Carrollian Supersymmetry and SYK-like models

Oguzhan Kasikci,1, ∗ Mehmet Ozkan,1, † Yi Pang,2, ‡ and Utku Zorba1, §

1
Department of Physics, Istanbul Technical University, Maslak 34469 Istanbul, Turkey
2
Center for Joint Quantum Studies and Department of Physics,
School of Science, Tianjin University, Tianjin 300350, China
(Dated: November 2, 2023)
This work challenges the conventional notion that in spacetime dimension higher than one, a super-
symmetric Lagrangian invariably consists of purely bosonic terms, purely fermionic terms, as well as
boson-fermion mixing terms. By recasting a relativistic Lagrangian in terms of its non-relativistic
and ultra-relativistic sectors, we reveal that an ultra-relativistic (Carrollian) supersymmetric La-
grangian can exist without a purely bosonic contribution. Based on this result, we demonstrate
a link between higher-dimensional Carrollian and (0+1)-dimensional quantum mechanical models,
arXiv:2311.00039v1 [hep-th] 31 Oct 2023

yielding higher-order extensions of supersymmetric SYK models in which purely bosonic higher or-
der terms are absent. Given that supersymmetry plays an essential role in improving the quantum
behavior and solubility, our findings may lead to interesting applications in non-AdS holography.

Supersymmetry, as a symmetry under the exchange
of bosons and fermions, implies that typically a super-
symmetric theory decomposes into three parts: a purely
bosonic part, a purely fermionic part and a set of mixing
terms that couples the fermionic and the bosonic sec-
tors to each other. In many known examples, this state-
ment appears to hold generically in diverse dimensions
higher than one with any number of supercharges and
any number of derivatives. In this work, we question
this statement by decomposing a relativistic Lagrangian
into its Galilean (non-relativistic) and Carrollian (ultra-
relativistic) sectors. Our analysis underscores the piv-
otal role played by the non-relativistic sector organized
according to the powers of the speed of light. Most no-
tably, we show that it is possible for a supersymmetric
ultra-relativistic Lagrangian to exist without a strict re-
quirement for a purely bosonic part.
The ultra-relativistic corner of the Bronstein hyper-
cube has been under intense investigation in recent years,
expanding our knowledge beyond the standard frame-
work of relativistic physics. This endeavor has revitalized
Carrollian physics [1–3] in many areas, including dynam-
ics of null hypersurfaces [4–11], celestial holography and
conformal Carroll holography [12–18], condensed mat-
ter systems [19–23], hydrodynamics [24–27], Carrollian
gravity [28, 29] and black holes [30]. Based on the suc-
cessful experience of relativistic supersymmetry, Carroll
supersymmetry is expected to be a vital component of
various studies for a number of reasons. It improves the
quantum behavior of non-supersymmetric models and of-
fers more tools for solving a model. Nevertheless, it is
poorly understood at this stage as most studies focus on
two-derivative scalar field theories and their quantization
[22, 31–34].
Another perspective to better understand the Carroll
supersymmetry comes from the fact that a higher dimen-
sional Carrollian model without spatial derivatives can
be viewed as a collection of infinitely many quantum me-
chanical models, each labelled by its spatial coordinates.
Consequently, it appears that Carroll symmetry can also
emerge in quantum mechanical systems. From this view-
point, the (0 + 1)-dimensional Sachdev-Ye-Kitaev (SYK)
[35, 36] or SYK-like models [37–40] are naturally related
to Carrollian models [41] . As we will show, the lack of a
purely bosonic part simultaneously leads to the absence
of spatial derivatives in supersymmetric Carrollian mod-
els. Thus our result unveils another connection between
Carrollian models and quantum mechanical models. In
particular this means that one can obtain extensions of
the recently studied supersymmetric SYK-like models as
a by-product of higher-dimensional models.
We start our discussion by fixing our relativistic nota-
tion as xµ = ct, xi along with η = diag(− , + , . . . , +)


where η is the Minkowski metric. Assuming that in
a relativistic Lagrangian the speed of light c only ap-
pears via time derivatives (there are no other c-dependent
parameters), and consequently when writing out the c
dependence explicitly, the Lagrangian is of the form
L= N −n
P
n=0 c Ln , where Ln represents the Lagrangian
that is n-th order in time derivatives and N denotes the
highest number of time derivatives. Thus, a purely spa-
tial derivative part of a Lagrangian comes with no speed
of light factor, hence surviving in the c → ∞ limit, which
is the so-called Galilean limit of a relativistic theory.
On the other hand, the supersymmetry transforma-
tions δQ is first order in derivatives and is of the general
form
G 1 ′
δQ = δQ + δQ , (1)
c
where the superscript G refers to the Galillean part that
survives in the c → ∞ limit. The invariance under rela-
tivistic supersymmetry (up to total derivative) then im-
ply
1
N
X −1
1 G
δ′ L + ′


0 = δQ L = N +1 Q N
δQ Ln + δQ Ln+1
c cn+1
n=0
G
+δQ L0 . (2)
 2

′
This structure indicates that LN is invariant under δQ The L0 part of the action is non-vanishing as the c → ∞
G
while L0 is invariant under δQ , and transformations of limit only replaces the Lorentz indices with spatial in-
other Ln s ought to cancel with each other. As a concrete dices, i.e., L0 ∼ φ(∇2 φ∇2 φ − ∂i ∂j φ∂ i ∂ j φ) where ∇2 =
example, consider the relativistic Lagrangian for a 3D ∂ i ∂i . Note, however, that model has an identically van-
N = 1 scalar multiplet with the field content (φ, ψ, f ) ishing L4 as terms with highest time derivatives cancel
with each other. Thus, we conclude that while Ln for
1 2 1
LS = 2
φ̇ + ψ̄γ0 ψ̇ n > 0 might be identically vanishing, the Galilei model
2c 8c given by L0 is usually nonvanishing.
1 1 1
− ∂i φ∂ φ − ψ̄γ i ∂i ψ + f 2 ,
i
(3) The vanishing of the L4 in our example is an encour-
2 8 8 aging signal to work out the structure of the opposite
which is invariant under the supersymmetry transforma- end, which is the Carrollian supersymmetry arising in
tion rules the c → 0 limit of relativistic supersymmetry. From the
1 1 1 algebra (5), it suggest that we need to
√ rescale the super-
√
δf = ǭγ0 ψ̇ − ǭγ i ∂i ψ , δφ = ǭψ ,
2c 2 4 symmetry generators by a factor of c, i.e., Q → cQ
1 1 [44]. Furthermore, we need to rescale the fields with cer-
δψ = − γ0 φ̇ǫ + γ i ∂i φǫ − f ǫ . (4)
c 2 tain powers of c such that neither the transformation
Here, φ and f are real scalar fields while ψ is a two- rules, nor the Lagrangian diverges at c → 0 limit. For
component Majorana spinor. Obviously, the O(c−2 ) La- instance, for the 3D N = 1 scalar multiplet model (3),
grangian is invariant under δQ ′
transformation rules since the c-scaling can be assigned as follows
δQ φ = 0. Similarly, the O(c0 ) Lagrangian, which is the
′ √ √
φ → cφ , ψ → cψ , f →f, ǫ → cǫ , (8)
G
Galilean supersymmetric model, is invariant under δQ
transformations. Finally, note that in the c → ∞ limit, so that supersymmetry generator squares to the time
the time derivatives drop out from the transformation translation in the c → 0 limit. The resulting Lagrangian
rules which is expected since the supersymmetry algebra contains only time derivatives which is invariant under
is of the form the Carrollian boosts, i.e., xi → xi , t → t + bi xi . More
1 generically, after rescaling various fields and transforma-
{Q, Q} ∼ − γ0 H + γ i Pi , (5) tion rules, we have
c
indicating that in the Galilean limit, supercharges square N
X
to the spatial translations. Before finishing our discus- L = cn Ln , C
δQ = δQ ′
+ cδQ . (9)
sion on Galilean models, we point out another important n=0

property of the expansion (2). Writing out the O(c−1 )
term in the expansion of the Lagrangian, we obtain
N −1
1 X 1 G
δ′ L + ′


δQ L = N +1 Q N n+1
δQ Ln + δQ Ln+1
c n=1
c were O(c−N ) and possible O(c−n ) (n < N ) terms which
1 ′ G

G
+ δQ L0 + δQ L1 + δQ L0 . (6)
c Therefore, L0 would contain the time derivatives of the
This form of the Lagrangian indicates that if L0 vanishes
identically, then L1 becomes Galilean invariant. If that
happens, we multiply the entire Lagrangian with a fac-
tor of c before taking c → ∞ limit to obtain the Galilean
invariant model described by L1 . As suggested by the
example (3), the action L1 contains only fermions but
no bosons. Thus, the vanishing of L0 appears to indi-
cate that the Galilean supersymmetric model can exist
without referring to any purely bosonic terms. Neverthe-
less, as L0 contains only spatial derivatives but no time
derivatives, it seems unlikely to obtain an identically van-
ishing L0 . The reason is that a theory with purely spatial
derivatives merely comes from replacing the Lorentz in-
dices with spatial indices, hence the vanishing of L0 im-
plies the vanishing of the original relativistic model itself
[42]. To be concrete, consider the following example [43]
L ∼ φ(φφ − ∂µ ∂ν φ∂ µ ∂ ν φ) . (7)
Note that the terms with N -th order time derivative have
been rescaled properly so that the leading term in L
comes out as L0 , and the c → 0 limit yield a finite result.
The L0 consists of terms that, in the original Lagrangian
acquire the same powers of c after rescaling the fields.
bosonic fields as well as derivatives of fermions, and terms
involving auxiliary fields while its purely spatial deriva-
tive part appears at O(cN ). To see the structure of the
supersymmetry transformations, consider
1

∂ 1 1

i


Qα = √ + (γ0 θ)α ∂t − c γ θ α ∂i , (10)
c ∂ θ̄α 4 4
√
where we rescale θ → cθ. The structure of the super-
symmetry generator and the transformation parameter
√
ǫ, that is ǫ → cǫ, imply that the transformation rules
for a superfield Φ decomposes as
δQ Φ = [ǭQ, Φ] = ǭQC , Φ + c [ǭQ′ , Φ] ,
 
(11)
which explains the structure of the supersymmetry trans-
C ′
formations δQ = δQ + cδQ . Here, we assume that the
structure of a superfield is not deformed due to rescal-
ing the fields, and the entire superfield scales with some
 3

power of c. In the case of scalar multiplet, the rescaling given as O(c−4 ), identically vanishes as F00 = 0. Once
θ̄θ
√+ 2i f , is given by Φ → cΦ
of the superfield Φ = φ + iθ̄ψ the fields are rescaled in accordance with (8), the lowest-
according to (8) and θ → cθ. The invariance of the order composite expression for Fµν and ϕ are given by
Lagrangian now implies that
c1/2 0 1 1
N ϕ= (γ φ̇1 ψ2 − γ 0 φ̇2 ψ1 − f1 ψ2 + f2 ψ1 ) ,
C
X 8 2 2
C
cn δQ ′


0 = δL = δQ L0 + Ln + δQ Ln−1 1
n=1 F0i = (ψ̄2 γi ψ̇1 − ψ̄1 γi ψ̇2 ) , Fij = O(c) . (19)
8
+cN +1 δQ
′
LN . (12)
Most importantly, the lowest order F0i does not contain
As in the case of (7), the contribution to L0 from O(c−N ) any purely bosonic part as they are included in O(c) due
may be identically vanishing. In this case, the resulting to rescaling of the fields. Consequently, the Carroll limit
Carroll invariant models does not necessarily contain a gives rise to the following supersymmetric model consist-
purely bosonic term. Instead, it could consists of terms ing of four-fields interactions
that are either purely fermionic or fermions coupled to
derivative of bosons. 1
L4f = αbi bi + 2αλ̄γ0 λ̇ , (20)
As an illustrative example with a supersymmetric Car- 2
rollian action that does not contain a purely bosonic part, where
we consider the recently proposed spacetime subsystem
symmetric model [43] 1
bi = (ψ̄2 γi ψ̇1 − ψ̄1 γi ψ̇2 ) ,
8
1 2 1 2 α 1 1 1
Lsub = φ̇ + φ̇ + (φ̇1 ∂i φ2 − φ̇2 ∂i φ1 )2 , (13) λ = (γ 0 φ̇1 ψ2 − γ 0 φ̇2 ψ1 − f1 ψ2 + f2 ψ1 ) . (21)
2 1 2 2 2 8 2 2
whose relativistic origin is given by [45] This is an example of a supersymmetric model that con-
tains only purely fermionic part and mixing terms, but
1 1 α
L = − ∂µ φ1 ∂ µ φ1 − ∂µ φ2 ∂ µ φ2 − Fµν F µν , (14) no purely bosonic part. The supersymmetry of the model
2 2 4 can also be checked explicitly by noticing the transforma-
where tion rules for bi and λ are given by

Fµν = ∂µ φ1 ∂ν φ2 − ∂ν φ1 ∂µ φ2 . (15) 1 i
δbi = ǭγi λ̇ , δλ = b γi0 ǫ . (22)
4
The two-derivative part has a well-defined limit which
can be read off from (3) after rescaling according to (8) Upon using the transformation rules, the Lagrangian (20)
followed by the c → 0 limit. The four-derivative terms is invariant up to a total derivative term 12 ∂t (ǭγi λbi ).
can be supersymmetrized by considering the vector mul- As mentioned, the Carrollian supersymmetric model
tiplet action without a bosonic part (20) does not contain a spatial
derivative, which enables us to calculate a higher-order
1 correction to N = 2 SYK model [38, 40] by reducing
LV = − Fµν F µν − 2ϕ̄∂/ϕ . (16)
4 (20) to (0 + 1) dimensions. This is achieved by first gen-
eralizing our construction to N -number of scalar mul-
which is invariant under the transformation rules tiplets with a cubic potential term. In this case, the
1 µν two-derivative action is given by
δAµ = −ǭγµ ϕ , δϕ = γ Fµν ǫ , (17)
8
1 1 1
where Fµν = 2∂[µ Aν] . Based on the transformation rules LN S = − ∂µ φA ∂ µ φA − ψ̄ A ∂/ψ A + f A f A
2 8 8
of the scalar multiplet (4) and the vector multiplet, the 1
+CABC φA φB f C + φA ψ̄ B ψ C ,


fields of the vector multiplet can be realized as composites (23)
2
built from the scalar multiplet
where CABC is fully symmetric in its indices and A, B =
1 1 1 1, 2, . . . , N counts the number of multiplets. The higher-
ϕ = (∂/φ1 ψ2 − ∂/φ2 ψ1 − f1 ψ2 + f2 ψ1 ) ,
8 2 2 derivative part is still given by the vector multiplet action
1 (16), but with the following composite expressions for
Fµν = ∂µ φ1 ∂ν φ2 − ∂ν φ1 ∂µ φ2 + ψ̄1 γµ ∂ν ψ2
8 (Fµν , ϕ)
1 1 1
− ψ̄1 γν ∂µ ψ2 − ψ̄2 γµ ∂ν ψ1 + ψ̄2 γν ∂µ ψ1 . (18) 1 1
8 8 8 / A ψB − f AψB ,


ϕ = CAB ∂φ
8 2
These composite expressions indicate that the lowest- 1
A B
= CAB ∂µ φ ∂ν φ + ψ̄ A γ[µ ∂ν] ψ B ,


order Lagrangian, which contains only time-derivatives Fµν (24)
4
 4

where CAB is antisymmetric in its indices. Upon rescal- 1

δψ1A = −φ̇A ǫ2 − f A ǫ1 ,
ing the fields in accordance with (8) and CABC with a 2
factor of c−2 , and finally taking the c → 0 limit, we ob- A A 1 A
δψ2 = φ̇ ǫ1 − f ǫ2 ,
tain the Carrollian supersymmetric N scalar multiplet 2
model A i
ǫ1 ψ̇1 + ǫ2 ψ̇2A .
A


δf = (30)
2
1 1 1
LCN S = − φ̇A φ̇A + ψ̄ A γ0 ψ̇ A + f A f A The higher-derivative part of the Lagrangian (28), which
2 8 8
 1 A B C does not contain any purely bosonic terms, represents
A B C
+CABC φ φ f + φ ψ̄ ψ an example of an off-shell extension of the N = 2 SYK-
2
1 like model. It is worthwhile to mention that the higher
i
+ αbi b + 2αλ̄γ0 λ̇ . (25) derivative Lagrangian (20) resembles the N = 1 SYK
2
model for two vector multiplets with CABC = 0 stud-
Here, the composite expressions for bi and λ are given by ied in [37], if the composite objects (bi , λi ) are viewed
as fundamental fields equipped with the transformation
1 rules (22). Thus, our higher-derivative SYK-like action
bi = CAB ψ̄ A γi ψ̇ B ,
8 is in fact the sum of SYK (built out of composite fields)
1 1 and SYK-like models.
λi = CAB φ̇A (γ0 ψ B )i − CAB f A ψiB . (26)
8 2 In this work, we show that the Carrollian supersym-
This model is invariant under the following set of trans- metry has the curious feature that a supersymmetric La-
formation rules grangian does not require a purely bosonic part. By tak-
ing Carrollian limit of relativistic models we are able to
1 1 A provide a concrete supersymmetric example with such
δf A = ǭγ0 ψ̇ A , δφA = ǭψ ,
2 4 a feature. A better understanding might be possible
1 by constructing a Carrollian superspace in which case
δψ A = −γ0 φ̇A ǫ − f A ǫ . (27)
2 various supersymmetric results could easily be obtained
and their mathematical structure can be better studied.
Here, since the Lagrangian (25) does not contain any
As the 3D example given here does not contain spatial
spatial derivative, it can be viewed as a collection of in-
derivatives, it is straightforward to relate it to a SYK-
finitely many identical (0+1)-dimensional model labeled
like model with N = 2 off-shell supersymmetry extended
by (x, y). Choosing gamma-matrices as γ0 = iσ2 , γ1 =
by higher-order interactions which lacks a purely bosonic
σ1 , γ2 = σ3 , the (0 + 1)-dimensional model is given by
contribution.
1 1 i Another intriguing aspect is that, as we have shown,
LESYK = − φ̇A φ̇A + f A f A + ψ1A ψ̇1A + ψ2A ψ̇2A


2 8 8 the SYK-like model descending from a composite 3D
N = 1 supersymmetric vector multiplet action lacks the
+CABC φA φB f C + iφA ψ1B ψ2C


cubic interaction term characterized by CABC as λ̄γ · F λ
1 vanishes identically in D = 3 due to the symmetry of
+ α b21 + b22 + 4iλ1 λ̇1 + 4iλ2 λ̇2 ,


(28)
2 gamma matrices. However, such a term does exist in
which is a new 2-derivative extensions of N = 2 SYK-like D ≥ 4 [46], which indicates that along our procedure,
model. The subscript 1, 2 refer to the components of the their ultra-relativistic sector should lead to a complete
3D spinor, i.e., ψ = (ψ1 , ψ2 ) and we define SYK model based on the composite fields, with the cu-
bic interaction term. We would like to elaborate on this
i point in a future publication. Finally, it should also be
CAB ψ1A ψ̇1B − ψ2A ψ̇2B ,


b1 =
8 interesting to investigate effects of the higher order terms
i on various physical quantities in N = 2 model extending
= − CAB ψ1A ψ̇2B + ψ2A ψ̇1B ,


b2
8 earlier results on this subject [37, 39, 40, 47].
1 1 Acknowledgements.— We are grateful to E. Bergshoeff,
λ1 = CAB φ̇A ψ2B − CAB f A ψ1B , J. Distler, J. Rong and N. Su for useful communica-
8 2
1 1 tions. M.O. and U.Z. are supported in part by TUBITAK
λ2 = − CAB φ̇A ψ1B − CAB f A ψ1B . (29) grant 121F064. M.O. acknowledges the support by
8 2
the Outstanding Young Scientist Award of the Turk-
This model is invariant under the following set of off- ish Academy of Sciences (TUBA-GEBIP). The work of
shell transformation rules obtained by recasting the 3D Y.P. is supported in part by National Natural Science
transformation rules (27) in terms of (0+1)-dimensional Foundation of China (NSFC) under grant No. 12175164.
variables Y.P. also acknowledges support by the National Key
i Research and Development Program under grant No.
δφA = ǫ1 ψ2A − ǫ2 ψ1A ,


4 2022YFE0134300.

∗
kasikcio@itu.edu.tr
†
ozkanmehm@itu.edu.tr
‡
pangyi1@tju.edu.cn
§
zorba@itu.edu.tr
[1] J.-M. Lévy-Leblond, “Une nouvelle limite
non-relativiste du groupe de Poincaré,” Ann. Inst.
Henri Poincaré III 3 (1965) 1.
[2] J.-M. Lévy-Leblond, “Possible kinematics,” J. Math.
Phys. 9 (1968) 1605.
[3] M. Henneaux, “Geometry of Zero Signature
Space-times,” Bull. Soc. Math. Belg. 31 (1979) 47–63.
[4] C. Duval, G. W. Gibbons, and P. A. Horvathy,
“Conformal Carroll groups and BMS symmetry,”
Class. Quant. Grav. 31 (2014) 092001,
arXiv:1402.5894 [gr-qc].
[5] L. Donnay and C. Marteau, “Carrollian Physics at the
Black Hole Horizon,”
Class. Quant. Grav. 36 no. 16, (2019) 165002,
arXiv:1903.09654 [hep-th].
[6] L. Ciambelli, R. G. Leigh, C. Marteau, and P. M.
Petropoulos, “Carroll Structures, Null Geometry and
Conformal Isometries,”
Phys. Rev. D 100 no. 4, (2019) 046010,
arXiv:1905.02221 [hep-th].
[7] D. Grumiller, A. Pérez, M. M. Sheikh-Jabbari,
R. Troncoso, and C. Zwikel, “Spacetime structure near
generic horizons and soft hair,”
Phys. Rev. Lett. 124 no. 4, (2020) 041601,
arXiv:1908.09833 [hep-th].
[8] R. F. Penna, “Near-horizon Carroll symmetry and black
hole Love numbers,” arXiv:1812.05643 [hep-th].
[9] L. Freidel and P. Jai-akson, “Carrollian hydrodynamics
from symmetries,”
Class. Quant. Grav. 40 no. 5, (2023) 055009,
arXiv:2209.03328 [hep-th].
[10] J. Redondo-Yuste and L. Lehner, “Non-linear black hole
dynamics and Carrollian fluids,” JHEP 02 (2023) 240,
arXiv:2212.06175 [gr-qc].
[11] L. Marsot, P.-M. Zhang, and P. Horvathy, “Anyonic
spin-Hall effect on the black hole horizon,”
Phys. Rev. D 106 no. 12, (2022) L121503,
arXiv:2207.06302 [gr-qc].
[12] J. Hartong, “Holographic Reconstruction of 3D Flat
Space-Time,” JHEP 10 (2016) 104,
arXiv:1511.01387 [hep-th].
[13] A. Bagchi, R. Basu, A. Kakkar, and A. Mehra, “Flat
Holography: Aspects of the dual field theory,”
JHEP 12 (2016) 147, arXiv:1609.06203 [hep-th].
[14] L. Donnay, A. Fiorucci, Y. Herfray, and R. Ruzziconi,
“Carrollian Perspective on Celestial Holography,”
Phys. Rev. Lett. 129 no. 7, (2022) 071602,
arXiv:2202.04702 [hep-th].
[15] A. Bagchi, S. Banerjee, R. Basu, and S. Dutta,
“Scattering Amplitudes: Celestial and Carrollian,”
Phys. Rev. Lett. 128 no. 24, (2022) 241601,
arXiv:2202.08438 [hep-th].
[16] L. Donnay, A. Fiorucci, Y. Herfray, and R. Ruzziconi,
“Bridging Carrollian and celestial holography,”
Phys. Rev. D 107 no. 12, (2023) 126027,
arXiv:2212.12553 [hep-th].
[17] A. Bagchi, P. Dhivakar, and S. Dutta, “AdS Witten
5
diagrams to Carrollian correlators,”
JHEP 04 (2023) 135, arXiv:2303.07388 [hep-th].
[18] A. Saha, “Carrollian approach to 1 + 3D flat
holography,” JHEP 06 (2023) 051,
arXiv:2304.02696 [hep-th].
[19] L. Bidussi, J. Hartong, E. Have, J. Musaeus, and
S. Prohazka, “Fractons, dipole symmetries and curved
spacetime,” SciPost Phys. 12 no. 6, (2022) 205,
arXiv:2111.03668 [hep-th].
[20] L. Marsot, P. M. Zhang, M. Chernodub, and P. A.
Horvathy, “Hall effects in Carroll dynamics,”
Phys. Rept. 1028 (2023) 1–60,
arXiv:2212.02360 [hep-th].
[21] X. Huang, “A Chern-Simons theory for dipole
symmetry,” SciPost Phys. 15 (2023) 153,
arXiv:2305.02492 [cond-mat.str-el].
[22] J. Figueroa-O’Farrill, A. Pérez, and S. Prohazka,
“Quantum Carroll/fracton particles,”
JHEP 10 (2023) 041, arXiv:2307.05674 [hep-th].
[23] P. M. Zhang, H.-X. Zeng, and P. A. Horvathy,
“MultiCarroll dynamics,” arXiv:2306.07002 [gr-qc].
[24] J. de Boer, J. Hartong, N. A. Obers, W. Sybesma, and
S. Vandoren, “Perfect Fluids,”
SciPost Phys. 5 no. 1, (2018) 003,
arXiv:1710.04708 [hep-th].
[25] L. Ciambelli, C. Marteau, A. C. Petkou, P. M.
Petropoulos, and K. Siampos, “Covariant Galilean
versus Carrollian hydrodynamics from relativistic
fluids,” Class. Quant. Grav. 35 no. 16, (2018) 165001,
arXiv:1802.05286 [hep-th].
[26] L. Ciambelli, C. Marteau, A. C. Petkou, P. M.
Petropoulos, and K. Siampos, “Flat holography and
Carrollian fluids,” JHEP 07 (2018) 165,
arXiv:1802.06809 [hep-th].
[27] A. Bagchi, A. Banerjee, R. Basu, M. Islam, and
S. Mondal, “Magic fermions: Carroll and flat bands,”
JHEP 03 (2023) 227, arXiv:2211.11640 [hep-th].
[28] E. Bergshoeff, J. Gomis, B. Rollier, J. Rosseel, and
T. ter Veldhuis, “Carroll versus Galilei Gravity,”
JHEP 03 (2017) 165, arXiv:1701.06156 [hep-th].
[29] D. Hansen, N. A. Obers, G. Oling, and B. T. Sogaard,
“Carroll Expansion of General Relativity,”
SciPost Phys. 13 no. 3, (2022) 055,
arXiv:2112.12684 [hep-th].
[30] F. Ecker, D. Grumiller, J. Hartong, A. Pérez,
S. Prohazka, and R. Troncoso, “Carroll black holes,”
arXiv:2308.10947 [hep-th].
[31] B. Chen, R. Liu, H. Sun, and Y.-f. Zheng,
“Constructing Carrollian Field Theories from Null
Reduction,” arXiv:2301.06011 [hep-th].
[32] K. Banerjee, R. Basu, B. Krishnan, S. Maulik,
A. Mehra, and A. Ray, “One-loop quantum effects in
Carroll scalars,”
Phys. Rev. D 108 no. 8, (2023) 085022,
arXiv:2307.03901 [hep-th].
[33] J. de Boer, J. Hartong, N. A. Obers, W. Sybesma, and
S. Vandoren, “Carroll stories,” JHEP 09 (2023) 148,
arXiv:2307.06827 [hep-th].
[34] K. Koutrolikos and M. Najafizadeh, “Super Carrollian
and Super Galilean Field Theories,”
arXiv:2309.16786 [hep-th].
[35] S. Sachdev and J. Ye, “Gapless spin-fluid ground state
in a random quantum Heisenberg magnet,”
Phys. Rev. Lett. 70 no. 21, (May, 1993) 3339–3342,

arXiv:cond-mat/9212030 [cond-mat].
[36] A. Kitaev, “A simple model of quantum holography,”
http://online.kitp.ucsb.edu/online/entangled15/kitaev/;
//online.kitp.ucsb.edu/online/entangled15/kitaev2/.
Talks at KITP, April 7, 2015 and May 27, 2015 .
[37] W. Fu, D. Gaiotto, J. Maldacena, and S. Sachdev,
“Supersymmetric Sachdev-Ye-Kitaev models,”
Phys. Rev. D 95 no. 2, (2017) 026009,
arXiv:1610.08917 [hep-th]. [Addendum: Phys.Rev.D
95, 069904 (2017)].
[38] D. Anninos, T. Anous, and F. Denef, “Disordered
Quivers and Cold Horizons,” JHEP 12 (2016) 071,
arXiv:1603.00453 [hep-th].
[39] J. Murugan, D. Stanford, and E. Witten, “More on
Supersymmetric and 2d Analogs of the SYK Model,”
JHEP 08 (2017) 146, arXiv:1706.05362 [hep-th].
[40] A. Biggs, J. Maldacena, and V. Narovlansky, “A
supersymmetric SYK model with a curious low energy
behavior,” arXiv:2309.08818 [hep-th].
[41] This is different from higher dimensional generalization
of SYK model [48] where interactions between nearest
sides are turned on.
[42] Exceptional cases may arise when the indices of the
Levi-Civita tensor are contracted with derivatives, in
6
which case one ends up with a single term in the
Lagrangian with an explicit c-factor. Such models are
relativistic by themselves, and we exclude those in our
analysis.
[43] S. A. Baig, J. Distler, A. Karch, A. Raz, and H.-Y. Sun,
“Spacetime Subsystem Symmetries,”
arXiv:2303.15590 [hep-th].
[44] E. Bergshoeff, J. Gomis, and L. Parra, “The
Symmetries of the Carroll Superparticle,”
J. Phys. A 49 no. 18, (2016) 185402,
arXiv:1503.06083 [hep-th].
[45] O. Kasikci, M. Ozkan, and Y. Pang, “Carrollian origin
of spacetime subsystem symmetry,”
Phys. Rev. D 108 no. 4, (2023) 045020,
arXiv:2304.11331 [hep-th].
[46] D. Z. Freedman and A. Van Proeyen, Supergravity.
Cambridge Univ. Press, Cambridge, UK, 5, 2012.
[47] C. Peng, M. Spradlin, and A. Volovich, “Correlators in
the N = 2 Supersymmetric SYK Model,”
JHEP 10 (2017) 202, arXiv:1706.06078 [hep-th].
[48] W. Cai, X.-H. Ge, and G.-H. Yang, “Diffusion in higher
dimensional SYK model with complex fermions,”
JHEP 01 (2018) 076, arXiv:1711.07903 [hep-th].