Tensor hierarchy from deformation quantisation holds only within the QP-manifold and DFT setting

What the tensor hierarchy and DFT already fixed

Double field theory makes T-duality manifest by extending the tangent bundle to TM⊕T*M and promoting the structure group from GL(D) to O(D,D), combining the metric and Kalb-Ramond field into a generalised metric and elevating diffeomorphisms and B-field transformations to generalised diffeomorphisms [1]. Hull and Zwiebach showed that the gauge algebra of doubled torus fields is T-duality covariant, reduces to the Courant bracket for restricted parameters, and fails the Jacobi identity in a controlled way [7]. Hull later clarified that finite gauge transformations reveal an underlying gerbe structure and that a constant split-signature metric does not restrict the doubled geometry provided it is treated as a generalised tensor [8]. The reducibility of these generalised diffeomorphisms — exact one-forms generating vanishing gauge transformations — is algebraically captured by the tensor hierarchy, a differential graded Lie algebra concentrated in positive degrees whose subspaces R1, R2, … hold generalised vectors, gauge-for-gauge parameters and higher reducibility data [1].

The QP-manifold frontier and the missing dilaton

By Roytenberg's theorem, degree-2 QP-manifolds are in one-to-one correspondence with Courant algebroids, the algebraic backbone of generalised geometry, and the most general degree-3 Hamiltonian on T*[2]T[1]M reproduces the Courant algebroid axioms once the classical master equation {S,S}=0 is imposed [1]. This framework encodes the kinematical gauge structure of generalised geometry but does not by itself specify the dynamical NS-NS background, and it fails to incorporate the generalised dilaton or its descendants, leaving the geometric origin of the dilaton obscured [1]. A parallel line of work developed the frame formulation of DFT using Clifford algebras and Dirac generating operators, where a Dirac-type operator plays the role of the differential, generates Courant algebroid structures via derived brackets, and naturally contains the dilaton flux as an ambiguity [1]. Boffo and Schupp had already explored a gravitational action with stringy Q and R fluxes via deformed differential graded Poisson algebras on graded symplectic manifolds, establishing a precursor for deforming the QP structure [3].

How the star product generates the dilaton and completes the hierarchy

The anchor paper deforms the algebra of functions on the QP-manifold using a normal-ordered graded Moyal–Weyl star product controlled by a central coordinate ħ of degree 2, so that ξA⋆ξB = ξAξB + ħηAB [1]. Because a generic degree-2 generator α ∼ MAB ξA⋆ξB then contains a trace part proportional to ħ, the generalised dilaton is not added by hand but arises from the generic structure of the deformed algebra, and the structure group is automatically extended from O(D,D) to O(D,D)×R+ [1]. The deformed master equation {S,S}⋆=0 governs the kinematic data, yielding the section condition and the Bianchi identities for the DFT fluxes, while the one-form flux is largely fixed in terms of the frame's divergence up to a residual gradient ambiguity whose resolution requires the generalised dilaton [1]. The paper presents two realisations: a strict differential graded Lie algebra built from the full algebra of functions on the deformed QP-manifold, and a cochain complex obtained by projecting to the momentum-independent subspace that reproduces the tensor hierarchy representations conjectured in the literature and can be read as a classical BV complex of the linearised theory [1].

Where this sits relative to superspace and double-copy approaches

Butter's superspace treatment of N=1 supersymmetric DFT found that maintaining the torsion constraints requires expanding the local tangent space group of O(D)×O(D) to include a tower of higher-dimension generators, including a hook-representation generator gauging the spin-connection shift symmetry, and proposed a relation to the super-Maxwell∞ algebra [2]. The type II superspace extension similarly enhances the local tangent space beyond the double Lorentz group to eliminate unphysical supervielbein components and define covariant torsion and curvature tensors, producing an infinite hierarchy of local tangent space symmetries connected to super-Maxwell∞ [6]. The anchor paper's approach differs in method: rather than enlarging the tangent space by hand, it derives the additional structure from the star-product deformation and identifies the resulting cochain complex as the algebraic data needed for covariant curvature and torsion tensors in generalised Cartan geometry [1]. Bonezzi, Chiaffrino, Díaz-Jaramillo and Hohm constructed weakly constrained DFT to quartic order as the double copy of Yang-Mills theory, building the L∞ algebra on a double-copied tensor product space via homotopy transfer and a nonlocal shift, and verifying the three-brackets and Jacobi identities up to homotopy for the gauge sector [5]. That work targets interacting quartic dynamics in the weakly constrained toroidal setting, whereas the anchor paper targets the complete classical kinematics and linearised BV complex in the strongly constrained QP setting, so the two are complementary rather than directly competing [1][5].

Boundaries: formal, linearised, and conditional on local double Lorentz invariance

The construction is limited to formal deformation quantisation and the linearised level: no full nonlinear quantum corrections or global geometric existence proof are provided [1]. Dynamics are not derived from the deformed master equation alone; they emerge from an independent action principle in which a physical generalised metric defines a chiral grading operator splitting the Hamiltonian into left and right eigenspaces, and demanding local double Lorentz invariance together with a Z2 symmetry exchanging the chiral sectors uniquely fixes the DFT action up to overall normalisation [1]. The projected differential on the momentum-independent subspace is no longer an inner derivation and the Leibniz rule fails by a term controlled by the odd symplectic structure, so the strict DGLA structure is lost and the kinematics are governed instead by a cochain complex with an associative star product and a Dirac generating operator as differential [1]. The authors flag this failure as suggestive of a non-commutative BV structure in which the second-order part of Q plays the role of a BV Laplacian, and note that the expansion {S,S}⋆ = {S,S}cl + ħΔS + … takes the form of a quantum master equation, but the algebraic properties of this projected structure remain an open problem [1]. Extension to exceptional field theory is substantially more involved because exceptional groups introduce non-trivial representation structure already at positive degrees, and standard QP-manifold approaches struggle to reach the negative-degree representations [1]. A separate limitation signal comes from outside the DFT literature: Nasr and Al-Qashbari's study of generalized U-birecurrent Finsler spaces shows that curvature-tensor identities in generalized geometric settings depend sensitively on the chosen connection and recurrence order, a reminder that algebraic constructions of curvature tensors do not automatically transfer across geometric frameworks [4].

About These Sources

This research page is built on 8 studies (7 peer-reviewed, 1 preprint) — published from 2009 to 2026, 2 from 2024 or later, collectively cited 537 times — selected as the most relevant from 9 studies that passed quality screening, drawn from 64 papers retrieved from a database of over 500 million.

Sources used in this answer