Self-dual holography:

2026

Read Self-dual holography: (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-30276.txt.

Introduction

Evgeny Skvortsov and Richard Van Dongen build the Fefferman-Graham growth and map for self-dual theories of any spin, a step toward self-dual holography. Self-dual theories, UV finite, should have AdS/CFT twins, and higher-spin versions make the twins simpler. Chiral higher-spin gravity stands as the top self-dual theory, twin to part of Chern-Simons matter theories. They find bulk propagators and four-point AdS/CFT links in slimmed Chiral higher-spin gravity, a higher-spin form of self-dual Yang-Mills theory.

Catalog record

YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE112451
LICENSEhttp://creativecommons.org/licenses/by-sa/4.0/
RIGHTSCC BY SA
SOURCEarXiv

Source: https://arxiv.org/abs/2605.30276

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Self-dual holography: Abstract Self-dual theories, being UV-finite, should have their own AdS/CFT dualities. Higher-spin extensions of self-dual theories are attractive to simplify the CFT duals. The maximal self-dual theory is Chiral higher-spin gravity, which should be dual to a subsector of Chern–Simons matter theories. As a step toward establishing self-dual holography, we develop the Fefferman–Graham expansion and holographic dictionary for arbitrary spin self-dual theories. As an application, we derive bulk-to-bulk propagators and compute three- and four-point AdS/CFT correlators in a contraction of Chiral higher-spin gravity, which is a higher-spin extension of self-dual Yang–Mills theory. four-point AdS/CFT correlators in higher-spin gravity Evgeny Skvortsov * * * Also affiliated with Lebedev Institute of Physics. & Richard van Dongen Service de Physique de l’Univers, Champs et Gravitation, Université de Mons, 20 place du Parc, 7000 Mons, Belgium 1 Introduction The main goal of this paper is to further develop self-dual holography [ 1 , 2 , 3 , 4 , 5 , 6 , 7 ] , i.e. to identify the role of self-dual theories in (A)dS/CFT and celestial holography. Concretely, we concentrate on self-dual theories with massless fields of arbitrary spin in AdS 4 . The maximal such theory — the “M-theory” of self-dual ones — is chiral higher-spin gravity [ 8 , 9 , 10 , 2 , 11 ] . Its spectrum has fields of all spins and it features all interactions compatible with self-duality. Self-dual theories are of great interest for many reasons, to mention a few: (a) they are consistent truncations of some bigger nonchiral theories and all solutions of self-dual theories are also solutions of the parent ones; (b) self-dual theories are integrable; (c) all amplitudes of self-dual theories are also amplitudes of their completions, i.e. self-dual theories allow one to compute some, but not all, amplitudes of more physical theories; (d) they do not have UV divergences; (e) they admit simple twistor formulations; (f) the parent theories can be represented as expansions over the self-dual sectors, which is richer than the usual perturbative expansion over a trivial vacuum. Therefore, self-dual theories provide a very useful approximation of the parent theories. Self-dual theories admit higher-spin extensions of various types, see the figure 1 and [ 10 , 12 , 13 , 14 , 15 , 16 ] for more details that the figure can display. Provided we restrict ourselves to either gauge or gravitational interactions, there exist simple higher-spin extensions of self-dual Yang-Mills theory (SDYM) and self-dual gravity (SDGR) — HS-SDYM and HS-SDGR. All self-dual theories, in particular those with genuine higher-spin higher-derivative nonabelian interactions, are contained in chiral higher-spin gravity, which thereby incorporates all interactions that are compatible with self-duality. The amplitudes of SDYM and SDGR are contained in the amplitudes of Chiral HiSGRA. Conversely, HS-SDYM and HS-SDGR compute some of the amplitudes of Chiral theory, which is the property we will use in the paper. Figure 1: The “map” of self-dual theories. Along x we plot spin and along y we display the “number of derivatives” in cubic vertices. The smallest theories are SDYM and SDGR, which admit higher-spin extensions: HS-SDYM and HS-SDGR. The maximal self-dual theory is chiral higher-spin gravity. There are many intermediate cases [ 12 , 16 ] and the complete classification is not yet available. The x -line itself is forbidden as there are no vertices with vanishing total helicity except for the scalar self-interaction. Self-dual theories admit gauge- and Lorentz-invariant formulations, e.g. Chalmers–Siegel [ 17 ] and Krasnov [ 18 ] formulations for SDYM and SDGR, respectively. Covariant formulations for HS-SDYM and HS-SDGR were constructed in [ 13 ] . However, the structure of the interactions is perhaps best exhibited in the light-cone gauge and in flat space. As is well-known, given any three helicities \lambda_{1,2,3} such that |\sum_{i}\lambda_{i}|>0 (no discrimination of higher-spins), there is a unique cubic vertex V_{\lambda_{1},\lambda_{2},\lambda_{3}} [ 19 , 20 ] and the corresponding three-point amplitude reads [ 21 ] \displaystyle V_{\lambda_{1},\lambda_{2},\lambda_{3}}\Big|_{\text{on-shell}}\sim\begin{cases}[12]^{\lambda_{1}+\lambda_{2}-\lambda_{3}}[23]^{\lambda_{2}+\lambda_{3}-\lambda_{1}}[13]^{\lambda_{1}+\lambda_{3}-\lambda_{2}}\,,&\sum\lambda_{i}>0\,,\\ \langle 12\rangle^{-\lambda_{1}-\lambda_{2}+\lambda_{3}}\langle 23\rangle^{-\lambda_{2}-\lambda_{3}+\lambda_{1}}\langle 13\rangle^{-\lambda_{1}-\lambda_{3}+\lambda_{2}}\,,&\sum\lambda_{i}<0\,.\end{cases} (1.1) For example, SDYM has a single interacting vertex V_{+1,+1,-1} , which is “half” of the Yang-Mills cubic vertex V_{+1,+1,-1}+V_{-1,-1,+1} . SDGR has only V_{+2,+2,-2} , which is again “half” of the cubic vertex of the Einstein-Hilbert action. HS-SDYM has all spins and \sum_{i}\lambda_{i}=1 , 2 2 2 Let us fix the preferred value of \Lambda=\sum\lambda_{i} to be positive. For every statement below there is a similar one with \sum\lambda_{i}<0 . The number of (transverse) derivatives in a vertex in the light-cone gauge is |\Lambda| . i.e. all vertices are single-derivative of the Yang-Mills type. HS-SDGR also has all spins but \sum_{i}\lambda_{i}=2 , i.e. all interactions are two-derivative gravitational-type interactions. Chiral HiSGRA has all spins and all possible interactions with \sum_{i}\lambda_{i}>0 : under quite general assumptions having a genuine higher-spin interaction forces all helicities to be present and also fixes the coupling constants to be [ 8 , 9 , 10 ] \displaystyle V_{\text{Chiral}} \displaystyle=\sum_{\lambda_{1},\lambda_{2},\lambda_{3}}C_{\lambda_{1},\lambda_{2},\lambda_{3}}V_{\lambda_{1},\lambda_{2},\lambda_{3}}\,, \displaystyle C_{\lambda_{1},\lambda_{2},\lambda_{3}}=\frac{\kappa\,(l_{p})^{\lambda_{1}+\lambda_{2}+\lambda_{3}-1}}{\Gamma(\lambda_{1}+\lambda_{2}+\lambda_{3})}\,. (1.2) As a result, there is a simple action for Chiral theory in the light-cone gauge in flat space. It is also possible to covariantize the theory at least at the level of equations of motion [ 2 , 11 ] . The equations of motion are smooth in the cosmological constant and Chiral theory exists also in \text{(A)dS}_{4} , which makes it an attractive model for holography since it is a perturbatively local and (supposed to be [ 22 , 23 ] ) UV-finite higher-spin gravity. Self-dual theories, being UV-finite, provide a natural playground for AdS/CFT correspondence (or for any other type of holography). Indeed, one needs a consistent theory of quantum gravity to have a complete model on the bulk side of the duality. (Super)gravity fails to be renormalizable and calls for a stringy completion, which complicates the model. Alternatively, self-dual theories are interesting toy models of quantum gravity. There is a good chance they can lead to nontrivial exactly soluble models of holography. There does not seem to be any other easily tractable framework in between self-dual theories and string theory that has propagating degrees of freedom, most notably the graviton. The last point requires more explanation since the statement has varied with time. When looking for simple models of the AdS/CFT correspondence it was initially noted that vector models (both the free and the critical one) should have “simpler” duals [ 24 , 25 , 26 ] : conserved (higher-spin) currents present in the free vector model (or large- N critical vector model) are dual to massless (higher-spin) fields and, hence, the AdS theory has to be a higher-spin gravity. Vector models are, in some sense, the smallest CFTs and the bulk dual has a finite number of fields for a given spin (say, one). Later, this idea was extended to Chern–Simons matter theories [ 27 ] , i.e. to vector models with bosonic and/or fermionic matter coupled to Chern–Simons theory, and played an important role in establishing the three-dimensional bosonization duality [ 27 , 28 , 29 ] . However, not every CFT has a quasi-classical and (perturbatively) local dual. The bulk coupling is of order 1/N for vector models and, hence, the dual admits a perturbative expansion. What prevents the dual from existing is the fact [

Read the complete text