of slope-stable n-instanton bundles of charge k greater than k0. Next, we study the acyclic extension of instantons on Fano threefolds with curvilinear Kuznetsov component and give a monadic description when the intermediate Jacobian is trivial. Finally, we provide several features of a general element in the main component of the moduli space of intantons, such as and generic splitting over rational curves contained in X and stable restriction to a K3 section S of X, and give applications to Lagrangian subvarieties of moduli spaces of sheaves on S.
Forward citations
Cited by 1 Pith paper
-
Ulrich bundles on double coverings of projective space
The general double cover of P^3 branched along a surface of degree 4, 6, or 8 carries a stable rank 2 Ulrich bundle, with moduli components of dimension 5, 6, and 0.
Reference graph
Works this paper leans on
- [1]
-
[2]
Michael F. Atiyah, Nigel J. Hitchin, Vladimir G. Drinfeld, and Yuri I. Manin, Construction of instantons, Phys. Lett. A 65 (1978), no. 3, 185--187
work page 1978
-
[3]
Allen B. Altman and Steven L. Kleiman, Compactifying the P icard scheme , Adv. Math. 35 (1980), no. 1, 50--112
work page 1980
-
[4]
Enrique Arrondo and Ignacio Sols, Classification of smooth congruences of low degree, J. Reine Angew. Math. 393 (1989), 199--219
work page 1989
-
[5]
Atiyah, Geometry of yang–mills fields, pp
Michael F. Atiyah, Geometry of yang–mills fields, pp. 216--221, Springer Berlin Heidelberg, Berlin, Heidelberg, 1978
work page 1978
-
[6]
Maria Chiara Brambilla and Daniele Faenzi, Moduli spaces of rank-2 ACM bundles on prime F ano threefolds , Michigan Math. J. 60 (2011), no. 1, 113--148
work page 2011
-
[7]
, Rank-two stable sheaves with odd determinant on F ano threefolds of genus nine , Math. Z. 275 (2013), no. 1-2, 185--210
work page 2013
-
[8]
, Vector bundles on F ano threefolds of genus 7 and B rill-- N oether loci , Internat. J. Math. 25 (2014), no. 3, 1450023
work page 2014
Show all 45 references
-
[9]
Arend Bayer, Alexander Kuznetsov, and Emanuele Macr \'i , Mukai bundles on F ano threefolds , arXiv e-print math.AG/2402.07154, 2024
2024
-
[10]
Ciro Ciliberto, Flaminio Flamini, and Andreas Leopold Knutsen, Elliptic curves, ACM bundles and U lrich bundles on prime F ano threefolds , Collect. Math. 75 (2024), no. 3, 795--822
2024
-
[11]
Gaia Comaschi and Marcos Jardim, Instanton sheaves on F ano threefolds , Manuscr. Math. 175 (2024), no. 1-2, 293--343
2024
-
[12]
Gaia Comaschi, Marcos Jardim, Cristian Martinez, and Dapeng Mu, Instanton sheaves: The next frontier, S \ a o Paulo J. Math. Sci. 18 (2024), no. 2, 639--675
2024
-
[13]
Algebra 33 (2005), no
Daniele Faenzi, Bundles over the F ano threefold V_5 , Comm. Algebra 33 (2005), no. 9, 3061--3080
2005
-
[14]
, Bundles over F ano threefolds of type V_ 22 , Ann. Mat. Pura Appl. (4) 186 (2007), no. 1, 1--24
2007
-
[15]
144 (2014), no
, Even and odd instanton bundles on F ano threefolds of P icard number one , Manuscripta Math. 144 (2014), no. 1-2, 199--239
2014
-
[16]
Daniele Faenzi and Alessandro Verra, Fano threefolds of genus 10, C oble cubics and their higher rank versions , Preprint in preparation, 2025
2025
-
[17]
Nagaraj, On prime F ano threefolds of genus 9 , Internat
Laurent Gruson, Fatima Laytimi, and Donihakkalu S. Nagaraj, On prime F ano threefolds of genus 9 , Internat. J. Math. 17 (2006), no. 3, 253--261
2006
-
[18]
Differential Geom
Lothar G \"o ttsche, Hiraku Nakajima, and K \=o ta Yoshioka, Instanton counting and D onaldson invariants , J. Differential Geom. 80 (2008), no. 3, 343--390
2008
-
[19]
52, Springer, 1977
Robin Hartshorne, Algebraic geometry, Graduate Texts in Mathematics, vol. 52, Springer, 1977
1977
-
[20]
, Stable reflexive sheaves, Math. Ann. 254 (1980), 121--176
1980
-
[21]
1124, Springer, Berlin, 1985, pp
Robin Hartshorne and Andr \'e Hirschowitz, Smoothing A lgebraic S pace C urves , Algebraic Geometry, Sitges (Barcelona), 1983, Lecture Notes in Mathematics, vol. 1124, Springer, Berlin, 1985, pp. 98--131
1983
-
[22]
Daniel Huybrechts and Manfred Lehn, The geometry of moduli spaces of sheaves, 2 ed., Cambridge Mathematical Library, Cambridge University Press, Cambridge, 2010
2010
-
[23]
Marcos Jardim, Instanton sheaves on complex projective spaces, Collect. Math. 57 (2006), no. 1, 69--91
2006
-
[24]
Prata, and Henrique N
Marcos Jardim, Gr \'e goire Menet, Daniela M. Prata, and Henrique N. S \'a Earp, Holomorphic bundles for higher dimensional gauge theory, Bull. Lond. Math. Soc. 49 (2017), no. 1, 117--132
2017
-
[25]
Marcos Jardim and Misha Verbitsky, Trihyperk\"ahler reduction and instanton bundles on C P ^3 , Compos. Math. 150 (2014), no. 11, 1836--1868
2014
-
[26]
Shoshichi Kobayashi and Takushiro Ochiai, Characterizations of complex projective spaces and hyperquadrics, J. Math. Kyoto Univ. 13 (1973), 31--47
1973
-
[27]
Siegmund Kosarew and Christian Okonek, Global moduli spaces and simple holomorphic bundles, Publ. Res. Inst. Math. Sci. 25 (1989), no. 1, 1--19
1989
-
[28]
Alexander Kuznetsov, Yuri Prokhorov, and Constantin Shramov, Hilbert schemes of lines and conics and automorphism groups of F ano threefolds , Japanese J. Math. 13 (2016), 109--185
2016
-
[29]
Kuznetsov, An exception set of vector bundles on the varieties V_ 22 , Vestnik Moskov
Alexander G. Kuznetsov, An exception set of vector bundles on the varieties V_ 22 , Vestnik Moskov. Univ. Ser. I Mat. Mekh. (1996), no. 3, 41--44, 92
1996
-
[30]
, Hyperplane sections and derived categories, Izv. Math. 70 (2006), no. 3, 447
2006
-
[31]
10 (2012), no
Alexander Kuznetsov, Instanton bundles on F ano threefolds , Open Math. 10 (2012), no. 4, 1198--1231
2012
-
[32]
Algebra 83 (1983), no
Herbert Lange, Universal families of extensions, J. Algebra 83 (1983), no. 1, 101--112
1983
-
[33]
38 (1979), no
Knud L nsted and Steven Lawrence Kleiman, Basics on families of H yperelliptic C urves , Compositio Math. 38 (1979), no. 1, 83--111
1979
-
[34]
Masaki Maruyama, Boundedness of semistable sheaves of small ranks, Nagoya Math. J. 78 (1980), 65--94
1980
-
[35]
Shigeru Mukai, Symplectic structure of the moduli space of sheaves on an abelian or K3 surface , Invent. Math. 77 (1984), 101--116
1984
-
[36]
I , Vector Bundles on Algebraic Varieties ( B ombay, 1984), Tata Inst
, On the moduli space of bundles on K3 surfaces. I , Vector Bundles on Algebraic Varieties ( B ombay, 1984), Tata Inst. Fund. Res. Stud. Math., vol. 11, Tata Inst. Fund. Res., Bombay, 1987, pp. 341--413
1984
-
[37]
357--377
, Curves, K3 surfaces and F ano 3-folds of genus 10 , Algebraic Geometry and Commutative Algebra (Hiroaki Hijikata, Heisuke Hironaka, Masaki Maruyama, Hideyuki Matsumura, Masayoshi Miyanishi, Tadao Oda, and Kenji Ueno, eds.), Academic Press, 1988, pp. 357--377
1988
-
[38]
Hiraku Nakajima and K\-ota Yoshioka, Instanton counting on blowup. I . 4-dimensional pure gauge theory , Invent. Math. 162 (2005), no. 2, 313--355
2005
-
[39]
Orlov, Exceptional set of vector bundles on the variety V_5 , Vestnik Moskov
Dmitri O. Orlov, Exceptional set of vector bundles on the variety V_5 , Vestnik Moskov. Univ. Ser. I Mat. Mekh. (1991), no. 5, 69--71. 1294662
1991
-
[40]
Giorgio Ottaviani, Spinor bundles on quadrics, Trans. Amer. Math. Soc. 307 (1988), no. 1, 301--316
1988
-
[41]
Prokhorov, Exotic F ano varieties , Vestnik Moskov
Yuri G. Prokhorov, Exotic F ano varieties , Vestnik Moskov. Univ. Ser. I Mat. Mekh. (1990), no. 3, 34--37, 111
1990
-
[42]
Parshin and Igor R
Aleksei N. Parshin and Igor R. Shafarevich (eds.), Algebraic geometry. V , Encyclopaedia of Mathematical Sciences, vol. 47, Springer-Verlag, Berlin, 1999
1999
-
[43]
Tikhomirov, Moduli of mathematical instanton vector bundles with odd c_2 on projective space , Izv
Alexander S. Tikhomirov, Moduli of mathematical instanton vector bundles with odd c_2 on projective space , Izv. Ross. Akad. Nauk Ser. Mat. 76 (2012), no. 5, 143--224
2012
-
[44]
, Moduli of mathematical instanton vector bundles with even c_2 on projective space , Izv. Ross. Akad. Nauk Ser. Mat. 77 (2013), no. 6, 139--168
2013
-
[45]
Tyurin, Fano versus C alabi- Y au , The F ano C onference, Univ
Andrei N. Tyurin, Fano versus C alabi- Y au , The F ano C onference, Univ. Torino, Turin, 2004, pp. 701--734
2004
Reviewed August 16, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.