A finite 6d supergravity landscape from anomalies

Yuta Hamada, and Gregory J. Loges · 2025

Read A finite 6d supergravity landscape from anomalies by Yuta Hamada, and Gregory J. Loges (2025) 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-2507-20949.txt.

Introduction

Hamada and Loges present a 2025 physics paper proving that six-dimensional supergravity models form a finite family once all anomaly rules apply. Local anomaly freedom alone leaves endless model choices. Adding global anomaly freedom changes the game. The pair boil the rules down to a giant linear program and study its dual with standard tools.

A universal cap falls out: at most 3,003 tensor multiplets in any consistent model. Finiteness of the whole landscape follows from that bound. One extreme model saturates the cap with a huge repeated gauge group, oddly echoing a known F-theory champion. The paper turns an open sea of models into a countable shore.

Catalog record

CREATORYuta Hamada, and Gregory J. Loges
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE104 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2507-20949

Browse the full library · About the Tower of Babel

Full text

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

A finite 6d supergravity landscape from anomalies Preprint: KEK-TH-2741 Yuta Hamada, and Gregory J. Loges Abstract 6d supergravities with non-abelian gauge group are subject to many consistency conditions. While the absence of local gauge and gravitational anomalies allows for infinitely many models, we show that those conditions stemming from the absence of both local and global anomalies together are strong enough to leave only finitely many consistent models. To do this we distill the consequences of anomaly cancellation into a high-dimensional linear program whose dual can be efficiently studied using standard techniques. We obtain a universal bound on the number of tensor multiplets T ≤ 11 ⋅ 273 = 3003 T\leq 11\cdot 273=3003 and show that this leads to a finite landscape of consistent non-abelian models. Interestingly, the model which saturates this bound has gauge group [ E 8 × F 4 × ( G 2 × SU ⁡ ( 2 ) ) 2 ] 273 [E_{8}\times F_{4}\times(G_{2}\times\SU(2))^{2}]^{273} , which bears a striking resemblance to the model which saturates the bound T ≤ 193 T\leq 193 for F-theory constructions. 1 Introduction It has long been appreciated that amongst all low-energy effective theories only a very small fraction are compatible with the principles of quantum gravity. Several general arguments have been made that the quantum gravity landscape is in fact finite (e.g. see Vafa:2005ui ; Acharya:2006zw ; Grimm:2021vpn ; Hamada:2021yxy ), and this has been borne out very concretely in high dimensions in the presence of supersymmetry. For example, in ten dimensions minimal supersymmetry and absence of local gauge and gravitational anomalies is enough to essentially fix the gauge group Green:1984sg and the short list of possibilities exactly matches those found in string theory, a phenomenon which has come to be known as string universality or the string lamppost principle . Ambitiously, one may hope to show that this continues all the way to four dimensions and in the absence of low-energy supersymmetry. Understanding string universality has since progressed, most successfully for d>6 and with supersymmetry Kim:2019ths ; Montero:2020icj ; Cvetic:2020kuw ; Hamada:2021bbz ; Bedroya:2021fbu . Continuing down to d=6 presents a new challenge since it is at this point that the minimal number of supercharges drops to eight and matter supermultiplets in arbitrary representations first appear. Much work has gone into understanding the subtleties of the consistency conditions for these models Kumar:2010ru ; Seiberg:2011dr ; Monnier:2018nfs ; Lee:2019skh ; Tarazi:2021duw ; Dierigl:2022zll ; Lee:2022swr ; Kim:2024tdh ; Kim:2024hxe ; Lockhart:2025lea . Recently in Kim:2024hxe it was argued by Kim, Vafa and Xu that the landscape of consistent 6d, \mathcal{N}=(1,0) supergravities with non-abelian gauge group is in fact finite. This is thanks to a sharp upper bound on the number of massless modes derived using the presence of a so-called H-string in the BPS spectrum. In particular, they find the bounds T\leq 193 and \rank G\leq 480 , both of which are saturated by known F-theory constructions; this provides additional evidence for string universality. A similar bound of T\leq 567 was also recently achieved by Birkar and Lee Birkar:2025rcg as a consequence of techniques developed to bound Hodge numbers of elliptic Calabi-Yau 3-folds. In the present work we give an argument for the finiteness of this 6d landscape which is complementary to both of these. Our upper bound on T will come from analyzing the structure of anomaly lattices subject to constraints coming from the absence of Dai-Freed anomalies; consequently our working assumptions are somewhat different than those of Kim:2024hxe and Birkar:2025rcg . Dai-Freed Dai:1994kq global anomalies arise when properly accounting for the fact that differential forms cannot fully capture all of the features of an (anti-)self-dual 2 -form gauge field. These anomalies are detected by placing the theory on a non-trivial curved background, such as the lens spaces that we consider in this work. In the context of 6d supergravity, Dai-Freed anomalies have recently been studied in Basile:2023zng ; Dierigl:2025rfn and shown to be quite discerning. The logical organization of our argument is summarized in figure 1 . After reviewing the basic structure of 6d supergravities and their local anomalies in section 2 , section 3 is dedicated to distilling the consequences of absence of Dai-Freed anomalies into a linear program which provides a strong lower bound on T in terms of other data. In section 4 the various bounds are brought together to prove finiteness and obtain upper bounds on T and the gauge group in progressively more restrictive settings. Finally, we conclude with a discussion in section 5 . Appendix A contains a proof of the uniqueness of (3;1^{T}) for the gravitational anomaly vector and appendices B and C contain data supporting the calculations of the main text. Finite landscape T\leq 3003 T\geq f^{\ast}(N) § 4.2 \Delta+29T\leq 273 if b_{i}^{2}\leq-1 , b_{i}^{0}=0 § 3.2 Dai-Freed anomalies § 3.1 b_{0}=(3;1^{T}) b_{0}\in\mathrm{I}_{1,T} or b_{0}/2\in\mathrm{I\!I}_{1,T} is primitive § A Finite number of models for fixed T String probes & anomaly inflow § 4.1 [\Gamma:\Gamma^{\mathrm{BPS}}]\leq C<\infty \exists\mathcal{J} with \mathcal{J}^{2}>0 and \mathcal{J}\cdot b_{i}>0 b_{i}^{2}\leq h(\Delta_{i}) Figure 1: Logical organization of the finiteness argument. Assumptions are in blue and the absence of local gauge anomalies is used implicitly throughout. 2 Preliminaries In this section we review the structure of 6d, \mathcal{N}=(1,0) supergravities and set our notation. First of all, with minimal supersymmetry there are four types of supermultiplets – gravity, tensor, vector and hyper – and a model \mathcal{M} amounts to a choice for the number of tensor multiplets T , gauge group G and charged (i.e. non-trivial) hypermultiplet representations \mathcal{H}_{\mathrm{ch}} . As we will review below, the number of hypermultiplets transforming in the trivial representation is determined by anomaly cancellation and therefore implicitly fixed by these data. The gauge group, G=\prod_{i}G_{i} , we take to be non-abelian and reserve i,j to index data corresponding to the simple factors. There are T+1 tensor fields, one in the gravity multiplet which is self-dual and one in each tensor multiplet which is anti-self-dual. We reserve \alpha,\beta\in\{0,1,\ldots,T\} to index these, with “ 0 ” always referring to the self-dual field. To each gauge factor is associated a vector b_{i} and there is a corresponding vector 1 1 1 The notation a\coloneqq-b_{0} is common. b_{0} associated with the gravity multiplet; we refer to these collectively using the indices I,J . These vectors live in a space of signature (1,T) with inner product \Omega which we abbreviate using u\cdot v\coloneqq u^{\alpha}\Omega_{\alpha\beta}v^{\beta} and u^{2}\coloneqq u\cdot u . Finally, the scalars in the tensor multiplets are written as \mathcal{J}\in\SO(1,T)/\SO(T) ; these parametrize the gauge kinetic terms \propto\mathcal{J}\cdot b_{i} , which should be non-negative to avoid ghosts. Local anomalies are captured by the 8 -form polynomial I_{8} which receives contributions from the chiral matter in all multiplets Alvarez-Gaume:1983ihn . In order to be canceled via the Green-Schwarz-West-Sagnotti (GSWS) mechanism, we must be able to write Green:1984sg ; Green:1984bx ; Sagnotti:1992qw I_{8}+\frac{1}{2}\Omega_{\alpha\beta}X_{4}^{\alpha}\wedge X_{4}^{\beta}\stackrel{{\scriptstyle!}}{{=}}0\,, (1) where the 4 -form X_{4} takes the following form: X_{4}^{\alpha}=-\frac{b_{0}^{\alpha}}{2}\tr\mathcal{R}^{2}+\frac{1}{2}\sum_{i}\frac{b_{i}^{\alpha}}{\lambda_{i}}\tr\mathcal{F}_{i}^{2}\,. (2) The normalization constants \lambda_{i} are listed in table 1 . Having added the appropriate couplings between B^{\alpha} and X_{4} in the action, the Bianchi identities read \mathrm{d}{H^{\alpha}}=X_{4}^{\alpha}\,. (3) Imposing the above cancellation requires that the irreducible terms ( \tr\mathcal{R}^{4} and \tr\mathcal{F}_{i}^{4} ) in I_{8} vanish, leading to the

Read the complete text