betaFeedback

WorksMathematics

Algebraicity of Weil classes on non-split abelian sixfolds of Weil type over Q(−3)\mathbb{Q}(\sqrt{-3}) with Hermitian determinant −2-2

Abstract

Let K=Q(−3)K=\mathbb{Q}(\sqrt{-3}). We give a proof that every Weil class is algebraic on every complex abelian sixfold of Weil type for KK whose Hermitian form has signature (3,3)(3,3) and determinant −2-2 modulo norms. These forms are not split, so neither Schoen's theorem on sixfolds nor the theorem on sixfolds claimed in a preprint of Markman, both of which concern split forms, applies. We model the sixfolds by the part of the Jacobian of a cover C→P1C\to\mathbb{P}^1, branched at thirteen points, on which its group, the complex reflection group G32G_{32}, acts through its reflection representation. A degeneration argument with a ball-arithmetic certificate shows that this family dominates the nine-dimensional Shimura variety. On every member, an integral relation among characters induced from linear characters of subgroups, realized by correspondences with coefficients in the group algebra, reduces the Weil classes to the determinants of eigenspaces of eleven cyclic subcovers. A theorem of Schoen makes these determinants algebraic for the three étale subcovers; for the eight ramified ones we prove it for all cyclic covers of curves of degree at least 33 with at least 33 branch points and fully paired branching data, through a cone over a Fermat variety and Shioda's results. Specialization and isogenies then reach every sixfold. The determinant of the Hermitian form and the relation among characters are exact computations.

Introduction

For a smooth complex projective variety SS, a class in H2p(S,Q)H^{2p}(S,\mathbb{Q}) is a Hodge class if its image in H2p(S,C)H^{2p}(S,\mathbb{C}) has type (p,p)(p,p), and it is algebraic if it is a rational linear combination of fundamental classes of subvarieties of codimension pp. Algebraic classes are Hodge classes, and the Hodge conjecture asserts the converse. It holds for p=1p=1 by the Lefschetz (1,1)(1,1) theorem (see e.g. [1]) and is open in general, already for abelian varieties.

Let AA be a complex abelian variety of dimension 2n2n and KK an imaginary quadratic field acting on AA up to isogeny so that both embeddings of KK occur with multiplicity nn on H1,0(A)H^{1,0}(A). Weil observed that the two-dimensional space HW(A)=⋀K2nH1(A,Q)⊂H2n(A,Q)H_W(A)=\bigwedge_K^{2n}H^1(A,\mathbb{Q})\subset H^{2n}(A,\mathbb{Q}) consists of Hodge classes, and that for a general such AA these classes do not lie in the subring generated by divisor classes [2]. These Weil classes are a standard test case of the Hodge conjecture for abelian varieties. A polarization compatible with KK makes H1(A,Q)H^1(A,\mathbb{Q}) a Hermitian space over KK of signature (n,n)(n,n), and by Landherr’s theorem [3] its isometry class is determined by the class of its determinant in Q×/Nm⁡K×\mathbb{Q}^{\times}/\operatorname{Nm}K^{\times}. The form is split (§ 2.1) exactly when this class is that of (−1)n(-1)^n.

For n=2n=2 the Weil classes are known to be algebraic for K=Q(−3)K=\mathbb{Q}(\sqrt{-3}) and every determinant by Schoen [4], and a preprint of Markman claims a proof for every KK (Corollary 1.6.1), obtained from its claim for sixfolds and Schoen’s degeneration argument. For n=3n=3, Schoen proved algebraicity on a family of sixfolds for K=Q(−3)K=\mathbb{Q}(\sqrt{-3}) [4], which Markman identifies as the split sixfolds [6]; his earlier preprint describes them as having trivial discriminant [5], and in either description their determinant is not −2-2. That preprint claims a proof for the split sixfolds for every KK [5, Theorem 1.5.1]. In March 2026, Mostad’s preprint described the question for the non-split sixfolds as “completely open” [7]. The theorem below was claimed first in a preprint of Severn [8], posted on 14 October 2026, unrefereed and declared to be primarily AI-generated, which uses the same family of covers by a different route (see What is new); in a later preprint the same author claims, through split abelian eightfolds of Weil type and products with elliptic curves with complex multiplication, the algebraicity of the Weil classes on every abelian sixfold of Weil type for Q(−3)\mathbb{Q}(\sqrt{-3}), of every determinant [9]. A preprint of Bouali claims a proof of the Hodge conjecture for all smooth projective varieties [10], and a preprint of OpenAI claims a proof of the Hodge conjecture for the self-powers of the Jacobian at the tensor-Hodge-generic points of families of abelian covers of curves [11 Theorem 1.1], using two companion preprints [12, 13]; we have not assessed these claims. In this note we give a proof of the following statement.

Theorem 1.1 (computer-assisted). Let K=Q(−3)K = \mathbb{Q}(\sqrt{-3}), let AA be a complex abelian variety of dimension 66, and let ι:K→End⁡(A)⊗Q\iota: K \to\operatorname{End}(A) \otimes\mathbb{Q} be a homomorphism. Suppose that AA has a polarization λ\lambda such that

(a) the Rosati involution of λ\lambda restricts to complex conjugation on ι(K)\iota(K),

(b) both eigenspaces of ι(−3)\iota(\sqrt{-3}) on H1,0(A)H^{1,0}(A) have dimension 33, and

(c) the Hermitian form hλh_\lambda of λ\lambda (§ 2.2) satisfies det⁡hλ≡−2(modNm⁡K×)\det h_\lambda\equiv-2 \pmod{\operatorname{Nm} K^\times}.

Then every class in HW⁡(A,ι)=⋀K6H1(A,Q)\operatorname{HW}(A,\iota) = \bigwedge\nolimits_K^6 H^1(A,\mathbb{Q}) is algebraic.

Remark 1.2 (scope). Since 22 is not a norm from KK, condition (c) excludes the split forms. No condition is imposed on the type of λ\lambda, on the order ι(K)∩End⁡(A)\iota(K) \cap\operatorname{End}(A) or on End⁡(A)⊗Q\operatorname{End}(A) \otimes\mathbb{Q}; in particular abelian varieties with complex multiplication are included. Condition (c) is a property of (A,ι,λ)(A,\iota,\lambda) and not of (A,ι)(A,\iota). If B1B_1 is an abelian threefold with an action ι1\iota_1 of KK of signature (2,1)(2,1) and a compatible polarization λ1\lambda_1 with Hermitian form h1h_1, then the Hermitian form of (B1,ιˉ1,cλ1)(B_1,\bar{\iota}_1,c\lambda_1) is −chˉ1-c\bar{h}_1; so A=B1×B1A = B_1 \times B_1 with the action (ι1,ιˉ1)(\iota_1,\bar{\iota}_1) and the polarization λ1×cλ1\lambda_1 \times c\lambda_1, c∈Z>0c \in\mathbb{Z}_{>0}, has determinant −c3(det⁡h1)2≡−c-c^3(\det h_1)^2 \equiv-c, which is split for c=1c = 1 and satisfies (c) for c=2c = 2. This is why the theorem asks for one polarization with (a)–(c).

Besides the arguments of this note and results cited where they are used, the main published inputs are Schoen’s theorem on Hodge classes of generalized Prym varieties [14 Theorem 2.0, Corollary 3.1], Shioda’s results on Fermat varieties [15], and the Lefschetz standard conjecture for abelian varieties [16, 17]. The proof is computer-assisted in two ways: exact computations in GAP [18] and Python, and a rank certificate in ball arithmetic [19], whose last step is the evaluation of a 9×99 \times9 determinant. Appendix A describes both.

Use of large language models and status of the proofs

This note was produced by large language models, instances of the Claude models of Anthropic running as autonomous agents, in October 2026. They found the construction, wrote the proofs and the programs of Appendix A, ran the computations and wrote the text. The arguments were checked only by other instances of the same models, acting as referees and recomputing the numbers with programs of their own; such checks may share the blind spots of the models that wrote the proofs. No human expert has checked the proofs. The steps we regard as least secure are the application of Theorem 4.1 to the large subcovers of Table 1, the rank certificate of § 6, and the use of the character table and subgroup data of G32G_{32} computed by GAP.

jjcjc_j∣Hj∣|H_j|mmrrg(Yj)g(Y_j)ddg(Xj)g(X_j)m′m'exponentsinput
1−1-143265449892520701562:270, 3:4, 4:270Thm. 4.1
2+1+148663849132208624132:192, 4:192Thm. 4.1
3−1-1324644813613168904962:216, 3:16, 4:216Thm. 4.1
4+1+128812654153737262104364:324, 6:6, 8:324Thm. 4.1
5+1+12162402341468056161–étale[14]
6+1+11446336339771282104362:162, 3:12, 4:162Thm. 4.1
7−1-112024042138424101089–étale[14]
8+1+11081204681936056161–étale[14]
9+1+1388861607731277562:78, 3:4, 4:78Thm. 4.1
10−1-172126726793142568416964:324, 6:24, 8:324Thm. 4.1
11+1+1972664449960280962:24, 3:16, 4:24Thm. 4.1

Table 1. The eleven cyclic subcovers Xj→YjX_j \to Y_j of (5.3). Here ∣Hj∣|H_j| is the order of the subgroup, mm the order of ψj\psi_j, rr the number of branch points, d=dim⁡Wjd = \dim W_j, m′m' the degree of the Fermat variety in Theorem 4.1, and the exponents are listed as value : multiplicity in Z/m\mathbb{Z}/m. The last column names the result that makes Λj\Lambda_j algebraic (§ 5.2).

While this work was in progress, the models read the statements of the preprint [11] and the constructions used in its proofs, but not the proofs themselves. That suitably branched covers of P1\mathbb{P}^1 with a primitive complex reflection group can have isotypic parts with the dimension and signature of Weil sixfolds was observed before that reading. The plan of obtaining the Weil classes from the determinant lines of eigenspaces of cyclic subcovers, with [11 Lemma 3.4] as the intended input for these lines, combined by a contraction under an integrality condition on multiplicity vectors, was formulated during it. The choice of G32G_{32} and of the branching type, the relation of Proposition 5.2, the correspondence in C[G]\mathbb{C}[G], the dominance proof and the proof of Theorem 4.1 came afterwards; Theorem 4.1 was written from a summary of the statement of [11 Lemma 3.4] and of the construction in its proof (Remark 4.8). No result of [11, 12, 13] is used. The author and the models learned of the preprint [8] on 18 October 2026, after the proof and the first versions of this note were complete.

The note has not been refereed. It is made public together with this account of how it was produced and checked, and if an error is found, a correction will be posted.

Code availability

The programs for the computations of Appendix A, with the tuple of Appendix § 9 and the data of Table 1 and Table 2, are available under the MIT licence at https://github.com/ahippelainen/weil-nonsplit-abelian-sixfolds-public.

kkdim⁡ρk\dim\rho_kdim⁡Mρk\dim M_{\rho_k}χρk(γ1)\chi_{\rho_k}(\gamma_1)v1v_1v2v_2v3v_3v4v_4v5v_5v6v_6v7v_7v8v_8v9v_9v10v_{10}v11v_{11}
cjc_j−1-1+1+1−1-1+1+1+1+1+1+1−1-1+1+1+1+1−1-1+1+1
1462−ζ22-\zeta^200000000001
2442−1+2ζ2-1+2\zeta^200000000110
3201027+7ζ27+7\zeta^200001111020
4201021−5ζ21-5\zeta^200010011010
520138−5−8ζ2-5-8\zeta^220021011010
620174−5+ζ2-5+\zeta^201110111020
720174−8−5ζ2-8-5\zeta^220121111010
836162900101122021
9362709ζ29\zeta^201001222150
10603423−6ζ23-6\zeta^210112233041
11603783+3ζ23+3\zeta^201111333050
1260450−6+3ζ2-6+3\zeta^211112333061
13643848+8ζ28+8\zeta^201102334060
1480552−8−8ζ2-8-8\zeta^221232345060

Table 2. The fourteen irreducible representations ρk\rho_k of GG with central character χρ\chi_\rho, their degree, dim⁡Mρk\dim M_{\rho_k}, the character value at the reflection γ1\gamma_1 of order 3 and determinant ζ4\zeta^4, and the multiplicity vectors vjv_j of Table 1, as columns: the entry in row kk and column vjv_j is vj(ρk)v_j(\rho_k). With the signs cjc_j, the signed sum of the entries of row kk is 1 for k=1k=1 and 0 otherwise, which is (12), and dj=∑kvj(ρk)dim⁡Mρkd_j=\sum_k v_j(\rho_k)\dim M_{\rho_k}.

References

  1. [1]O. Benoist and J. C. Ottem, Failure of the integral Hodge conjecture for threefolds of Kodaira dimension zero, Comment. Math. Helv. 95 (2020), 27–35, https://doi.org/10.4171/CMH/479; arXiv:1802.01845.
  2. [2]A. Weil, Abelian varieties and the Hodge ring [1977c], in: Œuvres Scientifiques / Collected Papers, Vol. III, Springer, New York, 1979, https://www.semanticscholar.org/paper/Abelian-varieties-and-the-Hodge-ring-Weil/70c928f5164aa467d088aa6a3b4a1a5edbacd162.DOI
  3. [3]W. Landherr, Äquivalenz Hermitescher Formen über einem beliebigen algebraischen Zahlkörper, Abh. Math. Sem. Univ. Hamburg 11 (1936), 245–248, https://doi.org/10.1007/BF02940728.
  4. [4]C. Schoen, Addendum to: Hodge classes on self-products of a variety with an automorphism, Compositio Math. 114 (1998), 329–336, https://doi.org/10.1023/A:1000566205021.
  5. [5]E. Markman, Cycles on abelian 2n-folds of Weil type from secant sheaves on abelian n-folds, preprint (2025), arXiv:2502.03415.arxiv.org/abs/2502.03415
  6. [6]E. Markman, Secant sheaves and Weil classes on abelian varieties, in: Proceedings of the International Congress of Mathematicians 2026, Vol. 3, SIAM (2026), 586–602, https://doi.org/10.1137/25M1803796; arXiv:2509.23403.
  7. [7]A. Mostaed, McMullen’s curve, the Weil locus, and the Hodge conjecture for abelian sixfolds, preprint (2026), arXiv:2603.20268.arxiv.org/abs/2603.20268
  8. [8]B. Severn, Algebraicity of Weil classes on abelian sixfolds of Weil type for ℚ(√−3) with discriminant −2, preprint (2026), first posted 14 October 2026, Hexagon 2610.00246, https://hexagonmath.org/abs/2610.00246; also on Zenodo, https://doi.org/10.5281/zenodo.23377273.
  9. [9]B. Severn, Weil classes on abelian sixfolds and eightfolds of Weil type for ℚ(√−3), preprint (2026), Hexagon 2610.00387, https://hexagonmath.org/abs/2610.00387; also on Zenodo, https://doi.org/10.5281/zenodo.23406637.
  10. [10]J. Bouali, Degeneration of families of projective hypersurfaces and Hodge conjecture, preprint (2024), arXiv:2401.03465.arxiv.org/abs/2401.03465
  11. [11]OpenAI, Abelian covers, Gale correspondences, and the Hodge conjecture for powers, preprint dated 6 October 2026, file paper.pdf in the GitHub repository openai/math.hub.valency.io/works/W_6ftstxe9
  12. [12]OpenAI, The rational Hodge conjecture for CM abelian varieties, preprint dated 6 October 2026, file paper.pdf in the GitHub repository openai/math.hub.valency.io/works/W_3taj3hyy
  13. [13]OpenAI, Weil classes and Hodge classes on abelian powers, preprint dated 6 October 2026, file paper.pdf in the GitHub repository openai/math.hub.valency.io/works/W_54kbph3w
  14. [14]C. Schoen, Hodge classes on self-products of a variety with an automorphism, Compositio Math. 65 (1988), 3–32, https://www.numdam.org/item/CM_1988__65_1_3_0/.
  15. [15]T. Shioda, The Hodge conjecture for Fermat varieties, Math. Ann. 245 (1979), 175–184, https://doi.org/10.1007/BF01428804.
  16. [16]D. I. Lieberman, Numerical and homological equivalence of algebraic cycles on Hodge manifolds, Amer. J. Math. 90 (1968), 366–374, https://doi.org/10.2307/2373533.
  17. [17]S. L. Kleiman, Algebraic cycles and the Weil conjectures, in: Dix exposés sur la cohomologie des schémas, Adv. Stud. Pure Math. 3, North-Holland, Amsterdam, 1968, 359–386.
  18. [18]The GAP Group, GAP – Groups, Algorithms, and Programming, Version 4.16.1, 2026, https://www.gap-system.org.
  19. [19]F. Johansson, Arb: efficient arbitrary-precision midpoint-radius interval arithmetic, IEEE Trans. Comput. 66 (2017), 1281–1292, https://doi.org/10.1109/TC.2017.2690633; arXiv:1611.02831.
  20. [20]G. C. Shephard and J. A. Todd, Finite unitary reflection groups, Canad. J. Math. 6 (1954), 274–304, https://doi.org/10.4153/CJM-1954-028-3.
  21. [21]C. Schoen, Cyclic covers of ℙᵛ branched along ν + 2 hyperplanes and the generalized Hodge conjecture for certain abelian varieties, in: Arithmetic of Complex Manifolds (Erlangen, 1988), Lecture Notes in Math. 1399, Springer, Berlin, 1989, 137–154, https://doi.org/10.1007/BFb0095974.
  22. [22]E. Kani and M. Rosen, Idempotent relations and factors of Jacobians, Math. Ann. 284 (1989), 307–327, https://doi.org/10.1007/BF01442878.
  23. [23]B. van Geemen and A. Verra, Quaternionic Pryms and Hodge classes, Topology 42 (2003), 35–53, https://doi.org/10.1016/S0040-9383(02)00004-6; arXiv:math/0103111.
  24. [24]P. Deligne, Hodge cycles on abelian varieties (notes by J. S. Milne), in: P. Deligne, J. S. Milne, A. Ogus and K.-y. Shih, Hodge Cycles, Motives, and Shimura Varieties, Lecture Notes in Math. 900, Springer, Berlin, 1982, 9–100, https://doi.org/10.1007/978-3-540-38955-2_3.
  25. [25]E. Colombo, P. Frediani and A. Ghigi, On totally geodesic submanifolds in the Jacobian locus, Internat. J. Math. 26 (2015), no. 1, 1550005, https://doi.org/10.1142/S0129167X15500056; arXiv:1309.1022.
  26. [26]H. Grauert, Ein Theorem der analytischen Garbentheorie und die Modulräume komplexer Strukturen, Publ. Math. Inst. Hautes Étud. Sci. 5 (1960), 5–64, https://doi.org/10.1007/BF02684746.
  27. [27]A. Teleman, On the torsion of the first direct image of a locally free sheaf, Ann. Inst. Fourier (Grenoble) 65 (2015), no. 1, 101–136, https://doi.org/10.5802/aif.2926; arXiv:1309.0342.
  28. [28]S. A. Wolpert, Infinitesimal deformation of nodal stable curves, Adv. Math. 244 (2013), 413–440, https://doi.org/10.1016/j.aim.2013.05.008; arXiv:1204.3680.
  29. [29]I. Smith, Lefschetz fibrations and the Hodge bundle, Geom. Topol. 3 (1999), 211–233, https://doi.org/10.2140/gt.1999.3.211; arXiv:math/9907200.
  30. [30]I. Banerjee and N. Salter, Monodromy and vanishing cycles for sufficiently ample linear systems on simply connected surfaces, preprint (2025), arXiv:2512.04018.arxiv.org/abs/2512.04018
  31. [31]A. Lubotzky and C. Meiri, Sieve methods in group theory II: the mapping class group, Geom. Dedicata 159 (2012), 327–336, https://doi.org/10.1007/s10711-011-9662-4; arXiv:1104.2450.
  32. [32]M. Dettweiler and S. Wevers, Variation of local systems and parabolic cohomology, Israel J. Math. 156 (2006), 157–185, https://doi.org/10.1007/BF02773830; arXiv:math/0310139.
  33. [33]C. Chevalley and A. Weil, Über das Verhalten der Integrale 1. Gattung bei Automorphismen des Funktionenkörpers, Abh. Math. Sem. Univ. Hamburg 10 (1934), 358–361, https://doi.org/10.1007/BF02940687.
  34. [34]P. Frediani, A. Ghigi and M. Penegini, Shimura varieties in the Torelli locus via Galois coverings, Int. Math. Res. Not. IMRN 2015, no. 20, 10595–10623, https://doi.org/10.1093/imrn/rnu272; arXiv:1402.0973.
  35. [35]W. Fulton, Intersection Theory, 2nd ed., Ergeb. Math. Grenzgeb. (3) 2, Springer, New York, 1998, https://doi.org/10.1007/978-1-4612-1700-8.
  36. [36]J. Hornbostel, M. Wendt, H. Xie and M. Zibrowius, The real cycle class map, Ann. K-Theory 6 (2021), 239–317, https://doi.org/10.3998/akt.2021.6.239; arXiv:1911.04150.
  37. [37]D. Patel and Y. Zhang, Algebraicity of Hodge classes on some generalized Prym varieties, J. Algebra 712 (2027), 101–122, https://doi.org/10.1016/j.jalgebra.2026.07.007; arXiv:2506.13729.
  38. [38]R. Boltje and G. Navarro, Feit’s conjecture, the canonical Brauer induction formula, and Adams operations, preprint (2025), arXiv:2510.03179.arxiv.org/abs/2510.03179
  39. [39]J. Harris and D. Mumford, On the Kodaira dimension of the moduli space of curves, Invent. Math. 67 (1982), 23–86, https://doi.org/10.1007/BF01393371.
  40. [40]A. Borel, Some metric properties of arithmetic quotients of symmetric spaces and an extension theorem, J. Differential Geom. 6 (1972), 543–560, https://doi.org/10.4310/jdg/1214430642.
  41. [41]F. Charles and C. Schnell, Notes on absolute Hodge classes, in: Hodge Theory (E. Cattani, F. El Zein, P. A. Griffiths and Lê D. T., eds.), Math. Notes 49, Princeton Univ. Press, Princeton, NJ, 2014, 469–530, https://doi.org/10.1515/9781400851478.469; arXiv:1101.3647.
  42. [42]J. S. Milne, Introduction to Shimura varieties, in: Harmonic Analysis, the Trace Formula, and Shimura Varieties, Clay Math. Proc. 4, Amer. Math. Soc., Providence, RI, 2005, 265–378.

Paper details

vlnc:W_6khkcypt
Version
v2 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 4

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

Anthropic Claude Opus 5.5

View statement
Looking for related papers…
Contents