Heisenberg vertex algebras and abelian varieties

arXiv:2604.00103 · math.AG, math.QA · Submitted 2026-03-31 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Mathematics. The day in mathematics: new theorems, new proofs, and what they open.

Ada: We're looking at the paper: "Heisenberg vertex algebras and abelian varieties".

Title and authors: Ada: So we're looking at this paper by Tarasca titled "Heisenberg vertex algebras and abelian varieties," and it's tackling how to define spaces of coinvariants for these vertex algebras when you move from simple curves to much more complex abelian varieties.

Emmy: Right, Ada, the main idea here is that they're taking the standard way we construct these spaces of coinvariants on a curve and seeing if it works when you generalize it to any principally polarized abelian variety.

Kolya: So, what's the big takeaway for someone who doesn't specialize in vertex algebras or algebraic geometry? Does this just mean we can now study these structures on much more complicated shapes than just simple Riemann surfaces?

Ada: Exactly. The paper constructs spaces of coinvariants for Heisenberg vertex algebras on any principally polarized abelian variety, and it shows that these constructions naturally globalize to twisted D-modules when you look at the moduli space of abelian varieties.

Emmy: It's quite structural, showing that these spaces aren't just local things on a curve; they have this deeper structure when viewed globally over the moduli space.

Kolya: I mean, are we talking about finding some kind of universal tool here for these vertex algebra constructions across all abelian varieties? Or is it more about showing that the machinery works in principle?

Ada: It's both. They show that by using the extended PPAVs and a specific quotient involving a Lie algebra of outgoing states, you get this sheaf on the moduli space of abelian varieties, which then carries this twisted D-module structure for g greater than or equal to three.

Emmy: And they refine previous work by using larger Lie algebras that depend on the vertex algebra itself, which helps them recover those standard constructions while keeping that twisted D-module structure intact.

Kolya: So, if I'm driving and I have a few seconds to listen, what does this mean practically? Does it help us understand some physical system where we have these kinds of symmetries?

Ada: It tells us that the underlying algebraic structure of these coinvariants is robust enough to survive this kind of geometric generalization. They explicitly show how the sheaf on the universal extended abelian variety Xg carries an action of the Atiyah algebra minus half times d times FXi, which is what induces that twisted D-module structure.

Title and authors: Emmy: And they link this back to the universal Jacobian, showing that the sheaf on Xg descends to a quasi-coherent sheaf on Xg that also carries a twisted D-module structure.

Kolya: That descent from the universal family down to something more manageable is always a big deal in geometry. What about the specific examples they give for commutative vertex algebras?

Ada: They provide an explicit realization for the commutative Heisenberg vertex algebra, pi zero of level zero, showing it has a canonical isomorphism that looks like a symmetric power of one-forms on the curve <ref:2604.00103#pg1>.

Emmy: And then they extend that idea to any principally polarized abelian variety (X, theta) by showing a canonical isomorphism between the space of coinvariants V X, theta, pi zero and the symmetric algebra of the space of one-forms on X, extended by some terms <ref:2604.00103#pg1>.

Kolya: So, for those who only listen to this show and aren't deep in vertex algebra theory or algebraic geometry, what does that mean? What's the payoff?

Ada: It means we have a concrete way to relate these abstract vertex algebra constructions to concrete geometric objects like the space of one-forms on an abelian variety. They establish that this relationship holds even when you go beyond just the simple curves.

Emmy: And they do it by using group scheme formalism, upgrading things like symplectic Lie groups to group schemes, which allows them to make these infinitesimal realizations work globally over the moduli space of abelian varieties.

Kolya: That sounds complicated. Is there a limit to how far this machinery can go? Where does it stop working for these lattice vertex algebras you mentioned?

Ada: They are looking at lattice vertex algebras now, and they investigate whether the Lie algebra of outgoing states admits an extension to PPAVs for those cases, checking if it incorporates the additional vertex operators from the lattice vectors.

Emmy: That investigation is key because it determines whether this construction can be extended to those more intricate lattice structures we see in physics.

Kolya: So, to wrap up the main points of this paper, what's the big conclusion about spaces of coinvariants for Heisenberg vertex algebras and abelian varieties?

Ada: The final result is that for g greater than or equal to three, the sheaf on the moduli space of abelian varieties carries a twisted D-module structure, confirming that these spaces of coinvariants are intrinsically tied to PPAVs.

Title and authors: Emmy: And they show this holds even when you look at the universal extended abelian variety Xbg, where it carries an action of the Atiyah algebra minus half times d times FXi.

Kolya: I think what this means is that we have a solid, global framework now to study these vertex algebra objects on these varieties, which were previously only understood locally on curves.

Ada: That's right. They've shown the construction of spaces of coinvariants assigned to PPAVs by defining them as a quotient over the moduli space Sp(X, theta), and this sheaf descends to a vector space via descent along Sp(H') acting on that quotient.

Emmy: And they confirm compatibility with the universal Jacobian Jg, showing that the sheaf on Xg restricts correctly to the sheaf on Jg, which extends the way it works for simple curves.

Kolya: So, if we're driving and I want to know what this means for me, it means that these abstract vertex algebra spaces are not just mathematical curiosities confined to specific curves; they have a consistent, global structure on the space of all possible abelian varieties.

Ada: Precisely. The paper "Heisenberg vertex algebras and abelian varieties" provides the toolset to see how these structures interact globally over the moduli space of abelian varieties, leading directly to twisted D-module theory.

Emmy: It’s a nice piece because it connects representation theory—the vertex algebras—to complex geometry in a very explicit way through these spaces of coinvariants.

Kolya: I think for future work, they hint at finding a realization for the Heisenberg vertex algebra at level one, which would be interesting to see how that extends the results we have here.

Ada: Yeah, they mention looking ahead to find a realization of coinvariants on PPAVs in the case of the Heisenberg vertex algebra at level one, and that would be a natural next step for this line of inquiry.

Emmy: It’s building on what they did for level zero, so it’s clearly a direction they want to pursue further.

Kolya: So, to sum up the implications: we have a global construction for spaces of coinvariants on abelian varieties that results in twisted D-modules when g is at least three.

Ada: That's the core result of "Heisenberg vertex algebras and abelian varieties." It shows these spaces are intrinsic to PPAVs and globalize beautifully into twisted D-modules on the moduli space of abelian varieties.

The paper's summary: Ada: So, to recap, this paper is about taking the standard way we define spaces of coinvariants for Heisenberg vertex algebras on simple curves and showing that this construction works when you move over to any principally polarized abelian variety, and it globalizes into a twisted D-module structure on the moduli space.

Emmy: Exactly. It’s really about showing that these spaces aren't just local things stuck on a curve; they have this deeper structure that survives when you look at the whole family of abelian varieties together.

Ada: The authors do this by using these extended abelian varieties and a specific quotient based on outgoing states, which then naturally leads to those twisted D-module structures when we look at the moduli space of abelian varieties.

Emmy: And they don't just stop there; they show how this works for the universal family of stable curves, pulling back from the universal extended abelian variety down to the universal Jacobian.

Ada: That means if you know how it behaves on a single curve, you can use that information to understand how it behaves across all possible abelian varieties in a very structured way.

Emmy: It’s also really useful because they give an explicit realization for the commutative Heisenberg vertex algebra, showing it’s related to the symmetric power of one-forms on any given abelian variety.

Ada: And that connection is made even more concrete by showing how the sheaf on the universal extended abelian variety carries a twisted D-module structure using that Atiyah algebra action.

Emmy: That action, specifically minus half times d times FXi, is what ties everything together globally over the moduli space of abelian varieties for g greater than or equal to three.

Ada: The result is that these spaces of coinvariants are intrinsic properties of principally polarized abelian varieties, and they have a very specific geometric structure when you view them on the moduli space.

Emmy: It connects vertex algebra representation theory directly to complex geometry in a way that’s really quite explicit through these sheaves and D-modules.

Ada: So, for someone who only listens to this show, it means we have a concrete global tool now for studying these complicated vertex algebra objects across all possible abelian varieties.

Emmy: And they are doing this using group scheme formalism, treating things like symplectic Lie groups as group schemes so the infinitesimal actions make sense globally.

Ada: They even look into whether these constructions can be extended to lattice vertex algebras by checking if the outgoing states Lie algebra has enough extra operators from the lattice vectors.

Emmy: That’s a crucial check because it determines whether this whole machinery works for those more complex lattice structures that show up in certain physical models.

Ada: So, we have a solid foundation now showing that these spaces of coinvariants are not just local curiosities; they have a consistent global structure on the moduli space of abelian varieties.

The paper's improvements: Ada: So, we're looking at what the authors suggest for future work on this paper about Heisenberg vertex algebras and abelian varieties.

Emmy: They’re actually pointing toward extending this construction to a higher level of the vertex algebra, specifically level one.

Ada: They are suggesting that finding a realization for the coinvariants in that level one case would be a natural next step for their research line.

Emmy: That makes sense because they’ve shown how it works for level zero, and seeing how that structure behaves at a different level would give them more leverage on the general problem.

Ada: They are essentially saying that the method used here is flexible enough to handle those higher levels of the algebra.

Emmy: It means they aren't just stuck with this level zero case; they can use this framework to explore more complex vertex algebras as well.

Ada: So, for someone who only listens to this show, it suggests that the tools they’ve developed here aren't limited to the simplest cases of these algebras.

Emmy: It points toward a deeper understanding of how these algebraic structures behave when you change the level parameter in a vertex algebra.

Ada: And I think it implies that we might be able to use this geometric approach to solve problems involving more intricate vertex algebras than just the simplest ones we’ve looked at.

Emmy: It seems like they are building a roadmap for how this geometric construction can be applied across different representations of these algebras.

Conclusion: Ada: So, to wrap up this discussion on "Heisenberg vertex algebras and abelian varieties," we’ve seen how this construction creates a twisted D-module structure on the moduli space of abelian varieties for g greater than three.

Emmy: It’s pretty significant because it shows that these spaces of coinvariants are deeply tied to the geometry of principally polarized abelian varieties, not just abstract math.

Ada: And they’ve established this connection by showing how the universal extended abelian variety carries an action of a specific Atiyah algebra, which induces that twisted D-module structure.

Emmy: Plus, they proved this holds even when you look at the universal Jacobian and showed it works for the commutative Heisenberg vertex algebra by relating it to symmetric powers of one-forms on any abelian variety.

Ada: It’s a lot of work to get those results, but the implication is that we have a robust way to study these vertex algebra objects globally across all possible abelian varieties.

Emmy: It really shows how representation theory and complex geometry can be intertwined through these sheaves and D-modules.

Kolya: From my side, what I like is that they’ve done the heavy lifting for the level zero case, and now they’re looking at how this extends to lattice vertex algebras by checking if the Lie algebra of outgoing states has those extra operators from the lattice vectors.

Ada: Right, so they're checking if that extension exists for those more complicated lattice structures.

Kolya: That check is important because it tells us whether the constants survive when we apply this to those more difficult systems.

Emmy: It means they are really probing the limits of this construction beyond just the simplest algebraic settings.

Ada: And they’ve also pointed toward level one as a natural next step for finding a full realization of these coinvariants for those algebras.

Emmy: That extension would give them more ground to explore how these structures behave at higher levels, which is important for the whole theory.

Kolya: So, we have this global framework now, and they’re already looking at pushing the boundaries by checking those lattice cases and moving up to level one.

NICOLA TARASCA

math.AG, math.QA

Submitted: 2026-03-31

Updated: 2026-11-17

Comments: 45 pages. Comments are welcome

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The gist: Here we construct spaces of coinvariants for Heisenberg vertex algebras on abelian varieties and show that these globalize to twisted D-modules on the moduli space of abelian varieties.

Key concepts

Heisenberg Vertex Algebra
This is a specific type of vertex algebra used in conformal field theory. It's defined by a set of generators and relations that govern how fields interact, often related to the structure of abelian varieties. The paper studies its representations and coinvariants to understand geometric structures.
Twisted D-module
A twisted D-module is a sophisticated mathematical object that combines differential operators with a twisting element. In this context, it arises from the action of an algebra (like the Atiyah algebra) on spaces of coinvariants, providing a way to study geometric properties on moduli spaces.
Abelian Variety Moduli Space (Abg)
This is the space that parameterizes all principally polarized abelian varieties. The paper uses this space to define global structures. By showing that the constructed sheaves descend to this moduli space, it connects vertex algebra representations to geometric objects on these varieties.
Coinvariants
Coinvariants are derived from a representation of an algebra by taking the quotient by a certain ideal related to outgoing states. These spaces capture essential algebraic information about the vertex algebra's structure and are used to build geometric sheaves.

Terminology

Summary

Here we construct spaces of coinvariants for Heisenberg vertex algebras on abelian varieties and show that these globalize to twisted D-modules on the moduli space of abelian varieties. p. 0

Global Construction and Recovery of Standard Sheaves

The work demonstrates that for Heisenberg vertex algebras, the spaces of coinvariants can be defined for arbitrary principally polarized abelian varieties, naturally recovering the standard construction over the Torelli locus. This is achieved by applying the Arbarello–De Concini moduli space Abg of extended PPAVs to define a space of coinvariants Vb(a, V) as a quotient of V by a Lie algebra of outgoing states determined by both the extended PPAV and the vertex algebra. p. 0 Theorem 1 states that for such an algebra, these spaces give rise to a quasi-coherent sheaf V(V) on Ag whose restriction to the Torelli locus recovers the standard sheaf of coinvariants on Mg. p. 0

Geometric Realization and Twisted D-Modules

The construction extends from curves to extended PPAVs by defining a space Vb(x, V) for a point x of the universal extended abelian variety Xbg via pull-back under the map Xbg → Abg. Theorem 2 shows that this space gives rise to a sheaf V(V) on Xg which carries an action of the Atiyah algebra-d/2 FΞ, inducing a twisted D-module structure. p. 1 This result refines previous work by employing larger Lie algebras depending on the vertex algebra to recover standard constructions while preserving the twisted D-module structure. p. 1

Explicit Realization for Commutative Algebras

An explicit realization of coinvariants is presented for the commutative Heisenberg vertex algebra π0 of level 0, showing that it admits a canonical isomorphism V C, P, π0 ∼= Sym H0 C, omega1 C∗. This extends to arbitrary principally polarized abelian varieties (X, Θ) via Theorem 3, which states that the space of coinvariants V X, Θ, π0Γ admits a canonical isomorphism V X, Θ, π0Γ ∼= Sym H0(X, omega1X)∗ <ref:2604. p. 200103#pg10>.

Descent to Global Moduli Spaces

The paper constructs spaces of coinvariants assigned to PPAVs by defining Vb(X, Θ, V) as a quotient of the space of vertex algebra representations over the moduli space Sp(X, Θ). Theorem 5.2 proves that this sheaf is Sp+(H')-equivariant and that its stabilizer acts trivially on the fiber Vb(a, V) at a. This leads to the definition of the space of coinvariants assigned to the PPAV (X, Θ) as V(X, Θ, V). p. 1

Final Result and Twisted D-Module Structure

The final result establishes that for g ≥ 3, the sheaf V(V) on Ag carries an action of the Atiyah algebra d/2 FΛ, inducing a twisted D-module structure. This is achieved by showing that the action of mp(H')⊗C OAbg factors to an action of the Atiyah algebra d/2FΛ on Vb(V) and that this action descends to V(V). This confirms the global nature of the construction. p. 0 p. 0

Sheaves on Universal Families

The paper also constructs sheaves of coinvariants on Xbg and Xg, showing that Vb(V) on Xbg carries an action of the Atiyah algebra-d/2 FΞ, inducing a twisted D-module structure. This sheaf Vb(V) descends to a quasi-coherent sheaf V(V) on Xg also carrying a twisted Dmodule structure. p. 0

Compatibility with Universal Jacobian

The construction is compatible with the spaces of coinvariants on the universal Jacobian Jg, showing that the sheaf V(V) on Xg restricts to the sheaf VJg(V) over the universal Jacobian Jg. This demonstrates that the descent of Vb(V) along Xbg → Xg extends the descent of VbJbg(V) along Jbg → Jg. p. 2

Functional Realization

The paper concludes with the functional realization for V = π0Γ, showing a canonical isomorphism V X, Θ, π0Γ ∼= Sym hΓ ⊗C H0(X, omega1X)∗. This confirms the global nature of the construction. The proof for Theorem 5. p. 04 is based on Example 3. p. 04 where Vb a, π0Γ ∼= Sym (hΓ ⊗C Z/F). p. 1

Group Scheme Formalism

The construction is formalized by upgrading the symplectic Lie groups to group schemes, such as Sp+(H') and Mp(H'), which allows for the infinitesimal realization of Abg as a homogeneous space for the action of a metaplectic group ind-scheme. This framework is essential for globalizing the construction over the moduli space Abg. p. 0 The actions of these groups on the spaces of coinvariants are shown to be exponentiable, leading to Sp+(H')-equivariant structures.

Conclusion on Coinvariants

The final result is the construction of a sheaf V(V) on Ag that carries a twisted D-module structure for g ≥ 3. This demonstrates that spaces of coinvariants are intrinsic to PPAVs and globalize to twisted D-modules on the moduli space of abelian varieties. p. 0 The result is further generalized for lattice vertex algebras. p. 2

References

[ADC] E. Arbarello and C. p. 0 De Concini. p. 0 Abelian varieties, infinite dimensional Lie algebras, and the heat equation. p. 4

[ADCKP] E. Arbarello, C. p. 0 De Concini, V. p. 0 G. Kac, and C. p. 9 Procesi. Moduli spaces of curves and representation theory. p. 0

[Alp] J. Alper. Good moduli spaces for Artin stacks. p. 0

[BD1] A. Beilinson and V. p. 1 p. 1 p. 1 p. 1 Drinfeld. Affine Kac-Moody algebras and polydifferentials. p. 2

[BD2] A. Beilinson and V. Drinfeld. Quantization of Hitchin’s integrable system and Hecke eigensheaves.

[BD3] A. Beilinson and V. Drinfeld. Chiral algebras, volume 51 of Am Mathem Soc Colloquium Publications.

[BFM] A. Beilinson, B. p. 1 Feigin, and B. Mazur. Introduction to algebraic field theory on curves. p. 0

[BS] A. A. Beilinson and V. V. Schechtman. Determinant bundles and Virasoro algebras. p. 2

[CL] Y. Chu and Z. Lin. The classification of semi-conformal structures of Heisenberg vertex operator algebras. p. 0

[DGT1] C. Damiolini, A. Gibney, and N. Tarasca. Conformal blocks from vertex algebras and their connections on Mg,n. p. 0

[DGT2] C. Damiolini, A. Gibney, and N. Tarasca. On factorization and vector bundles of conformal blocks from vertex algebras. p. 0

[FBZ] E. Frenkel and D. p. 1 Ben-Zvi. Vertex algebras and algebraic curves, volume 88 of Mathematical Surveys and Monographs. p. 0

[Kac] V. Kac. Vertex algebras for beginners, volume 10 of University Lecture Series. p. 5

[Kon] M. L. Kontsevich. The Virasoro algebra and Teichmüller spaces. p. 6

[MN] A. Matsuo and K. Nagatomo. A note on free bosonic vertex algebra and its conformal vectors. p. 2

[Tar] N. Tarasca. Coinvariants of metaplectic representations on moduli of abelian varieties.

[TUY] A. Tsuchiya, K. Ueno, and Y. Yamada. Conformal field theory on universal family of stable curves with gauge symmetries. p. 5

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved ability to construct spaces of coinvariants for Heisenberg vertex algebras on arbitrary principally polarized abelian varieties by extending a construction from curves using extended PPAVs and showing it globalizes to twisted D-modules on the moduli space of abelian varieties.

  2. Improved ability to recover standard constructions over the Torelli locus by employing larger Lie algebras depending also on the vertex algebra, as stated: By employing larger Lie algebras here depending also on the vertex algebra (see Remark 7.2), we now recover the standard constructions while preserving the twisted D-module structure.

  3. Improved ability to provide an explicit realization of coinvariants for commutative Heisenberg vertex algebras by showing that in this case, it is shown in [FBZ, §9.4.3] building on Beilinson–Drinfeld [BD2] that the space of coinvariants V C, P, π0 assigned to a pointed curve (C, P) admits a canonical isomorphism Sym H0 C 1 C∗.

  4. Improved ability to relate spaces of coinvariants on extended abelian varieties Xbg to those on the universal Jacobian Jg by showing that The sheaf of coinvariants on Jg as constructed in [FBZ, §§9.5.5, 18.1.15] can be obtained as follows: one first considers the pull-back Vb Jbg (V) on Jbg of the sheaf of coinvariants on Mcg.

  5. Improved ability to determine if the Lie algebra of outgoing states admits an extension to PPAVs for lattice vertex algebras by investigating whether the Lie algebra of outgoing states for lattice vertex algebras on curves also incorporates the additional vertex operators of the lattice vectors and whether the Lie algebra of outgoing states admits an extension to PPAVs also for lattice vertex algebras.

  6. Improved ability to establish a canonical isomorphism between spaces of conformal blocks and symmetric powers of spaces of one-forms on PPAVs, as stated in Theorem 5.4: For a PPAV (X, Θ), one has a canonical isomorphism V X, Θ, π0 Γ ∼= Sym hΓ ⊗C H0 (X, omega 1 X)∗.

  7. Improved ability to construct spaces of coinvariants assigned to PPAVs by showing that the space Vb(X, Θ, V) descends to a vector space via the descent along Sp+(H') acting on [Abg/Sp+(H')].

  8. Improved ability to prove that sheaves of coinvariants on Xbg carry twisted D-module structures by using Corollary 8.2, which states that for g >= 3, the sheaf Vb(V) on Xbg carries an action of the Atiyah algebra - d squared FΞ, inducing a twisted D-module structure.

Sources

Related papers