a

r
X
i
v
:
0
8
0
9
.
5
2
3
4
v
4


[
h
e
p
-
t
h
]


1
8

J
a
n

2
0
0
9
AEI-2008-076
HU-EP-08/31
Imperial-TP-AT-2008-4
Structure of large spin expansion of anomalous dimensions
at strong coupling
M. Beccaria
a, 1
, V. Forini
b,2
, A. Tirziu
c,3
and A.A. Tseytlin
d,4
a
Physics Department, Salento University and INFN, 73100 Lecce, Italy
b
Humboldt-Universit¨ at zu Berlin, Institut f¨ ur Physik, D-12489 Berlin, Germany
c
Department of Physics, Purdue University, W. Lafayette, IN 47907-2036, USA
d
The Blackett Laboratory, Imperial College, London SW7 2AZ, U.K.
Abstract
The anomalous dimensions of planar N = 4 SYM theory operators like tr(ΦD
S
+
Φ) ex-
panded in large spin S have the asymptotics γ = f ln S + f
c
+
1
S
(f
11
ln S +f
10
) + ..., where
f (the universal scaling function or cusp anomaly), f
c
and f
mn
are given by power series in
the ‘t Hooft coupling λ. The subleading coefficients appear to be related by the so called
functional relation and parity (reciprocity) property of the function expressing γ in terms
of the conformal spin of the collinear group. Here we study the structure of such large
spin expansion at strong coupling via AdS/CFT, i.e. by using the dual description in terms
of folded spinning string in AdS
5
. The large spin expansion of the classical string energy
happens to have exactly the same structure as that of γ in the perturbative gauge theory.
Moreover, the functional relation and the reciprocity constraints on the coefficients are also
satisfied. We compute the leading string 1-loop corrections to the coefficients f
c
, f
11
, f
10
and
verify the functional/reciprocity relations at subleading
1

λ
order. This provides a strong
indication that these relations hold not only in weak coupling (gauge-theory) but also in
strong coupling (string-theory) perturbative expansions.
1
matteo.beccaria@le.infn.it
2
Recently moved to Albert-Einstein-Institut, MPG, Potsdam. forini@aei.mpg.de
3
atirziu@purdue.edu
4
Also at Lebedev Institute, Moscow. tseytlin@imperial.ac.uk
1 Introduction and summary
Recent advances in the study of duality between planar N = 4 SYM theory and free AdS
5
×S
5
superstring theory which utilise their integrability property led to important insights into the
structure of dependence of anomalous dimensions of gauge-invariant operators on the quantum
numbers like spin and on the t’Hooft coupling. While there was a remarkable recent progress in
understanding the asymptotic large spin limit in which the compactness of the spatial direction
of the world sheet may be ignored [1, 2],
1
it is important to study corrections to this limit.
Here we shall consider the famous example [6] of folded spinning string in AdS
5
dual to a
minimal twist gauge theory operator like tr(ΦD
S
+
Φ). Starting with the classical string energy
for the solution of [7, 6] and expanding it in large semiclassical spin parameter S one finds (see
[8] and below)
E =

λ E(S) , S =
S

λ
, (1.1)
E(S)
S≫1
= S +a
0
ln S +a
c
+
1
S
(a
11
ln S +a
10
)
+
1
S
2
(a
22
ln
2
S +a
21
ln S +a
20
) +O(
ln
3
S
S
3
) , (1.2)
with a
0
=
1
π
, a
c
=
1
π
(ln 8π −1), etc.
2
That means that in the semiclassical string theory limit
in which one first takes the string tension

λ

to be large for fixed S and then expands in large
S, the corresponding string energy can be written as (

λ ≫1,
S

λ
≫1)
E = S +f ln S +f
c
+
1
S
[f
11
ln S +f
10
]
+
1
S
2
[f
22
ln
2
S +f
21
ln S +f
20
] +O(
ln
3
S
S
3
) , (1.3)
where f =

λ a
0
+ .., f
c
=

λ a
c
+ ..., etc. The subleading coefficients simplify if we absorb
the constant f
c
into the ln S term, i.e. if we re-write (1.3) as
E = S +f ln(S/
˜
f
c
) +
1
S
[f
11
ln(S/
˜
f
c
) +f

10
]
+
1
S
2
[f
22
ln
2
(S/
˜
f
c
) +f

21
ln(S/
˜
f
c
) +f

20
] +O(
ln
3
S
S
3
) , (1.4)
where to leading order in
1

λ
expansion
f =

λ
π
,
˜
f
c
=
e

λ

, f
11
=
λ

2
, f

10
= 0, f
22
= −
λ
3/2

3
, f

21
=
5 λ
3/2
16π
3
, f

20
=
λ
3/2

3
1
Here we refer to the integral equations that describe the minimal anomalous dimension in the band [3, 4].
These equations were obtained from the all-loop Bethe ansatz by taking a special scaling limit [5, 3] which
describes a condensation of magnons and holes at the origin.
2
Note that the small S behaviour of the energy is quite different [9]: E =

2S[h0 + h1S + ...].
2
Following the analysis of quantum corrections to the folded string solution in [10], one may
conclude that this structure of the large S expansion is preserved by the α


1

λ
corrections,
with the coefficients f, f
c
, f
11
, ... being promoted to power series in
1

λ
, i.e. f
mk

¸
n
b
mk,n
(

λ)
n
.
Indeed, as we shall find below, the 1-loop corrections for leading coefficients in (1.4) are
3
f =

λ
π

1 −
3 ln 2

λ
+O(
1
λ
)

, (1.5)
˜
f
c
=
e

λ

1 +
1

λ
(3 ln 2 −c) +O(
1
λ
)

, (1.6)
f
11
=
λ

2

1 −
6 ln 2

λ
+O(
1
λ
)

, (1.7)
f

10
=
λ

2

0 −
0

λ
+O(
1
λ
)

. (1.8)
Here c is a constant that we were not able to determine with the method for evaluation of 1-loop
string correction we used below. Equivalently, in (1.3) we get the same f, f
11
and
f
c
= f ln
1
˜
f
c
=

λ
π

ln


λ
−1 +
1

λ

−3 ln 2 ln


λ
+ c

+O(
1
λ
)

, (1.9)
f
10
= f

10
+f
11
ln
1
˜
f
c
= f

10
+
f
c
f
f
11
(1.10)
=
λ

2

ln


λ
−1 +
1

λ

−6 ln 2

ln


λ

1
2
] + c

+O(
1
λ
)

. (1.11)
The (at first surprising) vanishing of the first two terms in f

10
in (1.8) is a consequence of the
relation f
10

fc
f
f
11
= 0 (here verified at first two leading orders, i.e. to order O(λ
0
) in f
10
).
In fact, since we also see that in (1.5),(1.7) f
11
=
1
2
f
2
, this relation is equivalent to f
10

1
2
f
c
f =
0. As we shall see below, these relations are consequences of the “functional relation” and
reciprocity at strong coupling. Note that these conditions thus determine 2 our of 4 coefficients
in the part of E up to order O(
ln
2
S
S
2
). We thus led to expect that f

10
= 0 should be true in to
all orders in the strong coupling expansion, suggesting the advantage of the form of E in (1.4)
over (1.3) and the importance of the function
˜
f
c
.
Reversing the usual logic, we may then conjecture that structurally same large spin expansion
should appear also at weak coupling, i.e. in the perturbative expressions for the corresponding
gauge theory anomalous dimensions. This is not, a priori, guaranteed since the limit taken on
the gauge theory side is different from the above string-theory limit: there one first expands the
anomalous dimension in small λ at fixed S and then takes S large in each of the λ
n
coefficients.
Yet, remarkably, expanding in large S the known 2-, 3- and 4-loop perturbative anomalous
dimensions of twist 2 and twist 3 operators in SYM theory one does find [11, 12, 8, 13, 15,
3
The leading correction to f was found in [10].
3
16, 17, 18] the expression of the form (1.3) with the coefficients given by power series in λ, i.e.
f
mk

¸
n
a
mk,n
λ
n
.
4
Assuming that the expansion (1.3) or
5
E −S =

¸
m=0
e
m
(λ, ln S)
S
m
, e
m
(λ, ln S) =
¸
k
f
mk
(λ) ln
k
S (1.12)
applies for any λ, and given the important role of the universal scaling function or cusp anomalous
dimension f(λ) [19], one may raise the question about the interpretation of other “interpolating”
functions f
mk
(λ) in (1.3).
From the gauge theory point of view, the function f(λ) appears in the asymptotics of anoma-
lous dimensions of gauge invariant operators as well as (for twist 2) in the IR asymptotics of
gluon scattering amplitudes related to UV cusp anomaly of light-like Wilson loops (for a review
and references see, e.g., [20]). On string side that corresponds, respectively, to the closed string
[6] and the open string [21, 22] sectors. They are connected in the strict large S limit since then
the ends of the folded spinning string reach the boundary of AdS
5
and thus the associated world
surface has a Wilson line interpretation [23]. This open string sector interpretation should not
be expected to apply to other subleading coefficients f
mk
(λ) since for finite S the end points of
the folded string no longer touch the boundary (cf., however, [24, 25] where the gauge theory
interpretation of the constant f
c
is discussed).
In fact, many of the f
mk
coefficients in (1.3),(1.12) are not actually independent, as was first
observed at few leading orders in weak coupling expansion and then given a general interpretation
in [15, 16]. According to [15], these coefficients are constrained by (i) the so called “functional
relation” suggested by the conformal invariance (which relates the leading f
mm
functions to
powers of the scaling function f and thus implies their universality) and also by (ii) the “parity
preserving relation” or “reciprocity” [12, 15, 16, 26, 18] (which relates some subleading non-
universal coefficients, e.g., f
10
to f
c
, f
32
to f
21
, etc.).
Our aim here will be to investigate the presence of such relations at strong coupling, i.e.
in the semiclassical string theory expansion for the spinning string states, extending earlier
observations made in [15].
6
4
The 4-loop prediction for twist 2 and 3 anomalous dimension at finite S [17] so far was not based on direct
gauge theory computation.
5
Here, as in (1.2),(1.3), we suppress the dependence on finite twist J.
6
Let us note also that the fine structure of the constant term fc in (1.3) for “non-minimal” operators and the
dual string states was studied in [3, 4] and in [27].
4
Before proceeding let us add an important clarification. While for low twists J = 2, 3 in
weak-coupling expansion the powers of ln S in e
m
in the anomalous dimension (1.12) appear
to be positive,
7
for J > 3 one finds also terms with negative powers like
J
k
ln
m
S
. Namely, for
J > 3 the leading term in E − S − J appears to be e
0
= k
1
(J) +
k
2
(J)
ln
2
S
+ ..., where for k
1
=
k
11
J + k
10
, k
2
= k
23
J
3
+ k
22
J
2
+ k
21
J + k
20
and k
2
vanishes for J = 2, 3.
8
The presence of
such terms was first observed in [3] (in the 1-loop approximation in the sl(2) sector) in the limit
of large J and S with j ≡
J
lnS
fixed and small, i.e. e
0
= (k
11
j + k
23
j
3
+ ...) ln S, and they are
likely to be present also for finite J with S ≫ 1.
9
At strong coupling, i.e. in the string-theory
semiclassical expansion, one cannot distinguish between finite values of J and J = 0; for large
J one finds (see [3] and section 2.2 below) that in a similar limit of large S with ℓ ≡
J

λln S
fixed e
0
= (n
1

2
+ n
2

4
+ ...) ln S. Thus the dependence on J and S is different at strong and
weak coupling and to relate the two expansions one needs a non-trivial resummation rather than
simple interpolation of coefficients in λ [40, 2].
Let us now review in more detail what is known at weak coupling (see [15] and references
there). The results of explicit higher-loop planar gauge-theory computations of anomalous di-
mensions γ(S, J, λ) of operators like tr(D
S
+
Φ
J
) (S is the Lorentz spin and J is the twist) were
interpreted in [15] in the following way (see also [16]). Observing that such Wilson-type op-
erators can be classified according to representations of the collinear SL(2, R) subgroup of the
SO(2, 4) conformal group [30] which are labeled by the conformal spin s =
1
2
(S + ∆) one may
argue that the anomalous dimension γ = ∆−S −J should be a function of S only through its
dependence on the conformal spin s. Since the scaling dimension ∆ is
10
∆ = S +J +γ(S, J) , (1.13)
that then leads to the following “functional relation” for γ
γ(S, J) = f(s; J) = f

S +
1
2
J +
1
2
γ(S, J); J

. (1.14)
Without further information, this relation is nothing more than a change of variable, since, at
least in perturbation theory, it is always possible to compute the function f in terms of the
7
Here we have in mind the minimal anomalous dimension in the band [3]; the
J
3
ln
2
S
terms appear [28] in
non-minimal dimensions even for low twists. We thank G. Korchemsky for this clarification.
8
We thank A. Rej for a discussion of the structure of these terms at 1-loop order in sl(2) sector (see also [3]).
9
There is a numerical evidence for the presence of
k
2
(J)
ln
2
S
term for J ≥ 4 from the analysis of the corresponding
1-loop Baxter equation (A. Rej, private communication). For a general method to derive higher order terms in
1/S expansion at fixed J see [29]. Let us mention also that coefficients in large S expansion beyond cusp anomaly
one may be also controlled by integral equations like the BES [1] one (see in this connection [54]).
10
Here we assume that Φ in the operator tr(D
S
+
Φ
J
) is a scalar field (as is the case in the sl(2) sector of SYM
theory). The relation between the notation used in [15] and ours is: N →S, L →J, J →C and j →s.
5
anomalous dimension γ(S, J). Nevertheless, the above reasoning suggests that f could be more
fundamental than γ.
11
This is what we shall assume below when referring to the functional
relation.
Suppressing the dependence on J in γ and f we may write this functional relation simply as
12
γ(S) = f

S +
1
2
γ(S)

. (1.15)
At weak coupling γ(S) =
¸

n=1
γ
n
(S)λ
n
; expanding the coefficients γ
n
in large S (for fixed
J) one finds that for all explicitly known perturbative gauge-theory results one gets the same
expansion as in (1.3)
γ(S)
S≫1
= f ln S +f
c
+
f
11
ln S +f
10
S
+
f
22
ln
2
S +f
21
ln S +f
20
S
2
+
+
f
33
ln
3
S +f
32
ln
2
S +f
31
ln S +f
30
S
3
+O

ln
4
S
S
4

, (1.16)
where the coefficients f, f
c
, f
11
, ... are power series in λ. Remarkably, the structure of this
expansion turns out to be perfectly consistent with the functional relation (1.15): the function f
starts with a logarithmic term (and is “simpler” than f, i.e. has no
ln
n
S
S
n
terms in it) so that the
coefficients of the leading
ln
m
S
S
m
terms are all determined by the scaling function f [15, 26, 4]
γ(S) = f ln

S +
1
2
f ln S +...

+... = f ln S +
f
2
2
ln S
S

f
3
8
ln
2
S
S
2
+
f
4
24
ln
3
S
S
3
+... . (1.17)
The universality (i.e. twist and flavor independence) of the scaling function or cusp anomalous
dimension f thus implies the universality of all of the coefficients f
mm
in (1.16) as they are
simply proportional to f
m+1
,
f
11
=
1
2
f
2
, f
22
= −
1
8
f
3
, f
33
=
1
24
f
4
, ... . (1.18)
These should be understood as relations between the functions f
mm
(λ) and f(λ) defined as
power series in λ.
Let us note that anomalous dimensions of operators with twist higher than two occupy a
band [3], the lower bound of which is the minimal dimension for given S and J. The relation
11
Apart from conformal invariance, for twist 2 case, this conclusion is also expected on the basis of the QCD
origin of the functional relation (1.14) [13]. In that context, the function f for twist 2 operators is closely related to a
special reformulation of the parton distribution functions evolution equation which aims at treating symmetrically
the space-like channel of deep inelastic scattering and the time-like crossed channel describing e
+
e

annihilation
[14]. In particular, the functional relation (1.14) turns out to be predictive within the large S expansion (1.12)
because the function f happens to be simpler than γ and, in particular, it does not contain ln
m
S/S
m
terms
[16, 15, 26].
12
In the approach of [13, 16], which recently received a nice confirmation in [31], this relation follows from a
suitable modification of the evolution equations governing the renormalization of the twist operators.
6
(1.17) is expected to apply for the minimal dimension in the band. Interestingly, as was found
at weak coupling, a similar relation also holds for the excited trajectories [4]. This is also what
we shall see at strong coupling on the example of the spiky string in section 2.3 (see (2.32)).
In Appendix F below we will summarize the known perturbative expansions for the minimal
anomalous dimensions of twist 2 and twist 3 operators of various flavors, as obtained from the
asymptotic Bethe ansatz of [32]. These results are indeed consistent with (1.17),(1.18) and thus
with the universality of the f
mm
coefficients in (1.16).
As for the subleading (
ln
k
S
S
m
, k < m) terms in (1.16), their coefficients are, at least partially,
controlled by special properties of the function f in the functional relation (1.15). Indeed, it was
observed on many examples that the function f should satisfy a “parity preserving relation” or
reciprocity property.
13
This property implies that the large S expansion of f(S) should run in
the inverse even powers of the quadratic Casimir of the collinear SL(2, R) group, namely, [15]
f(S) =

¸
n=0
a
n
(ln C)
C
2n
, (1.19)
where C is the “bare” quadratic Casimir defined in terms of the “canonical” value of the con-
formal spin s
0
as C
2
≡ s
0
(s
0
−1), s
0
=
1
2
(S + ∆
0
) = S +
1
2
J , i.e.
14
C
2
= (S +
1
2
J)(S +
1
2
J −1) . (1.20)
The reciprocity property (1.19) of the function f in the relation (1.15) then imposes constraints
on some of the coefficients of the subleading terms in the expansion (1.16):
f
10
=
1
2
f (f
c
−1 +J) , f
32
=
1
16
f [f
3
−2f
2
(f
c
−1 +J) −16f
21
] , ... (1.21)
where dots stand for similar expressions for f
31
, f
30
, f
55
, ..., f
50
, etc.
15
Again, these equations
relate functions f
mk
(λ) defined as power series in λ. For twist J = 2 we get simply
f
10
=
1
2
f(f
c
+ 1) , f
32
=
1
16
f [f
3
−2f
2
(f
c
+ 1) −16f
21
] , ... . (1.22)
13
The name reciprocity has its origin in the the formulation of this property, for twist 2 case, in terms of
the Mellin transform: F(x) = −xF(1/x), where f(S) =
R
1
0
dxx
S−1
F(x). For twist larger than 2 the “parity
preserving relation” was suggested in [15] as a more appropriate name. Here for simplicity we shall not make
this distinction and will often refer to reciprocity when implying the “parity invariance” property in (1.19),(1.20)
below.
14
Here we again consider the operator built of scalar fields. For twist 2, i.e. J = 2, one then has C
2
= S(S +1),
so that C = S +
1
2
+ O(
1
S
). For generic flavour one is to replace, in (1.20), S +
1
2
J → S + ℓJ, where ℓ =
1
2
, 1,
3
2
for a scalar, spinor or vector cases [30].
15
Note that using this expression for f10 and defining
˜
fc = e
−fc/f
to put (1.16) into the form (1.4) one finds
that f

10
=
1
2
f (−1 + J) (which, at strong coupling, is subleading to f11 term unless J ∼

λ).
7
These so called MVV [12] relations were first observed for twist 2 QCD anomalous dimensions
up to 3 loops. The large S expansions for the known twist 2 and twist 3 SYM anomalous
dimensions that we will present in Appendix F are indeed consistent with these relations, i.e.
with the reciprocity property of the function f.
16
It is natural to expect that the functional relation and the reciprocity property should hold
also at higher orders in small λ expansion. Since the planar perturbation theory should be
convergent, they should then also be visible at strong coupling [15], i.e. in the large spin
expansion of the corresponding semiclassical string energies.
One may also wonder if the reciprocity property may apply to higher twist operators above
the lower bound of the band [3, 4]. If that were the case, it could then be checked also at strong
coupling on the example of the spiky string solution of [34].
The agreement in the structure of the large S expansion found in perturbative gauge theory
and in perturbative string theory is already quite remarkable. This agreement is non-trivial
since, as was already mentioned, the gauge-theory and string-theory perturbative expansions
are organized differently: the gauge-theory limit is to expand in small λ at fixed S and then
expand the λ
n
coefficients in large S, while the semiclassical string-theory limit is to expand in
large λ with fixed S =
S

λ
and then expand the
1
(

λ)
n
terms in E in large S. Even assuming these
limits commute (which so far appears to be verified only for the leading universal ln S term) the
reason for the validity of the functional relation (1.15) and, moreover, of the reciprocity property
(1.19) is obscure on the semiclassical string theory side.
The functional relation (1.15) for the anomalous dimensions of Wilson-type operators on the
gauge theory side was argued [15] to follow from the invariance under the collinear SL(2, R)
subgroup of the conformal SO(2, 4) group. Given that this argument is based on the conformal
symmetry, one may think that it should then apply also on the string theory side. However,
as we will review in Appendix A, the realization of the conformal group on states represented
by classical spinning string solutions in global AdS
5
coordinates is a priori different from the
one used on the gauge-theory side (which is based on the collinear subgroup), so that the direct
connection is not obvious. The reason for the reciprocity property on the string theory side is
even far less clear.
16
Three-loop tests of reciprocity for QCD and for the universal twist 2 supermultiplet in N = 4 SYM were
discussed in [15, 16]. A four-loop test for the twist 3 anomalous dimension in the sl(2) sector was performed
in [26]. The case of twist 3 gauge field strength operators was analyzed in [33] (at three loops) and in [18] (at
four loops). In the latter paper it was also proved that even the wrapping-affected four loop result for the twist
two operators [17] is reciprocity respecting.
8
If one identifies the energy E and the spin S of a string rotating in a plane in global AdS
5
with dimension and Lorentz spin of the gauge theory operator like tr(D
S
+
Φ
J
), the functional
relation (1.15) would then imply that γ = E −S −J should be a function of s =
1
2
(E +S), i.e.
E −S −J = f(E +S, J) . (1.23)
As we shall discuss below (extending earlier observations in [8, 15]), not only the structure of
the large spin expansion on the string theory side happens to be the same as on the gauge
theory side but also its coefficients are indeed consistent with the functional relation and the
reciprocity for the minimal dimension case represented by the folded spinning string. This will
be demonstrated at the classical as well as 1-loop string theory level.
We shall also show that the functional relation but not the reciprocity appears to apply also
to the case of the classical spiky string solution.
The rest of this paper is organized as follows.
In section 2 we shall first consider the large spin expansion of the classical energy of folded
spinning string in AdS
5
and show that the large spin expansion has the structure (1.2) and the
functional and reciprocity relations between the coefficients are satisfied. We shall then include
(in section 2.2) the dependence on the angular momentum J in S
5
in the “long string” limit
(J ≪S). In section 2.3 we shall study the same large spin expansion for a spiky string in AdS
5
;
in this case we shall find that the reciprocity condition is violated which should be related to
the fact that the corresponding operator has higher than minimal dimension for a given spin.
In section 3 we shall return to the case of the folded spinning string in AdS
5
(i.e. assume that
J is negligible compared to S) and compute the 1-loop correction to the energy expanded in
large S, determining corrections to several leading coefficients. As result, we shall verify that the
string 1-loop corrections preserve the structure (1.3) of the large spin expansion and, moreover,
that the reciprocity condition is satisfied beyond the string tree level.
In Appendix A we shall make some comments on relation between different realizations of
conformal group. In Appendix B and C we shall review the folded spinning string solution
and discuss long-string or large-spin expansions used in the 1-loop computation in section 3. In
Appendix D we shall give details of large spin expansions for (S, J) string considered in section 2.
In Appendix E we shall discuss some consequences of the functional relation and the reciprocity
at strong coupling, pointing out a subtlety in the definition of the latter in the semiclassical
string expansion. In Appendix F we shall summarize the known weak coupling planar SYM
results for the large spin expansion of twist 2 and 3 anomalous dimensions up to 4-loop order
in the ‘t Hooft coupling.
9
2 Large spin expansion: classical string theory
2.1 Folded spinning string with J = 0
We shall start with a discussion of the limit when the S
5
momentum J of the string state can
be ignored, i.e. we shall concentrate only on the AdS
5
spin S dependence of the string energy.
This is the limit when the twist of the gauge theory operator is sufficiently small compared to
the Lorentz spin.
We review the folded spinning string solution [6] in Appendix B. The integrals of motion are
the energy E =

λ E and the spin S =

λS, which can be expressed in terms of the elliptic
functions E and K of an auxiliary variable η
17
E −S =
2
π

1 +η
η
¸
E


1
η

1

1 +η
−1

+K


1
η

, (2.1)
S =
2
π

1 +η
η
¸
E


1
η

−K


1
η

. (2.2)
To find the energy in terms of the spin one is to solve for η. Here we are interested in the
large spin expansion which corresponds to the long string limit (when the string ends are close
to the boundary of AdS
5
). For such long string one has η → 0. Solving (2.2) for small η and
substituting it into (2.1) one finds for E as a function of S
E = S +
ln
¯
S −1
π
+
ln
¯
S −1
2 π
2
S

2 ln
2
¯
S −9 ln
¯
S + 5
16 π
3
S
2
+
2 ln
3
¯
S −18 ln
2
¯
S + 33 ln
¯
S −14
48 π
4
S
3
+... ,
¯
S ≡ 8 π S , (2.3)
as was already claimed in (1.2).
The functional relation (1.15),(1.23) implies that E − S should be a function of E + S. It
is not immediately obvious from (2.2) (or from the form of the exact solution in global AdS
5
coordinates) why such a relation should be natural for any value of S. Still, the coefficients of
the leading (
ln S
S
)
m
terms in (2.3) happen, indeed, to be consistent with such a relation, with
the leading term in the function f being simply the logarithm (cf. (1.17))
E −S =

λ
π
ln

S +
1
2

λ
π
ln S +...

+... . (2.4)
Furthermore, it is possible to verify that the expansion of E − S also satisfies the reciprocity
property (1.19),(1.22). The large S expansion of the function f (its leading term in the strong-
coupling limit) is much simpler than that of the anomalous dimension E−S in (2.1) and contains
17
Here we follow the notation of [35]. Equivalently, one can express the conserved charges in terms of the
hypergeometric functions as in Appendix B.
10
only even powers of C
−1
∼ S
−1
(see (1.20))
f =

λ
˜
f ,
˜
f(S) =
1
π

ln
¯
S −1 +
ln
¯
S + 1
16π
2
S
2
+O

1
S
4

+O(
1

λ
) . (2.5)
Equivalently, we find that the MVV-like relations (1.22) are satisfied.
18
A more systematic analysis of the reciprocity (parity invariance) property of the function f
is possible with the help of an integral representation for it. Using that (1.15) implies
˜
f(S

) =
˜ γ

S


1
2
˜
f(S

)

, where S

= S +
1
2
˜ γ(S), ˜ γ(S) = E −S, and renaming S

→S we have
˜
f(S) =
1
2π i

Γ
dω ˜ γ(ω)
1 +
1
2
˜ γ

(ω)
ω −S +
1
2
˜ γ(ω)
, (2.6)
where the contour Γ encircles the pole of the integrand and prime stands for derivative.
19
It
is natural to replace the variable ω in (2.6) with the expression (2.2) for the semiclassical spin
S(η)
˜
f(S) =
1
2π i

Γ
dη ˜ γ(η)
˜ s

(η)
˜ s(η) −S
, (2.7)
where ˜ s(η) ≡ S(η) +
1
2
˜ γ(η) =
1
2
(E + S) is the “conformal spin” expressed in terms of the
semiclassical quantities. The integral then gives the function ˜ γ evaluated at the zero of the
denominator; this is the same as the statement that the anomalous dimension as a function of
the Lorentz spin is, effectively, a function of the conformal spin ˜ s.
To verify the reciprocity property of the function
˜
f(S) in (2.7) it is useful to redefine the
variable η as
20
η → −1 + 16η +

1 + 256 η
2
and examine the large S or small η limit of the
expressions. One finds that ˜ γ(η) is a series in even powers of η
˜ γ(η) = −
1 + ln η
π
+
4(ln η + 12)
π
η
2

6(62 ln η + 777)
π
η
4
+... , (2.8)
while the expression for the conformal spin runs in odd powers of η
˜ s(η) =
1
8πη
+
11 + 2 ln η

η −
877 + 92 ln η

η
3
+... . (2.9)
From the equation for the pole of the integrand in (2.7), ˜ s −S = 0, one can find the parameter
η in terms of the spin S, concluding that it is given by a power series in odd negative powers of
18
The definition of reciprocity condition in string semiclassical expansion is discussed in Appendix E.
19
The expression that multiplies ˜ γ in the integrand has residue 1, so that the integral is ˜ γ evaluated at the
pole ω = S −
1
2
˜ γ. Then defining x = S −
1
2
˜
f(S) we have 2S − 2x = ˜ γ which coincides with the equation for
the pole with x = ω. Note that assuming f exists, one can formally reconstruct it from γ using [15] f(S) =
P

k=1
1
k!
`

1
2
∂S
´
k−1
[γ(S)]
k
= γ −
1
4

2
)

+
1
24

3
)
′′
+· · · . This relation also arises by expanding the denominator
in (2.6) in small ˜ γ and integrating the resulting series.
20
This choice is not unique. An analogous transformation was used in [15].
11
S. As a result,
˜
f(S), which is same as ˜ γ(η) evaluated at the pole, should also run only in even
negative powers of S or C =
C

λ
(cf. (1.20)).
The above discussion has a straightforward generalization to the multifolded spinning string
case. The leading terms in the large S expansion of the energy of a string with m folds are (see
Appendix D)
E −S =
m
π

ln
¯
S −1 +
4
¯
S
(ln
¯
S −1) −
4
¯
S
2
(2 ln
2
¯
S −9 ln
¯
S + 5) +...

,
¯
S ≡

m
S . (2.10)
In this case it is possible to show again that the large S expansion is consistent with the
reciprocity property.
2.2 Folded spinning string with J = 0
Let us now consider the case when the S
5
angular momentum of the string is not negligible
compared to S, i.e. when the string state is dual to an operator with large spin S and large
twist J. The corresponding charges are the energy E =

λE and the two angular momenta
S =

λS and J =

λJ [10, 35]:
E = κ +
κ
ω
S ,
ω
2
−J
2
κ
2
−J
2
≡ 1 +η , (2.11)

κ
2
−J
2
=
2
π

η
K


1
η

, S =


ηω

κ
2
−J
2

E


1
η

−K


1
η

(2.12)
Here κ and ω (or η) are parameters of the classical solution which should we eliminated to find
E as a function of S and J.
We will be interested in large S expansion with S ≫J since only in this case the expansions
like (1.16),(1.19), i.e. going in the inverse powers of S with the coefficients being polynomials
in ln S, will apply (see also [10, 15]).
In the large S ≫J or long string limit, when η ≪1, one should distinguish between “small”
or “large” J cases [10, 36]. In the “slow long string” approximation (corresponding to taking
S to be large with ℓ ≡
J
ln S
fixed and then expanding in powers of ℓ) the leading terms in the
semiclassical energy read (cf. (2.3))
E −S −J ≈
1
π
(ln
¯
S −1) +
π J
2
2 ln
¯
S

π
3
J
4
8 ln
3
¯
S

1 −
1
ln
¯
S

+... (2.13)
+
4
¯
S

1
π
(ln
¯
S −1) +
π J
2
2 ln
2
¯
S


3
J
4
4 ln
4
¯
S

1 −
2
3 ln
¯
S

+...


4
¯
S
2

1
π
(2 ln
2
¯
S −9 ln
¯
S + 5) +π J
2

1 +
3
2 ln
¯
S

1
ln
2
¯
S

2
ln
3
¯
S

+...

12
where
¯
S ≡ 8πS, and dots stand for higher order corrections depending on J.
21
In the case of “fast long string”, when ln S ≪J ≪S, the corrections to the energy read
E −S −J ≈
1
π
2
J

1
2
ln
2
ˆ
S −ln
ˆ
S +
4 ln
ˆ
S
ˆ
S
+
4
ˆ
S
2

−2 ln
ˆ
S + 1 +
3
ln
ˆ
S
+
2
ln
2
ˆ
S
+...

+...

+
1
π
4
J
3


ln
4
ˆ
S
8

2
ˆ
S

3 ln
2
ˆ
S + ln
ˆ
S + 1 +
1
ln
ˆ
S
+
1
ln
2
ˆ
S
+...

(2.14)

2
ˆ
S
2

2 ln
3
ˆ
S −19 ln
2
ˆ
S + 11 ln
ˆ
S + 13 +
13
ln
ˆ
S
+
11
ln
2
ˆ
S
+...

+...

where
ˆ
S ≡
8S
J
=
8S
J
≫1. Dots in the square brackets indicate corrections in 1/
ˆ
S, corrections in
1/ ln
ˆ
S can be added in the round brackets and terms like ln(ln
ˆ
S) have been neglected.
The leading terms here can be summed up as [3]
E −S =

J
2
+
1
π
2
ln
2
8S
J
+... , (2.15)
where
lnS
J
≪1 plays the role of an expansion parameter.
Notice that in contrast to the slow long string case where the expansion (2.13) has the same
structure as in (1.12), in the fast long string case (2.14) we get higher powers of ln S not
suppressed by S, and so this case (cf. also its discussion in Appendix D) is somewhat outside
our main theme here.
To study the properties of the subleading corrections, one may again make use of the integral
representation for the functional relation as in (2.6). The discussion will apply to both the “slow”
and the “fast” long string limits. Here the “conformal spin” is ˜ s =
1
2
(S + E) = S +
1
2
J +
1
2
˜ γ,
while the “semiclassical” value of the Casimir operator in (1.20) is C =
C

λ
≈ S +
1
2
J. Then the
integral in (2.7) can be written as
˜
f(C) =
1
2π i

Γ
dη ˜ γ(η)
˜ s

(η)
˜ s(η) −C
, ˜ s(η) = S(η) +
1
2
˜ γ(η) . (2.16)
After a redefinition of η one can then show that the expansion of f in large C runs only in even
negative powers of C. Some details are given in Appendix D. In the kinematic region of “fast”
long strings, with 1 ≪ln S ≪J ≪S, this parity invariance property was already demonstrated
in a closely related way in [15].
2.3 Large spin expansion of energy of a spiky string in AdS
5
Let us now consider the spiky spinning string in AdS
5
[34], and find corrections to the leading
ln S term in its large spin expansion.
21
Note that the leading terms in expression of the previous subsection (2.3) dominate in the limit when
J
2
ln S

ln S
S
.
13
The integrals of motion here are the difference between the position of the spike and of the
middle of the valley between the two spikes, the spin and the energy [34]
22
∆θ =
π
n
=
sinh2ρ
0

2 sinh ρ
1
1

u
1
+u
0
¸
Π(
π
2
,
u
1
−u
0
u
1
−1
, p) −Π(
π
2
,
u
1
−u
0
u
1
+ 1
, p)

, (2.17)
S =
n cosh ρ
1

2 π

u
1
+u
0
¸
−(1 +u
0
)K(p) + (u
1
+u
0
)E(p) −
u
2
0
−1
u
1
+ 1
Π(
π
2
,
u
1
−u
0
u
1
+ 1
, p)

, (2.18)
E −ωS =
n

u
1
+u
0

2 π sinhρ
1
[K(p) −E(p)] , (2.19)
where n is the number of the spikes and
u
0
= cosh 2ρ
0
, u
1
= cosh 2ρ
1
, ω = coth ρ
1
, p =

u
1
−u
0
u
1
+u
0
. (2.20)
The string is rigidly rotating with the radial coordinate being ρ = ρ(σ), with ρ
0
and ρ
1
as its
minimal and maximal values (positions of the bottom of the valley between the spikes and the
spikes themselves). ρ
0
and ρ
1
are related by the condition (2.17). Solving for the remaining free
parameter gives E = E(S, n).
The large spin limit corresponds to ρ
1
→ ∞, i.e. to the case when the ends of the spikes
approach the boundary of AdS
5
. Let us set
y = e
−2ρ
1
, (2.21)
and expand in y →0. Then, at leading order, ∆θ =
π
n
= arcsin
1
u
0
+O(y) implies u
0
= cosh 2ρ
0
=
csc
π
n
and
E −S = −
n

ln y +O(y) , S =
n

1
y
+O(ln y) , (2.22)
i.e.
E −S =
n
2 π
ln
16 π S
n
+... . (2.23)
This is the result already found in [34], which reduces to the case of the folded string when
n = 2.
23
Expanding further near y ≃ 0 one gets
S =
n
4 π

1
y
+ ln y + 1 −2

u
2
0
−1 arccos

u
0
+ 1
2 u
0
+ ln
u
0
4

+... , (2.24)
∆θ =
π
n
= arcsin
1
u
0
+y

2 arcsin

u
0
+ 1
2 u
0
−π

+... , (2.25)
22
In the case of the multiply folded string with n spikes multiplying formulas one should multiply (2.19) by the
number m of the folds, and use that in this case ∆θ =
π
nm
. As a result, one is simply to substitute n →nm.
23
For n = 2 we have ∆θ =
π
2
(i.e. the angle between spikes is π), and thus ρ0 = 0 or u0 = 1.
14
where the second equation can be used to fix u
0
in terms of y and the number of spikes n.
Eliminating then y in favor of S, we have from (2.19)
E −S =
n

ln
¯
S +p
1
+
4
¯
S

ln
¯
S +p
2


4
¯
S
2

2 ln
2
¯
S +p
3
ln
¯
S +p
4

+
32
3
¯
S
3

2 ln
3
¯
S +p
5
ln
2
¯
S +p
6
ln
¯
S +p
7

+...

, (2.26)
where
¯
S =
16 π
n
S (2.27)
p
1
= −1 + ln sin
π
n
, p
2
= −1 + ln sin
π
n
+
π(n −2)
2n
cot
π
n
, (2.28)
p
3
= −10 +
2π(n −2)
n
cot
π
n
−2 cot
2
π
n
−4 ln csc
π
n
+ csc
2
π
n
,
p
4
= 6 −csc
2
π
n
+
π
2
(n −2)
2
2n
2

4π(n −2)
n
cot
π
n
+ cot
2
π
n

π
2
(n −2)
2
n
2
+ 1

+ ln csc
π
n

2 cot
2
π
n

2π(n −2)
n
cot
π
n
−csc
2
π
n
+ 2 ln csc
π
n
+ 10

, (2.29)
p
5
= −18 +O(n −2) , p
6
= 33 +O(n −2) , p
7
= −14 +O(n −2) . (2.30)
It is easy to check that (2.26) coincides with the energy (2.3) for the folded string in AdS
5
when
n = 2.
24
Retaining in (2.26) only the dominant contributions at each order of the above expansion we
obtain
E −S =
n

ln S +
n
2
8 π
2
S
ln S −
n
3
64 π
3
S
2
ln
2
S +
n
4
384 π
4
S
3
ln
3
S +... . (2.31)
This may be rewritten as
E −S =

λ n

ln

S +
1
2

λ n
2 π
ln S

+... , (2.32)
implying that the functional relation is satisfied (cf. (2.4)).
However, the reciprocity property is not respected in this case. Indeed, the analog of the
function
˜
f(S) in (2.5) has the following expansion
˜
f(S) =
n

ln
¯
S +q
1
+
q
2
¯
S
+
1
¯
S
2
(q
3
ln
¯
S +q
4
) +
1
¯
S
3
(q
5
ln
¯
S +q
6
)...

+... , (2.33)
where
q
1
= −1 + ln sin
π
n
, q
2
=
2π(n −2)
n
cot
π
n
, q
3
= 4 csc
2
π
n
, (2.34)
q
4
= 4 + 2π
2

n −2
n

2
(1 −2 csc
2
π
n
) + 4 ln sin
π
n
csc
2
π
n
, (2.35)
q
5
= O(n −2) , q
6
= O(n −2) , (2.36)
24
Note also that the form of p1 is consistent with the interpretation of the subleading term in the energy in
[27].
15
where q
5
, q
6
are non-zero for n = 2. The expansion (2.33), even if considerably simpler compared
to the energy (2.26), is not parity invariant under S →−S. It is interesting though that higher
powers of lnS appear to cancel in the subleading terms in (2.33).
25
The parity invariance is
restored in the case of the folded string when n = 2, where indeed (2.33) coincides with (2.5).
This breakdown of parity invariance for a string with n > 2 spikes is not totally surprising, as
such spiky string should correspond to an operator with non-minimal anomalous dimension for
a given spin, while the reciprocity was checked at weak coupling only for the minimal anomalous
dimensions. Indeed, anomalous dimensions of operators of twist higher than two with trajectories
close to the upper boundary of the band also do not respect the reciprocity as was seen recently
in the twist three case at weak coupling in [4].
It is interesting that our strong-coupling result (2.32),(2.33) has close similary with weak-
coupling one found for n = 3 in [4]: the functional relation (2.32) is still satisfied, and the parity
invariance is broken at level 1/S. Interestingly, the 1/S coefficient ∼ nq
2
in (2.33) (cf. (2.27))
is proportional, for n = 3, to

3, which is, suprisingly, the same factor appearing also in the
corresponding expression at weak coupling [4].
26
In general, this coefficient should be a function
of λ interpolating from weak to strong coupling but its dependence on n might be the same for
any λ.
3 Large spin expansion of folded string energy: 1-loop order
Let us now go back to the folded spinning string case of section 2.1 and compute the leading
1-loop corrections to its energy (2.3) in the large S expansion. We shall follow the general
approach for computation of quantum string corrections developed in [10] where the 1-loop shift
of the ln S term was found.
27
We shall find the 1-loop corrections to the subleading terms
in (2.3) by applying a perturbative procedure similar to the one used in [9] in the small spin
expansion case.
Our aim will be to verify that: (i) the structure of the large spin expansion (2.3) remains
the same also with the 1-loop corrections included, and (ii) the constraints on the coefficients
imposed by the functional relation and the reciprocity remain to be satisfied at the 1-loop order.
The fluctuation action in the conformal gauge expanded to quadratic order in fluctuations
25
This feature of the
˜
f-function is in a marked contrast with the anomalous dimension, whose large S expansion
includes growing powers of ln S in the coefficients of 1/S
n
terms. This reduction of singularity of the large S
expansion of
˜
f was observed also at weak coupling [26, 18, 4].
26
We thank G. Korchemsky for this observation.
27
The 2-loop correction to the scaling function was found in [38, 39]; a generalization to non-zero J was
considered in [36, 40] (parallel results from the Bethe ansatz were found in [41, 42]).
16
near the folded spinning string solution
¯
S = −

λ


0

¯
L has the following bosonic part
(see [10] and Appendix B)
¯
L
B
= − ∂
a
˜
t∂
a
˜
t −µ
2
t
˜
t
2
+∂
a
˜
φ∂
a
˜
φ +µ
2
φ
˜
φ
2
+ 4˜ ρ(κsinh ρ ∂
0
¯
t −wcosh ρ ∂
0
¯
φ) +∂
a
˜ ρ∂
a
˜ ρ +µ
2
ρ
˜ ρ
2
+ ∂
a
β
u

a
β
u

2
β
β
2
u
+∂
a
ϕ∂
a
ϕ +∂
a
ζ
s

a
ζ
s
, (3.1)
where
µ
2
t
= 2ρ
′2
−κ
2
, µ
2
φ
= 2ρ
′2
−w
2
, µ
2
ρ
= 2ρ
′2
−w
2
−κ
2
, µ
2
β
= 2ρ
′2
. (3.2)
Here β
u
(u = 1, 2) are the two AdS
5
fluctuations transverse to the AdS
3
subspace in which the
string is moving, while ϕ, ζ
s
(s = 1, 2, 3, 4) are fluctuations in S
5
. The fermionic part of the
quadratic fluctuation Lagrangian can be put into the form [10]
˜
L
F
= 2i(
¯
Ψγ
a

a
Ψ−µ
F
¯
ΨΓ
234
Ψ) , µ
F
= ρ

, (3.3)
and can be interpreted as describing a system of 4+4 2d Majorana fermions with σ-dependent
mass ρ

. As explained in [9], after squaring the corresponding “Dirac” operator, the fermionic
contribution to the 1-loop partition function can be represented as

1
2

4 ln det∆
F
+
+ 4 ln det∆
F

, ∆
F
±
≡ −∂
a

a
±ρ
′′

′2
. (3.4)
In the leading-order computation in the long-string limit that we are going to discuss below the
term ±ρ
′′
in the effective fermionic mass squared term in (3.4) can be ignored: as follows from
(B.12), ρ
′′
= O(η) and since according to (3.4) half of fermions has +ρ
′′
and half −ρ
′′
in their
mass term, the leading O(η) contribution to the partition function can come only from the ρ
′2
term in ∆
F
±
. We shall assume this when writing the fermionic contribution below.
Switching to euclidean signature (τ → iτ), the 1-loop correction to the energy can be found
from the 2d effective action
E
1
=
Γ
1
κT
, T ≡

dτ →∞ . (3.5)
Since the spinning string solution is stationary, both the bosonic and the fermionic fluctuation
Lagrangians do not depend on τ; thus, as in [9], we may compute the relevant 2d functional
determinants by reducing them to 1d functional determinants using
det[−∂
2
1
−∂
2
0
+m
2
] = T



det[−∂
2
1

2
+m
2
] , (3.6)
where m
2
is a generic mass term which may depend on σ.
17
Given that ρ(σ) is a complicated function (see (B.4)), we are unable to determine the fluctua-
tion spectrum exactly, and, as in [9], we will resort to perturbation theory in
1
S
or in parameter
η determining the maximal string length (see Appendix B and C). In (3.5) we have from (B.6)
κ = κ
0

η

(πκ
0
−2) +O(η
2
) , κ
0

1
π
ln
16
η
. (3.7)
Γ
1
will also be expected to have expansion in powers of η ∼
1
S
(see (B.9)) with the coefficients
containg powers of ln η.
To proceed, we need to expand the fluctuation Lagrangian in small η corresponding to large
S. Some relations needed below can be found in Appendix B. Let us first perform (as in [36])
the following rotation (
¯
t,
¯
φ) →(ξ, χ):
ξ = −
˜
t sinhρ +
˜
φ cosh ρ, χ = −
˜
φ sinhρ +
˜
t cosh ρ . (3.8)
Then the fluctuation Lagrangian takes the form
¯
L
B
= −∂
a
χ∂
a
χ + (µ
2
φ
sinh
2
ρ −µ
2
t
cosh
2
ρ +ρ
′2

2
+∂
a
ξ∂
a
ξ + (µ
2
φ
cosh
2
ρ −µ
2
t
sinh
2
ρ −ρ
′2

2
+4¯ ρ(κsinh
2
ρ −wcosh
2
ρ)∂
0
ξ +∂
a
¯ ρ∂
a
¯ ρ +µ
2
ρ
¯ ρ
2
+ 2ρ

(χξ

−ξχ

) +χξ(µ
2
φ
−µ
2
t
) sinh2ρ
+2˜ ρ ˙ χ(κ −w) sinh 2ρ +∂
a
β
u

a
β
u

2
β
β
2
u
+∂
a
ϕ∂
a
ϕ +∂
a
ζ
s

a
ζ
s
(3.9)
The reason for this rotation is that in the subsequent small η expansion the bosonic fluctuation
Lagrangian at order O(η
0
) will become σ-independent, i.e. will have constant coefficients as at
the leading order in long-string expansion considered in [10, 36].
28
Expanding the solution for ρ(σ) and the parameters κ and w in small η (see Appendix B),
the bosonic fluctuation Lagrangian becomes
˜
L
B
=
˜
L
0

˜
L
1
+..., where
˜
L
0
= − ∂
a
χ∂
a
χ +∂
a
ξ∂
a
ξ + 2κ
0
χξ

− 2κ
0
χ

ξ − 4κ
0
˜ ρ
˙
ξ
+ ∂
a
˜ ρ∂
a
˜ ρ +∂
a
β
u

a
β
u
+ 2κ
2
0
β
2
u
+∂
a
ϕ∂
a
ϕ +∂
a
ζ
s

a
ζ
s
, (3.10)
and
˜
L
1
= −κ
2
0
cosh(2κ
0
σ)ξ
2
−κ
2
0
cosh(2κ
0
σ)˜ ρ
2
−κ
2
0
sinh(2κ
0
σ) ξχ −
κ
0
π

0
π cosh(2κ
0
σ) −2]β
2
u
+ (χξ

−ξχ

)[
1
π

κ
0
2
cosh(2κ
0
σ)] − ˜ ρ ˙ χκ
0
sinh(2κ
0
σ) − ˜ ρ
˙
ξ[
2
π

0
cosh(2κ
0
σ)] . (3.11)
As already mentioned, the 1-loop effective action can be expressed in terms of 1d functional
determinants (with ∂
0
→ iω, see (3.6)). We shall denote the quadratic fluctuation operator in
28
As discussed below, we shall ignore the contribution of the turning points at σ =
π
2
and

2
and will treat the
fluctuation problem separately on each “quarter-string” interval.
18
the coupled (χ, ξ, ˜ ρ) sector as Q
ω
. Since
˜
L
0
has constant coefficients, the leading part of the
fluctuation operator coming from
˜
L
0
can be written as
Q
(0)
ω
=

¸
¸
¸
−(−∂
2
1

2
) 2κ
0

1
0
−2κ
0

1
−∂
2
1

2
−2ωκ
0
0 2ωκ
0
−∂
2
1

2
¸

. (3.12)
The 1-loop correction to the effective action is then
Γ
1
=
T


−∞

¸
−8 ln
det[−∂
2
1

2

′2
]
det[−∂
2
1

2

2
0
]
+ 2 ln
det[−∂
2
1

2
+ 2ρ
′2
]
det[−∂
2
1

2
+ 2κ
2
0
]
− ln
det
8
[−∂
2
1

2

2
0
]
det
2
[−∂
2
1

2
+ 2κ
2
0
] det
6
[−∂
2
1

2
]
+ ln
detQ
ω
detQ
(0)
ω
−ln
detP
ω
detQ
(0)
ω

, (3.13)
where
P
ω
=

¸
¸
¸
−(−∂
2
1

2
) 0 0
0 −∂
2
1

2
0
0 0 −∂
2
1

2
¸

. (3.14)
Also, Q
ω
= Q
(0)
ω
+ ηQ
(1)
ω
+ ..., where Q
(1)
ω
is the next to leading order coupled operator from
(3.11)
Q
(1)
ω
=

¸
¸
¸
0 Q
12
Q
13
Q
21
−κ
2
0
cosh(2κ
0
σ) Q
23
−Q
13
−Q
23
−κ
2
0
cosh(2κ
0
σ)
¸

. (3.15)
Here
Q
12
= −
κ
2
0
2
sinh(2κ
0
σ) +in[
1
π

κ
0
2
cosh(2κ
0
σ)] , (3.16)
Q
21
= −
κ
2
0
2
sinh(2κ
0
σ) −in[
1
π

κ
0
2
cosh(2κ
0
σ)] , (3.17)
Q
13
= −
κ
0
ω
2
sinh(2κ
0
σ), Q
23
= −ω[
1
π
+
κ
0
2
cosh(2κ
0
σ)] (3.18)
and we performed the Fourier transform in σ, i.e. replaced ∂
1
→in, n = 0, ±1, ..., as appropriate
for fluctuation fields which are 2π periodic in σ.
Our aim will be to determine the 1-loop correction to string energy to order η by computing
Γ
1
= Γ
(0)
1
+ Γ
(1)
1
+O(η
2
) , Γ
(1)
1
= O(η) . (3.19)
As in [9] the first, second and fourth terms in (3.13) can be computed to order O(η) using that
ln
det[O
(0)
+ηO
(1)
]
det O
(0)
= η Tr[(O
(0)
)
−1
O
(1)
] +O(η
2
) . (3.20)
19
While in [9] a similar contribution to the effective action happened to vanish since it was pro-
portional to the sum of squares of fluctuation masses,
29
here this leading term is no longer zero
as in the present case the expansion is around a nontrivial string background with different
propagators for different string fluctuations.
In (3.11) we used the expansion (B.11) of the solution ρ(σ) in small η. As discussed in
Appendices B and C, this expansion breaks down at the turning points where subleading terms
are of the same order as the leading term. As in the computation of the leading order in [10],
here we shall assume that one can ignore the contributions from the turning points. The classical
folded string solution is built out of four parts making up the closed string (e.g., the expansion
(B.11) used in (3.11) is defined for 0 ≤ σ ≤
π
2
).
30
The closed string fluctuations by definition
must be periodic in 0 ≤ σ ≤ 2π.
We shall assume that we can treat the problem “piece-wise” also at the fluctuation level.
Direct implementation of this may effectively bring back the turning-point contributions. We
shall parametrize our current lack of control of such terms by including the possible contribution
with an arbitrary coefficient in the final result.
We shall split the integral over σ as follows


0



1

¸ π
2
0
dσ +

π
π
2
dσ +

2
π
dσ +



2

. (3.21)
Considering the first interval (0,
π
2
), the order η contribution of the decoupled boson β
u
in (3.1)
and (3.10),(3.11) can be obtained as

ln
det[−∂
2
1

2
+ 2ρ
′2
]
det[−∂
2
1

2
+ 2κ
2
0
]

(1)
= −
ηκ
0
π

¸
n=−∞
1
n
2

2
+ 2κ
2
0
π
2
0


[πκ
0
cosh(2κ
0
σ) −2]
= −
ηκ
0


¸
n=−∞
sinh(πκ
0
) −2
n
2

2
+ 2κ
2
0
. (3.22)
Similarly, for the fermionic contribution (the first term in (3.13)) we get

ln
det[−∂
2
1

2

′2
]
det[−∂
2
1

2

2
0
]

(1)
= −
ηκ
0


¸
n=−∞
1
n
2

2

2
0
π
2
0


[πκ
0
cosh(2κ
0
σ) −2]
= −
ηκ
0


¸
n=−∞
sinh(πκ
0
) −2
n
2

2

2
0
. (3.23)
29
The mass sum rule implies the 1-loop UV finitness of the superstring; it was proven in general for any string
solution in [43].
30
Note that for the second and the fourth σ intervals where ρ decreases we need to use the minus sign in (B.2).
20
For the coupled part one finds

ln
detQ
ω
detQ
(0)
ω

(1)
= η
π
2
0


Tr[(Q
(0)
ω
)
−1
Q
(1)
ω
]
=
ηκ
0
π

¸
n=−∞
(n
2

2
)
2
−n
2
(n
2

2

2
0
) sinh(πκ
0
)
(n
2

2
)
2
(n
2

2
+ 4κ
2
0
)
. (3.24)
The contributions of the other three intervals of σ are the same.
Collecting the above results we observe that the final expression for the order η term in the
effective action Γ
(1)
1
is UV finite. Moreover, the part that does not contain the sinh(πκ
0
) factor
is IR finite, i.e. the non-trivial potentially IR divergent contributions of the two unphysical
AdS
5
massless modes (χ, ξ) (time-like and longitudinal) that appear in the coupled part of the
fluctuation Lagrangian cancel.
31
Explicitly, integrating first over ω we obtain
32
the order η contribution to the 1-loop effective
action (3.19) coming from (3.20) as
Γ
(1)
1
= −
T η


¸
n=−∞

A
n
+C
n
sinh(πκ
0
)

, (3.25)
where
A
n
=

0

n
2

2
0


0

n
2
+ 2κ
2
0


0

n
2
+ 4κ
2
0
, (3.26)
C
n
=
κ
0
2n
+
3n

0


0

n
2

2
0
+

0

n
2
+ 2κ
2
0

3n
2

0

n
2
+ 4κ
2
0
. (3.27)
The coefficient of the part proportional to sinh(πκ
0
) given by
¸
n
C
n
is UV finite but formally has
an IR singular contribution.
33
This term should be an artifact of our computational procedure
related to the problem with expansion in η in (B.11) near the turning points (see Appendices B
and C). Insisting on omitting the turning point contributions means that we should drop this
IR singular ∼ sinh(πκ
0
) term, and this is what we will do below.
34
We believe that in a more
systematic treatment that consistently treats the turning point contributions such terms will
be automatically absent (equivalently, in our present form of the expansion, such terms should
resum away, see also the discussion in Appendix C).
31
Their flat-space contribution is cancelled against the conformal gauge ghost contribution.
32
Let us mention that if we perform the sum over n first, then using the residue theorem in the integral over ω
we arrive back at the same sum as below.
33
The IR singular contribution goes away if one separates it before doing the integral over ω. Then we get for
the large κ0 behaviour of
P
n
Cn: 2
P

n=1
Cn = κ0(ln κ0 −6 ln2 −
3
2
+ γE) + O((κ0)
0
).
34
To justify the expansion in (B.11) we need to omit the turning point contribution. That can be done by
shifting the upper limit of the integration over σ in (3.22)-(3.24) to
π
2
− ǫ, ǫ → +0. Then the coefficient of Cn
will become sinh[(π −2ǫ)κ0] ∼ η
−1+
2
π
ǫ
and thus is subleading compared to the contribution of order O(η
0
).
21
Computing the remaining
¸
n
A
n
contribution in (3.25),(3.26) using the Euler-MacLaurin
formula

¸
n=1
f(n) =


1
dn f(n) +
f(1) +f(∞)
2
+

¸
k=1
B
2k
(2k)!
[f
(2k−1)
(∞) −f
(2k−1)
(1)] , (3.28)
we can extract its large κ
0
behaviour

¸
n=−∞
A
n
= 12κ
0
ln 2 +O(e
−2πκ
0
) . (3.29)
The contribution of this term in Γ
(1)
1
in (3.25) to the energy (3.5) is then (using (B.9))
E
(1)
1
= −
3 ln 2
π
κ
0
κ
η . (3.30)
Let us now include the O(η
0
) contribution to Γ
1
(3.19) coming from the third and fifth terms
in (3.13). Since Q
(0)
ω
in (3.12) has no σ dependence, its functional determinant
detQ
(0)
ω
= −det
2
(−∂
2
1

2
) det(−∂
2
1

2
+ 4κ
2
0
) (3.31)
can be easily computed as a product over integer n of a matrix determinant (after ∂
1
→ in).
Since detP
ω
= −det
3
(−∂
2
1

2
) we may write the relevant contribution from (3.13) as
Γ
(0)
1
= −
T


−∞
dω ln
det
8
(−∂
2
1

2

2
0
)
det
2
(−∂
2
1

2
+ 2κ
2
0
) det
5
(−∂
2
1

2
) det(−∂
2
1

2
+ 4κ
2
0
)
. (3.32)
Since ln det(−∂
2
1
+ ω
2
+ κ
2
) =
¸

n=−∞
ln(n
2
+ ω
2
+ κ
2
), doing the integral over ω we finally
obtain the 1-loop correction to the string energy to order O(η) as
E
(0)
1
=
1


¸
n=−∞
¸
2

n
2
+ 2κ
2
0
+

n
2
+ 4κ
2
0
+ 5

n
2
−8

n
2

2
0

. (3.33)
This turns out to be a direct generalization of the leading-order result of [10] where κ should be
replaced by κ
0
in the fluctuation mass terms (but not in the overall
1
κ
factor due to t = κτ).
Using again the Euler-MacLaurin formula to transform the sum into an integral we find
35
E
(0)
1
=
1
κ

−3 ln 2 κ
2
0

5
12
+O(e
−2πκ
0
)

. (3.34)
This is, of course, in agreement with the result of [10] and also, for the subleading term, with
ref. [44].
36
35
Note that the sum of (3.26) is minus the derivative over κ0 of the sum in (3.33), which explains why the
corresponding coefficients are closely related.
36
Ref.[44] considered, following [10], the formal sum (3.32) with κ0 →κ and with the n = 0 term omitted (this
term was omitted in [10] since there it was subleading in the infinite κ limit). As a result, the expression in [44]
contained an extra (minus “zero mode”) term 3 −

2. Note that the coefficient in the exponent of the leading
e
−2πκ
0
exponential correction [44] is determined by the mass of the lightest mode – in the present case of the
fermionic mode. The exponential term in the square bracket has also a prefactor of

κ0.
22
As we shall argue in Appendix C, an additional contribution that may come from near turning
point regions can be parametrized as follows:
Γ
(2)
1
=
c
π
κ
0
T , E
(2)
1
=
c
π
κ
0
κ
, (3.35)
where c is an undetermined constant (we included factor of π for convenience).
Inverting the relation between S and η in (B.9) to order O(η) we get
η =
2
πS

ln(8πS) −3
π
2
S
2
+... , (3.36)
which, plugged into (3.7), gives
37
κ =
ln(8πS)
π

1

2
S
+... , κ
0
=
ln(8πS)
π
+
ln(8πS) −3

2
S
+... ,
κ
0
κ
= 1+
1
2πS
+... . (3.37)
This means that the dominant term in (3.34) is the −3 ln 2
κ
2
0
κ
one: the other terms −
5
12κ

1
ln S
and

κ
0
κ
e
−2πκ
0

1

ln S S
2
should be subleading and should be ignored in the approximation we
considered above where we dropped terms of higher order in η at earlier stages; these terms are
expected to cancel out in a more systematic treatment.
38
Thus we find that the 1-loop correction to the folded string energy to order O(
ln
2
S
S
2
) can be
written in the same form as the classical energy (2.3), i.e. as in (1.2),(1.3)
E
1
= b
0
ln S +b
c
+
b
11
ln S +b
10
S
+O(
ln
2
S
S
2
) . (3.38)
The contribution from E
(0)
1
in (3.34) to the 1-loop coefficients is
b
(0)
0
= −
3 ln 2
π
, b
(0)
c
= −
3 ln 2
π
ln 8π, b
(0)
11
= −
3 ln 2
π
2
, b
(0)
10
= −
3 ln 2
π
2

ln 8π −
5
2

, (3.39)
while E
(1)
1
in (3.30) and E
(2)
1
in (3.35) contribute as
b
(1)
10
= −
6 ln 2
π
2
, b
(2)
c
=
1
π
c , b
(2)
10
=
1

2
c . (3.40)
Thus finally we obtain for the full 1-loop coefficients
b
0
= −
3 ln 2
π
, b
c
=
1
π

−3 ln 2 ln 8π + c

, (3.41)
b
11
= −
3 ln 2
π
2
, b
10
=
1

2

−6 ln 2

ln 8π −
1
2

+ c

. (3.42)
37
Note that in the expression for the energy in (3.5) we need to keep κ to order O(η) to get the correction in
E1 also to order η ∼
1
S
.
38
The role of these subleading terms and their possible resummation remains to be studied in more detail.
23
The functional and reciprocity relations in (1.18),(1.22) at strong coupling are (see discussion
in Appendix E)
¯
f
11
=
1
2
¯
f
2
,
¯
f
10
=
1
2
¯
f
¯
f
c
,
¯
f ≡
f

λ
,
¯
f
c

f
c

λ
,
¯
f
1k

f
1k
λ
, (3.43)
¯
f = a
0
+
b
0

λ
+... ,
¯
f
c
= a
c
+
b
c

λ
+... ,
¯
f
1k
= a
1k
+
b
1k

λ
+... . (3.44)
They imply that the coefficients in (3.38) should obey (see (E.6),(E.7))
b
11
= a
0
b
0
, b
10
=
1
2
(a
0
b
c
+b
0
a
c
) . (3.45)
Recalling the values of the leading coefficients at the classical level in (2.3)
a
0
=
1
π
, a
c
=
1
π
(ln 8π −1) , (3.46)
we see that the relations (3.45) are indeed satisfied by the expressions in (3.41), i.e. the functional
and the reciprocity relations appear to apply also including string 1-loop corrections. Note that
this is true for any value of teh undetermined coefficient c.
Needless to say, it would be interesting to generalize the 1-loop computation of this section and
the check of reciprocity to the case of non-zero J and to attempt to relate the strong-coupling
version of the reciprocity discussed in Appendix E to its weak-coupling finite twist one in (1.21).
Acknowledgments
We are grateful to F. Alday, G. Arutyunov, B. Basso, L. Dixon, S. Frolov, N. Gromov, G.
Korchemsky, M. Kruczenski, T. McLoughlin, A. Rej, R. Roiban, D. Seminara and S. Theisen
for useful discussions. We thank G. Korchemsky and R. Roiban for very helpful comments on
the draft. A.A.T. thanks I.Y. Park for a collaboration on this topic back in 2005. The research
of V. Forini is supported by the SFB 647 ‘Space-Time-Matter’ grant and by the Alexander von
Humboldt foundation. A.T. was supported in part by NSF under grant PHY-0653357. Part
of this work was done while V.F. and A.A.T. were participants of the GGI Florence workshop
“Non-Perturbative Methods in Strongly Coupled Gauge Theories” and we thank the Galileo
Galilei Institute for Theoretical Physics for the hospitality and the INFN for partial support.
Note Added
As we have learned (N. Gromov, private communication) an independent way of evaluating the
1-loop correction to the folded string energy based on the algebraic curve approach to extracting
24
fluctuation frequencies [53] leads to the value c = 6 ln 2 + π for the undetermined coefficient in
(1.6),(1.9),(1.11),(3.35),(3.41),(3.42). This c contribution changes the 1-loop coefficient in (1.6)
from 3 ln 2 to −3 ln 2−π. It is interesting to note that for c = 6 ln 2+d, where d is a constant not
involving ln 2, the coupling redefinition

λ →

λ + 3 ln 2 (suggested to us by G. Korchemsky)
removes all ln 2 terms from the leading 1-loop coefficients in (1.5)–(1.11), just as it did in the
cusp anomaly coefficient in [37]. Namely, then we get
f =

λ
π
+O(
1

λ
), f
c
=

λ
π
[ln


λ
−1] +
d
π
+O(
1

λ
), f
10
=
λ

2
[ln


λ
−1] +
d

λ

2
+O(λ
0
).
Appendix A: Comments on conformal algebra realizations
Starting with a conformal theory in R
1,3
with the standard SO(2, 4) conformal group generators
P
m
, M
mn
, K
m
, D (m, n = 0, 1, 2, 3) one may define the collinear SL(2, R) subgroup as generated
by the following light-cone components [30]:
L
+
≡ −iP
+
, L


i
2
K

, L
0

i
2
(D +M
−+
), (A.1)
[L
0
, L
±
] = ±L
±
, [L
+
, L

] = −2L
0
. (A.2)
If the eigenvalue of D is dimension ∆ and the eigenvalue of M
−+
– the collinear projection of the
Lorentz spin S, then the eigenvalue of L
0
is the conformal spin s =
1
2
(∆+S). The corresponding
quadratic Casimir operator is C
2
= s(s −1).
At the same time, in the R
2,4
embedding representation of the global AdS
5
space the gener-
ators Σ
MN
(M, N = 0, 1, 2, 3, 4, 5 with the signature −+ + + +−) of SO(2, 4) linearly realised
on the embedding coordinates Y
M
can be related to the standard boundary conformal group
generators as (see, e.g., [45])
Σ
mn
= M
mn
, Σ
m4
=
1
2
(K
m
−P
m
) , Σ
m5
=
1
2
(K
m
+P
m
) , Σ
54
= D . (A.3)
Then the standard spin is S = Σ
12
= M
12
, the conformal spin is S

= Σ
34
=
1
2
(K
3
− P
3
), and
the conformal energy is the rotation generator in the 05 plane, i.e. the global AdS
5
energy,
E = Σ
05
=
1
2
(K
0
+P
0
).
In general, the energy E of a string state in global AdS
5
space with boundary R×S
3
should
be equal to the energy of the corresponding SYM state on R×S
3
. Through radial quantization
(and analytic continuation) this state may be associated to a local operator in R
1,3
that creates
it. The AdS
5
energy E = Σ
05
or conformal Hamiltonian generates an SO(2) subgroup while
the dilatation operator D = Σ
54
generates an SO(1, 1) subgroup of SO(2, 4).
39
After the
39
Their eigenvalues happen to be the same since the two representations (the unitary one classified by SO(4) ×
SO(2) and the one classified by SO(4) ×SO(1, 1)) are related by a global SO(2, 4) similarity transformation (see,
e.g., [46]).
25
Euclidean continuation of the embedding coordinate Y
0
→iY
0E
(to allow for the mapping from
R × S
3
to R
4
) one may exchange Y
0E
with Y
4
which exchanges the generator Σ
54
= D with
E = Σ
05
=
1
2
(P
0
+K
0
).
To relate the SO(1, 2) subgroup of SO(2, 2) which is a symmetry of global AdS
3
subspace of
AdS
5
where the folded spinning string is moving to the collinear SL(2, R) subgroup classifying
the operators like tr(ΦD
S
+
Φ) one is also to perform an additional analytic continuation that
interchanges the euclidean (12) plane with the hyberbolic (+−) plane. Since different choices
are formally related via SO(2, 4) transformations and a re-identification of the generators one
may expect that the two representations should be equivalent.
40
Still, the representations of SO(2, 2) or string states in AdS
3
are naturally labeled by (E, S),
and the relation to SO(1, 2) labels
1
2
(E+S) does not appear to be natural, unless one is interested
in the large spin expansion (see in this connection [47, 55]). That relation may possibly be made
more explicit by choosing a different set of coordinates in global AdS
5
in which the boundary is
not R×S
3
but AdS
3
×S
1
(see [48] where such coordinates in the boundary theory where used
to explain the leading E ∼ ln S behaviour).
Let us mention also that the relation (1.23) or E−S = f(E+S) is reminiscent of a light-cone
gauge expression, where f would be a light-cone Hamiltonian (cf. [49, 50, 47]).
Appendix B: Review of folded string solution with J = 0
In this Appendix we review the folded spinning string solution in AdS
3
[7, 6] and consider its
large spin expansion (see also [10]).
The solution is given by
t = κτ, φ = wτ, ρ = ρ(σ) , ds
2
= −cosh
2
ρ dt
2
+dρ
2
+ sinh
2
ρ dφ
2
, (B.1)
where ρ(σ) satisfies
ρ

= ±κ

1 −η sinh
2
ρ . (B.2)
Here ρ varies from 0 to its maximal value ρ
0
related to the parameter η by
coth
2
ρ
0
=
w
2
κ
2
≡ 1 +η . (B.3)
The solution in the interval 0 ≤ σ ≤
π
2
with the initial condition ρ(0) = 0 is
sinh ρ =
1

η
sn

κ

η σ, −
1
η

, 0 ≤ σ ≤
π
2
. (B.4)
40
The formal relation can be achieved by a continuation to euclid: by replacing null direction like x0 +x3 with
a complex one x1 + ix2, i.e. replacing the operator tr(ΦD
S
+
Φ) with tr(ΦD
S

Φ), where D∗ = D1 + iD2.
26
The condition satisfied at the turning point ρ
0
at σ =
π
2
is ρ

(
π
2
) = 0. To construct the full
(2π periodic) folded closed string solution one should glue together four such functions on
π
2
intervals to cover the full 0 ≤ σ ≤ 2π interval; e.g., for
π
2
< σ < π we have
sinh ρ =
1

η
sn

κ

η (π −σ), −
1
η

,
π
2
≤ σ ≤ π . (B.5)
The expressions for the parameter κ, the energy and the spin in terms of η are
κ =
1

η
2
F
1
(
1
2
,
1
2
; 1; −
1
η
), E =
1

η
2
F
1
(−
1
2
,
1
2
; 1; −
1
η
), S =

1 +η


η
2
F
1
(
1
2
,
3
2
; 2; −
1
η
) (B.6)
In this paper we are interested in the large spin or long string limit, i.e. small η expansion.
Since in section 3 we compute 1-loop correction only to order
1
S
we will only need expansions to
order O(η). Expanding κ, E, S in small η we obtain
41
κ = κ
0

η

(πκ
0
−2) +O(η
2
) , κ
0

1
π
ln
16
η
, (B.7)
E =
2
πη
+
πκ
0
+ 1


η
32π
(2πκ
0
−3) +O(η
2
) , (B.8)
S =
2
πη

πκ
0
−3


η
32π
(2πκ
0
+ 13) +O(η
2
) . (B.9)
Expanding the solution (B.4) in small η we obtain, for 0 < σ <
π
2
,
sinhρ = sinh(κσ) −
η
8

sinh(2κσ) −2κσ

cosh(κσ) +O(η
2
) , (B.10)
or, using (B.7),
sinhρ = sinh(κ
0
σ) −
η
8

sinh(2κ
0
σ) −
4
π
σ

cosh(κ
0
σ) +O(η
2
) . (B.11)
To leading order when η → 0, κ
0
→ ∞ the string touches the boundary of AdS
5

0
= ∞)
and the solution can be approximated (away from the turning points) by simply ρ = κ
0
σ. This
limiting case proved to be a useful framework for computing 1- and 2-loop string corrections
[10, 36, 38, 39]. At the next order in small η expansion the “ends” (turning points) of the string
are close to the boundary but no longer touch it.
We should add a word of caution about the use of the formal expansion in (B.10) or (B.11).
Notice that it goes in powers of η with coefficients containing lnη and is not, strictly speaking,
valid close enough to the turning points. Indeed, for σ =
π
2
we have sinh(2κ
0
σ) = sinh(πκ
0
) ≈
1
2
e
πκ
0
∼ η
−1
and similarly cosh(κ
0
σ) = sinh(
π
2
κ
0
) ∼ η
−1/2
. Hence at the turning point the order
41
Let us note that these expansions were found using pre-Mathematica 6 versions of Mathematica (Mathematica
6 apparently has some bug leading to inconsistent expansions for some elliptic and hypergeometric functions).
27
η term in (B.11) goes actually as η
−1/2
, i.e. is of the same order as the leading term sinh(κ
0
σ).
If σ is slightly away from the turning point the subleading terms are smaller than the leading
term but then the expansion and the contributions to the energy need to be resummed. The
resummation at the level of the string profile ρ(σ) is completely equivalent to its expansion near
the turning point σ =
π
2
, as will be discussed in Appendix C.
In section 3 we ignored the regions near the turning points and thus uses the formal expansion
(B.11) to compute it. Similar assumption was made in [10] in the computation of the 1-loop shift
of the coefficient of the ln S term in the energy (where it was indeed justified). A reason behind
this assumption is that the masses of string fluctuations in (3.2) depend on ρ
′2
which is small
near the turning points where ρ

= 0. Thus the correction to the leading result of [10] should
mainly come from the “internal” parts of the σ-interval, where the expansion (B.11) is justified.
We shall comment more on this point in Appendix C. In section 3 we included a possible term
which we may thus miss with an aribitrary coefficient yet to be deterined.
For the computation of the 1-loop correction in section 3 we will need the following expansions
ρ
′2
= κ
2
0

η

κ
0
[πκ
0
cosh(2κ
0
σ) −2] +O(η
2
) , (B.12)
κsinh ρ = κ
0
sinh(κ
0
σ) +
η

¸

0
σ cosh(κ
0
σ)
−[πκ
0
cosh(2κ
0
σ) + 3πκ
0
−4] sinh(κ
0
σ)
¸
+O(η
2
) , (B.13)
wcosh ρ = κ
0
cosh(κ
0
σ) +
η

¸

0
σ sinh(κ
0
σ)
−[πκ
0
cosh(2κ
0
σ) −3πκ
0
−4] cosh(κ
0
σ)
¸
+O(η
2
) . (B.14)
Notice again that these expansions are formally invalid at the turning point (but are justified
away from it). This is evident, e.g., in (B.12) where the leading term does not vanish at the
turning point where one should have ρ

= 0. What happens is that at σ =
π
2
the leading term
κ
2
0
gets cancelled agianst the sum of subleading terms which all are of the same order, i.e. are
proportional to κ
2
0
(see Appendix C).
The masses appearing in the fluctuation Lagrangian in section 3 are then expanded as follows
µ
2
t
= κ
2
0

ηκ
0
π

πκ
0
cosh(2κ
0
σ) −
1
2
πκ
0
−1

+O(η
2
) , (B.15)
µ
2
φ
= κ
2
0

ηκ
0
π

πκ
0
cosh(2κ
0
σ) +
1
2
πκ
0
−1

+O(η
2
) , (B.16)
µ
2
ρ
= −ηκ
2
0
cosh(2κ
0
σ) +O(η
2
) . (B.17)
Appendix C: Resummation of “long string” expansion near the turning points
Continuing the discussion of the previous Appendix B here we will show that it is possible
to resum systematically the terms ∼ e
2 nκ
0
σ
appearing in the formal η expansion of ρ(σ) in
28
(B.10). These terms are potentially dangerous since they scale at the turning point σ = π/2 as
e
nπ κ
0
= (16/η)
n
and spoil the perturbative expansion.
In the first “quarter-string” interval 0 ≤ σ ≤
π
2
, the function ρ(σ) obeys the differential
equation (B.2) with plus sign and ρ(0) = 0. Its formal expansion in powers of η (treating κ
0
in
(B.7) as a constant parameter) reads (cf. (B.10))
ρ(σ) = κ
0
σ +

σ


1
8
sinh (2κ
0
σ)

η
+


cosh (2κ
0
σ) σ


13σ
64π
+
1
16
sinh (2κ
0
σ) +
1
256
sinh (4κ
0
σ)

η
2
+


sinh(2κ
0
σ) σ
2
16π
2
+
29 cosh (2κ
0
σ) σ
256π
+
cosh (4κ
0
σ) σ
128π
+
23σ
192π
(C.1)

1
128
cosh (2κ
0
σ) sinh(2κ
0
σ) −
cosh (4κ
0
σ) sinh(2κ
0
σ)
3072

125 sinh (2σκ
0
)
3072

η
3
+O

η
4

.
Since here, in fact, κ
0
= −
1
π
ln
η
16
the hyperbolic functions potentially reduce the true order in
the η expansion. The dangerous terms can be easily identified by setting
t = e
κ
0
σ
, r(t) ≡ ρ

ln t
κ
0

, (C.2)
and neglecting exponentially suppressed terms in the above expansion. The next-to-leading
(NLO) result which is correct at O(η) can be written
r
NLO
(t) = ln t −
ηt
2
16
+
η
2
t
4
512

η
3
t
6
12288
+
η
4
t
8
262144
+· · · (C.3)
+
¸
ηt
2
32

η
2
t
4
512
+
η
3
t
6
8192
+· · · +
ln t
2πκ
0

1 −
ηt
2
8
+
η
2
t
4
128

η
3
t
6
2048
+· · ·

η +. . . .
All terms (η t
2
)
k
are O(1) at σ =
π
2
and the above infinite series need resummation. This can
be accomplished as follows. Introducing h(t) ≡ r(t) −ln t we get
1 + t h

(t) =
κ
κ
0

1 −η

t e
h
−t
−1
e
−h
2

2
, h(t) = r(t) −lnt . (C.4)
Taking the large t limit we arrive at the following equation for the leading order term h
LO
1 + t h

LO
(t) ≃

1 −
1
4
ηt
2
e
2 h
LO
(t)
, (C.5)
which can be integrated and gives
h
LO
(t) = −ln

1 +
η t
2
16

. (C.6)
As a check we can reexpand and find indeed
h
LO
(t) = −
ηt
2
16
+
η
2
t
4
512

η
3
t
6
12288
+
η
4
t
8
262144

η
5
t
10
5242880
+O

t
11

(C.7)
29
which are the leading terms in r(t).
The NLO approximation h
NLO
(t) is simply obtained by including an extra piece in the square
root and taking into account that the ratio κ/κ
0
has a non trivial expansion in η,
1 + t h

NLO
(t) =
κ
κ
0

1 −η
t
2
4
e
2 h
NLO
(t)
+
η
2
. (C.8)
Integrating this equation, substituting the necessary terms in the expansion of κ/κ
0
, and ne-
glecting all NNLO terms we then get
h
NLO
(t) = −ln

1 +
η t
2
16


ln t
2π κ
0
+
η
2
t
2
32
1
1 +
η t
2
16

1 −
2 ln t
π κ
0

. (C.9)
The expansion of (C.9) in powers of η gives
h
NLO
(t) =

ln t
2πκ
0

t
2
16

η +

t
4
512

t
2
ln t
16πκ
0
+
t
2
32

η
2
+


t
6
12288
+
t
4
ln t
256πκ
0

t
4
512

η
3
+

t
8
262144

t
6
ln t
4096πκ
0
+
t
6
8192

η
4
+


t
10
5242880
+
t
8
ln t
65536πκ
0

t
8
131072

η
5
+· · · , (C.10)
which agrees indeed with the expansion of r
NLO
in (C.3).
In terms of the string profile ρ(σ) = r(e
κ
0
σ
) = h(e
κ
0
σ
) −κ
0
σ this result can be written as
ρ
NLO
(σ) = κ
0
σ −ln

1 +

η
16

1−2σ/π

+ η

σ
2 π
+
1
2

η
16

1−2σ/π
1 +

η
16

1−2σ/π

1 −
2 σ
π

. (C.11)
This expression resums at NLO order the contributions near the boundary point σ =
π
2
. This
expression is not, of course, expected to be correct near σ = 0, but it must reproduce the exact
value of ρ(
π
2
) with order O(η) included. This is true since
ρ
NLO
(
π
2
) =

ln 2 −
1
2
ln η

+
η
4
+O

η
2

(C.12)
is in agreement with the exact value of ρ(
π
2
) which is
ρ(
π
2
) = arcsinh
1

η
=

ln 2 −
1
2
ln η

+
η
4


2
32
+

3
96
+O

η
4

. (C.13)
Also, as an additional check, we immediately reproduce that ρ

NLO
(
π
2
) = 0.
30
In order to obtain the resummation in a systematic way and to show that it comes from the
behaviour of the string profile around the turning point we can work out the expansion of the
differential equation for ρ(σ) around σ =
π
2
. To this aim, let us define
x = κ

π
2
−σ

, ˆ ρ(x) ≡ ρ

π
2

x
κ

, (C.14)
and solve the corresponding equation (cf. (B.2))
ˆ ρ

(x) = −

1 −η sinh
2
ˆ ρ(x) , ˆ ρ(0) = arcsinh
1

η
. (C.15)
perturbatively in η, i.e.
ˆ ρ(x) = arcsinh
1

η
+ ln sech x −
η
4
x tanh x
+
η
2
128

−1 + cosh(2x) −4 x
2
sech
2
x + 10 x tanh x

+O(η
3
) (C.16)
Here the value ˆ ρ(0) was left unexpanded. Expanding it consistently we get the final result
ˆ ρ(x) =

πκ
0
2
+ ln
sech x
2

+

1
4

1
4
xtanh x

η
+


1
32
x
2
sech
2
x +
1
128
cosh(2x) +
5
64
x tanh x −
13
128

η
2
+O(η
3
) (C.17)
If we now use the definition of x = κ

π
2
−σ

in this expression, we get precisely the NLO
resummation in Eq. (C.11) plus a new η
2
term which is beyond the order of accuracy of (C.11).
One may wonder if this systematic resummation of ρ(σ) can be used to resum the associated
contributions in the 1-loop correction to string energy discussed in section 3. This is not,
however, immediately clear. Plugging the expansion of ρ around σ =
π
2
in the Q
ω
operator in
(3.13) and denoting the resulting terms with label “fold” to indicate the expansion point, we
find
Q
ω
= Q
(0)
ω,fold
+η Q
(1)
ω,fold
+· · · , (C.18)
where
Q
(0)
ω,fold
=

¸
¸
¸
−n
2
−ω
2
−V
1
−V
2
V
1
n
2

2


0
ω
V
2
−2ωκ
0
−V
2
V
2
2ωκ
0
+V
2
n
2

2


0
ω
V
2
¸

, (C.19)
z ≡
π
2
−σ , V
1
=
κ
2
0
cosh
2

0
z)
+ 2inκ
0
tanh (κ
0
z) , V
2
=
ωκ
0
cosh
2

0
z)
. (C.20)
31
In the large κ
0
limit, we can make the following replacements (


0
dz δ
+
(z) = 1)
κ
0
sech
2

0
z) −→δ
+
(z) , κ
0
tanh(κ
0
z) −→κ
0
−ln(2) δ
+
(z) . (C.21)
After this substitution we can write
Q
(0)
ω,fold
= Q
(0)
ω
+Q
(0)

ω,fold
δ
+
(z), (C.22)
where Q
(0)
ω
is the same operator (3.12) we found in section 3 in the expansion valid near σ = 0,
while the new piece is
Q
(0)

ω,fold
=

¸
¸
¸
0 2 i n ln 2 −κ
0
−ω
−2 i n ln 2 −κ
0
−2κ
0
−ω
ω ω −2κ
0
¸

. (C.23)
This is a O(κ
0
) perturbation over Q
(0)
ω
whose matrix elements are O(κ
2
0
) (since n, ω ∼ κ
0
in the
combined sum and integral like in (3.24)). Unfortunately, higher order terms coming from this
term can be estimated to have the same order of magnitude and thus must be resummed. In
principle, the contribution from Q
(0)
ω,fold
must be treated exactly and separately, a task which we
leave for the future.
Still, it is encouraging to note that a possible non-zero contribution from the near-turning-
point region is expected to change the one-loop energy by a term proportional to
κ
0
κ
= 1 +

1
4

1
2 π κ
0

η +· · · = 1 +
1
2 π S
+· · · . (C.24)
This means that the induced modification of the coefficients in appearing in (3.38) must obey
δb
0
= δb
11
= 0 , δb
10
=
1
2 π
δb
c
. (C.25)
Remarkably, this is precisely what is required by the reciprocity conditions in (3.45),(3.46).
Appendix D: Details of large spin expansion for folded (S, J) spinning string
In this section we collect some details on large spin expansions used in Section 2.2.
In the “slow long strings” regime (S ≫1, J ≪S), the small η expansions for the “anomalous”
32
part of the energy and the conformal spin read
42
˜ γ
J≪1
= κ +
κ
ω
S −S −J ≈
¸

1 + ln η
π
+
4(ln η + 12)
π
η
2
+O(η
4
)

−J
+ π J
2
¸
(1 −ln η)
2 ln
2
η
−η
2

10
ln η
+
20
ln
2
η

44
ln
3
η

+O(η
4
)

+... ,
˜ s
J≪1
= S +
1
2
J +
1
2
˜ γ ≈
¸
1
8πη

2 ln η + 11

+O(η
3
)

+ π J
2
¸
1
16 η ln
2
η
−η

3
2 ln η

13
4 ln
2
η

11
2 ln
3
η

+O(η
3
)

+... . (D.1)
For the “fast long strings” (S ≫1, with ln S ≪J ≪S) one finds
˜ γ
ln S≪J≪S

1
π
2
J
¸
ln η

1 +
1
2
ln η

+ 44 η
2

1
11
ln
2
η −ln η −1

+O(η
4
)

+
1
π
4
J
3
¸

1
8

ln
4
η + 4 ln
3
η

+ 2 η
2
(−5 ln
4
η + 5 ln
3
η + 33 ln
2
η) +O(η
4
)

+... ,
˜ s
ln S≪J≪S
≈ J
¸

1
8 η ln η

1 −
11 + 12 ln η
2 ln
2
η

+O(η
3
)

+
1
π
2
J
2
¸

ln η
16 η
−η

3 ln
2
η
2
+ 4 ln η −
11
4

+O(η
3
)

+... . (D.2)
Since the function
˜
f =
f

λ
in (1.14) coincides with the anomalous dimension evaluated at zero of
the denominator in (2.16) in both cases we get an equation expressing the parameter η in terms
of only the odd powers of the Casimir C. From the power series expressions for ˜ γ, even in S, it
follows that the function
˜
f has expansion in even negative powers of the semiclassical Casimir C.
Explicitly, the first few corrections read, for slow long strings
˜
f ≈

ln 8πC −1
π
+
ln 8πC + 1
16 π
3
C
2
+O(1/C
4
)

−J (D.3)
+π J
2

1
2 ln 8πC

3
32π
2
C
2
ln 8πC
+O(1/C
4
)

+... ,
where C = S +
1
2
J and dots indicate corrections in J. For fast long strings,
˜
f ≈
1
π
2
J

ln
2
ˆ
C
2
−ln
ˆ
C +
1
16
ˆ
C
2

4 ln
ˆ
C + 3 +
3
ln
ˆ
C
+
7
ln
2
ˆ
C
+...

+O(1/
ˆ
C
4
)


1
π
4
J
3

ln
4
ˆ
C
8
+
1
32
ˆ
C
2

4 ln
3
ˆ
C + 5 ln
2
ˆ
C + 9 ln
ˆ
C
+ 16 +
24
ln
ˆ
C
+
34
ln
2
ˆ
C
+...

+O(1/
ˆ
C
4
)

+... , (D.4)
where
ˆ
C =
C
J
=
S
J
+
1
2
and dots inside round brackets indicate corrections in 1/ ln
ˆ
C. As was
already noted in section 2.2, the expansion in the case of the fast long strings is not of the same
type as in (1.12) and (1.19) assumed in the main part of this paper.
42
The expansions are obtained from (2.11) and (2.12) after the redefinition η →−1 + 16η +
p
1 + 256 η
2
.
33
Let us mention also that in the case of the m-folded string the interval 0 ≤ σ < 2π is split
into 4m segments: for 0 < σ <
π
2m
the function ρ(σ) increases reaching its maximal value ρ
0
,
then decreases to zero for
π
2m
≤ σ ≤
π
m
, etc. This implies the condition
2π =


0
dσ = 4 m

ρ
0
0


2
−J
2
) cosh
2
ρ −(ω
2
−J
2
) sinh
2
ρ
, (D.5)
which leads to a factor of m in front of the relevant expressions for E, S,

κ
2
−J
2
. The large
spin expansion is then similar to the m = 1 case. Once E is expressed in terms of S and J, the
parameter m enters only in combination with the string tension

λ

.
Appendix E: Higher order relations from reciprocity at strong coupling
The evidence for the functional relation and reciprocity (1.14),(1.19) at weak coupling suggests
that the corresponding constraints should hold also in strong-coupling expansion. As we have
seen, the large spin expansion of anomalous dimensions at strong coupling appears to have the
same structure as at weak coupling (1.16) where now
f ≡

λ
¯
f ,
¯
f = a
0
+
b
0

λ
+
c
0
(

λ)
2
+... , (E.1)
f
c


λ
¯
f
c
,
¯
f
c
= a
c
+
b
c

λ
+
c
c
(

λ)
2
+... , (E.2)
f
mk
≡ (

λ)
m+1
¯
f
mk
,
¯
f
mk
= a
mk
+
b
mk

λ
+
c
mk
(

λ)
2
+... . (E.3)
Assuming the functional relation or (1.17), one is then able to compute the coefficients f
mm
of
ln
m
S
S
m
in terms of the strong-coupling expansion coefficients in the scaling function f. The latter
are known up to 2-loop order directly from the string-theory computations [10, 39]
a
0
=
1
π
, b
0
= −
1
π
3 ln 2 , c
0
= −
1
π
K , .... (E.4)
and also to a high (in principle, arbitrarily high) order from the analytic strong coupling solution
[37] of the BES [1] equation for the function f. This means that f
mm
are then effectively
determined if the functional relation applies.
Assuming the validity of the reciprocity condition (1.19) should lead to additional constraints
on the subleading coefficients like (1.22) which here should be understood in terms of power
series in
1

λ
. As a result, one should find non-trivial relations between strong-coupling expansion
coefficients in (E.2) and (E.3).
34
There is, however, a subtlety in formulating the reciprocity condition in the context of large
spin expansion at strong coupling as defined by string semiclassical perturbation theory where
all non-zero charges are automatically large at large λ. For example, the case of finite twist
J = 2, 3, ... can not be distinguished from the formal case of J = 0. It is usually assumed that
the folded string in AdS
5
with zero angular momentum in S
5
describes an operator of small
twist, but that can be J = 2 or J = 3, etc. To establish a relation to the definition of reciprocity
in weakly coupled gauge theory expansion with finite twist one would need to consider the case
of semiclassical (S, J) string and then resum the series for its energy (both in J and in

λ) so
that the limit of finite J would make sense.
Here we shall assume that in checking the reciprocity (1.19) at subleading order in strong
coupling at J = 0 one may simply take the Casimir C in (1.19) as C =

λC, C = S, and ignore
the shifts in brackets in (1.21) or (1.22), i.e. getting
¯
f
10
=
1
2
¯
f
¯
f
c
,
¯
f
32
=
1
16
¯
f (
¯
f
3
−2
¯
f
2
¯
f
c
−16
¯
f
21
) , ... . (E.5)
Equivalently, these relations follow from (1.21) by noting that at strong coupling f ∼ f
c


λ
and thus terms of order 1 or J ≪sqrtλ can be ignored.
Multiplying the series in (E.1)–(E.3) we then find that some of the 1-loop coefficients can be
expressed in terms of the tree-level coefficients and the coefficients in f. Explicitly,
b
11
= a
0
b
0
, b
10
=
1
2
(a
0
b
c
+a
c
b
0
) , b
22
= −
3
8
a
2
0
b
0
(E.6)
b
33
=
1
6
a
3
0
b
0
, b
32
=
1
8
a
3
0
(2b
0
−b
c
) −a
21
b
0

3
8
a
2
0
a
c
b
0
−a
0
b
21
, ... (E.7)
We have verified the validity of these relations for b
11
and b
10
in section 3.
Appendix F: Large S expansions for twist 2 and twist 3 anomalous dimensions
at weak coupling
Here we shall collect the coefficients of large spin expansion of anomalous dimensions for planar
SYM operators of twist 2 and 3, up to four loops in the gauge coupling and up to 1/S
3
order.
43
They are derived from the closed expressions in terms of the harmonic sums that were obtained
(mainly exploiting the maximum transcendentality principle and the asymptotic Bethe ansatz),
respectively, in [17] (at four loops) for the twist two scalar sector, in [51, 17] for the twist three
scalar sector and in [33] (at three loops) and [18] (at four loops) for the “gauge” sector.
All these expansions were proven to satisfy the reciprocity property, for a review see [18].
The expansions are indeed of the generic form (1.16) where the coefficients satisfy the relations
(1.21) once J (and the flavor index ℓ, see footnote 9 in the Introduction) are fixed accordingly.
43
In the case of twist 3 operators, the anomalous dimensions we will consider here are the minimal in the band.
35
The coefficients of the leading ln
m
S/S
m
terms below are manifestly universal in twist and
flavor.
44
As far as these leading terms are concerned, there is no need to explicitly write down
the results for the twist two gaugino and gauge sectors, and the twist three gaugino sector.
Indeed, the closed formulas for their anomalous dimensions can be deduced from the one for the
twist two scalar case by just shifting the argument of the harmonic sums
45
but such shifts do
not affect the coefficients of the leading ln
m
S/S
m
terms. It is worth stressing again that this
universality, a well-known feature of the leading ln S coefficient (or cusp anomaly), is a nontrivial
consequence of the functional relation (1.14), as was noticed in [33] and emphasized in [4].
At weak coupling it is useful to rewrite (1.16) as
γ(S)
S≫1
= f ln
¯
S +
¯
f
c
+
f
11
ln
¯
S +
¯
f
10
S
+
f
22
ln
2
¯
S +
¯
f
21
ln
¯
S +
¯
f
20
S
2
+
+
f
33
ln
3
¯
S +
¯
f
32
ln
2
¯
S +
¯
f
31
ln
¯
S +
¯
f
30
S
3
+O

ln
4
¯
S
S
4

, (F.1)
where
¯
S = e
γ
E
S and the coefficients will be power series in
ˆ
λ =
λ
16π
2
. Then one finds:
44
The only exception being the four loop coefficient of the term ln
2
S/S
2
in the case of twist two scalar operators.
However, it seems reasonable to relate this exception to the wrapping-induced breakdown of the Bethe equations
at four loops for twist two operators.
45
It is well known that in N = 4 SYM all twist two operators belong to the same supermultiplet, and
their anomalous dimension is expressed in terms of a universal function with shifted arguments γ
ϕ
J=2
(S) =
γuniv(S) , γ
ψ
J=2
(S) = γuniv(S +1) , γ
A
J=2
(S) = γuniv(S +2). In [52] it was proved that the anomalous dimension
for twist three operators built out of gauginos is related to the one of the twist two universal supermultiplet as
γ
ψ
J=3
(S) = γ
ϕ
J=2
(S + 2).
36
Twist two scalar sector:
f = 8
ˆ
λ −

2
3
ˆ
λ
2
+
88π
4
45
ˆ
λ
3
−(
584π
6
315
+ 64ζ
2
3
)
ˆ
λ
4
,
¯
f
c
= −24ζ
3
ˆ
λ
2
+ (
16
3
π
2
ζ
3
+ 160ζ
5
)
ˆ
λ
3
+ (−
56
15
π
4
ζ
3

80
3
π
2
ζ
5
−1400ζ
7
)
ˆ
λ
4
f
11
= 32
ˆ
λ
2

64 π
2
3
ˆ
λ
3
+
96 π
4
5
ˆ
λ
4
,
¯
f
10
= 4
ˆ
λ −

2
3
ˆ
λ
2
+ (
44π
4
45
−96ζ
3
)
ˆ
λ
3
+ (−
292π
6
315
+
160
3
π
2
ζ
3
−32ζ
2
3
+ 640ζ
5
)
ˆ
λ
4
,
f
22
= −64
ˆ
λ
3
+ (64π
2
−128ζ
3
)
ˆ
λ
4
, (F.2)
¯
f
21
= −16
ˆ
λ
2
+ (128 +
16 π
2
3
)
ˆ
λ
3
+ (−128π
2

32 π
4
15
+ 448ζ
3
)
ˆ
λ
4
,
¯
f
20
= −
2
3
ˆ
λ + (24 +

2
9
)
ˆ
λ
2
−(
32π
2
3
+
22π
4
135
−48ζ
3
)
ˆ
λ
3
+(
136π
4
15
+
146π
6
945
−384ζ
3

32
3
π
2
ζ
3
+
16ζ
2
3
3
−320ζ
5
)
ˆ
λ
4
f
33
=
512
3
ˆ
λ
4
,
¯
f
32
= 64
ˆ
λ
3
+ (−768 −
64 π
2
3
+ 128ζ
3
)
ˆ
λ
4
,
¯
f
31
=
16
3
ˆ
λ
2
+ (−256 +
16π
2
9
)
ˆ
λ
3
+ (512 +
512π
2
3

64π
4
15
−576ζ
3
)
ˆ
λ
4
,
¯
f
30
= −
56
3
ˆ
λ
2
+ (96 +
40π
2
9
−16ζ
3
)
ˆ
λ
3
−(
224π
2
3
+
32π
4
15
−800ζ
3
+
64
9
π
2
ζ
3

320ζ
5
3
)
ˆ
λ
4
Twist three scalar sector:
f = 8
ˆ
λ −

2
3
ˆ
λ
2
+
88π
4
45
ˆ
λ
3
−(
584π
6
315
+ 64ζ
2
3
)
ˆ
λ
4
,
¯
f
c
= −8 ln 2
ˆ
λ + (
8
3
π
2
ln2 −8ζ
3
)
ˆ
λ
2
+ (−
88
45
π
4
ln 2 +
8
3
π
2
ζ
3
−8ζ
5
)
ˆ
λ
3
+
8
315
(73π
6
ln2 −84π
4
ζ
3
+ 2520 ln 2ζ
2
3
+ 105π
2
ζ
5
+ 17325ζ
7
)
ˆ
λ
4
,
f
11
= 32
ˆ
λ
2

64 π
2
3
ˆ
λ
3
+
96 π
4
5
ˆ
λ
4
,
¯
f
10
= 8
ˆ
λ + (−

2
3
−32 ln 2)
ˆ
λ
2
+ (
88π
4
45
+
64
3
π
2
ln 2 −32ζ
3
)
ˆ
λ
3

8
315
(73π
6
+ 756π
4
ln 2 −840π
2
ζ
3
+ 2520ζ
2
3
+ 1260ζ
5
)
ˆ
λ
4
,
f
22
= −64
ˆ
λ
3
+ 64π
2
ˆ
λ
4
,
¯
f
21
= −32
ˆ
λ
2
+ (128 +
64π
2
3
+ 128 ln 2)
ˆ
λ
3
+ (−256 −128π
2

96π
4
5
−128π
2
ln 2 + 256ζ
3
)
ˆ
λ
4
,
37
¯
f
20
= −

3
+ (48 +

2
9
+ 32 ln 2)
ˆ
λ
2
+(32 −
80π
2
3

88π
4
135
−128 ln 2 −
64
3
π
2
ln 2 −64 ln
2
2 + 32ζ
3
)
ˆ
λ
3
+(−512 −
32π
2
3
+
352π
4
15
+
584π
6
945
+ 256 ln 2 + 128π
2
ln 2 +
96
5
π
4
ln 2
+64π
2
ln
2
2 −128ζ
3

64
3
π
2
ζ
3
−256 ln 2ζ
3
+
64ζ
2
3
3
+ 32ζ
5
)
ˆ
λ
4
,
f
33
=
512
3
ˆ
λ
4
, (F.3)
¯
f
32
= 128
ˆ
λ
3
−(768 + 128π
2
+ 512 ln 2)
ˆ
λ
4
,
¯
f
31
=
64
3
ˆ
λ
2
+ (−512 −
128π
2
9
−256 ln 2)
ˆ
λ
3
+(768 +
1408π
2
3
+
64π
4
5
+ 1536 ln 2 + 256π
2
ln 2 + 512 ln
2
2 −512ζ
3
)
ˆ
λ
4
,
¯
f
30
= −(
224
3
+
64 ln 2
3
)
ˆ
λ
2
+ (128 +
352π
2
9
+ 512 ln 2 +
128
9
π
2
ln 2 + 128 ln
2
2 −
64ζ
3
3
)
ˆ
λ
3
+(896 −
448π
2
3

512π
4
15
−768 ln 2 −
1408
3
π
2
ln 2 −
64
5
π
4
ln 2 −768 ln
2
2
−128π
2
ln
2
2 −
512 ln
3
2
3
+ 640ζ
3
+
128
9
π
2
ζ
3
+ 512 ln 2ζ
3

64ζ
5
3
)
ˆ
λ
4
.
Twist three “gauge” sector:
f = 8
ˆ
λ −

2
3
ˆ
λ
2
+
88π
4
45
ˆ
λ
3
−(
584π
6
315
+ 64ζ
2
3
)
ˆ
λ
4
,
¯
f
c
= 8(1 −ln2)
ˆ
λ +
8
3
(−12 −π
2

2
ln 2 −3ζ
3
)
ˆ
λ
2

8
45
(−1440 −60π
2
−11π
4
+11π
4
ln 2 −15π
2
ζ
3
+ 45ζ
5
)
ˆ
λ
3
+
8
315
(−100800 −3360π
2
−336π
4
−73π
6
+ 73π
6
ln 2 −84π
4
ζ
3
−2520ζ
2
3
+ 2520 ln 2ζ
2
3
+ 105π
2
ζ
5
+ 17325ζ
7
)
ˆ
λ
4
f
11
= 32
ˆ
λ
2

64 π
2
3
ˆ
λ
3
+
96 π
4
5
ˆ
λ
4
,
¯
f
10
= 32
ˆ
λ + (32 −
32π
2
3
−32 ln 2)
ˆ
λ
2
+
32
45
(−180 −30π
2
+ 11π
4
+ 30π
2
ln 2 −45ζ
3
)
ˆ
λ
3

32
315
(−10080 −840π
2
−189π
4
+ 73π
6
+ 189π
4
ln 2 −210π
2
ζ
3
+ 2520ζ
2
3
+ 315ζ
5
)
ˆ
λ
4
f
22
= −64
ˆ
λ
3
+ 64π
2
ˆ
λ
4
,
¯
f
21
= −128
ˆ
λ
2
+ (
256π
2
3
+ 128 ln 2)
ˆ
λ
3
+ (256 −
384π
4
5
−128π
2
ln 2)
ˆ
λ
4
¯
f
20
= −
200
3
ˆ
λ + (16 +
200π
2
9
+ 128 ln 2)
ˆ
λ
2
+ (480 −
16π
2
3

440π
4
27

256
3
π
2
ln2
−64 ln
2
2 + 128ζ
3
)
ˆ
λ
3
+ (−2816 −
1120π
2
3
+
64π
4
15
+
2920π
6
189
−256 ln 2
+
384
5
π
4
ln 2 + 64π
2
ln
2
2 −128ζ
3

256
3
π
2
ζ
3
+
1600ζ
2
3
3
+ 128ζ
5
)
ˆ
λ
4
38
f
33
=
512
3
ˆ
λ
4
, (F.4)
¯
f
32
= 512
ˆ
λ
3
+ (−256 −512π
2
−512 ln 2)
ˆ
λ
4
¯
f
31
=
1600
3
ˆ
λ
2
+ (−640 −
3200π
2
9
−1024 ln 2)
ˆ
λ
3
+(−1792 +
1792π
2
3
+ 320π
4
+ 512 ln 2 + 1024π
2
ln 2 + 512 ln
2
2)
ˆ
λ
4
¯
f
30
= 192
ˆ
λ + (−
1120
3
−64π
2

1600 ln 2
3
)
ˆ
λ
2
+ (−
5824
3
+
1856π
2
9
+
704π
4
15
+ 640 ln 2
+
3200
9
π
2
ln2 + 512 ln
2
2 −
1600ζ
3
3
)
ˆ
λ
3
+ (
25984
3
+
15488π
2
9

544π
4
3

4672π
6
105
+1792 ln 2 −
1792
3
π
2
ln 2 −320π
4
ln 2 −256 ln
2
2 −512π
2
ln
2
2 −
512 ln
3
2
3
+ 1152ζ
3
+
3200
9
π
2
ζ
3
−1536ζ
2
3

1600ζ
5
3
)
ˆ
λ
4
39
References
[1] N. Beisert, B. Eden and M. Staudacher, “Transcendentality and crossing,” J. Stat. Mech.
0701, P021 (2007) [arXiv:hep-th/0610251].
[2] L. Freyhult, A. Rej and M. Staudacher, “A Generalized Scaling Function for AdS/CFT,”
arXiv:0712.2743 [hep-th].
[3] A. V. Belitsky, A. S. Gorsky and G. P. Korchemsky, Logarithmic scaling in gauge / string
correspondence, Nucl. Phys. B 748, 24 (2006) [arXiv:hep-th/0601112].
[4] A. V. Belitsky, G. P. Korchemsky and R. S. Pasechnik, “Fine structure of anomalous
dimensions in N=4 super Yang-Mills theory,” arXiv:0806.3657 [hep-ph].
[5] G. P. Korchemsky, “Quasiclassical QCD pomeron,” Nucl. Phys. B 462, 333 (1996)
[arXiv:hep-th/9508025].
[6] S. S. Gubser, I. R. Klebanov and A. M. Polyakov, “A semi-classical limit of the gauge/string
correspondence,” Nucl. Phys. B 636, 99 (2002) [hep-th/0204051].
[7] H. J. de Vega and I. L. Egusquiza, “Planetoid String Solutions in 3 + 1 Axisymmetric
Spacetimes,” Phys. Rev. D 54, 7513 (1996) [hep-th/9607056].
[8] I.-Y. Park and A.A. Tseytlin, unpublished (2005).
[9] A. Tirziu and A. A. Tseytlin, “Quantum corrections to energy of short spinning string in
AdS5,” arXiv:0806.4758 [hep-th].
[10] S. Frolov and A. A. Tseytlin, “Semiclassical quantization of rotating superstring in AdS(5)
x S(5),” JHEP 0206, 007 (2002) [arXiv:hep-th/0204226].
[11] A. V. Kotikov, L. N. Lipatov, A. I. Onishchenko and V. N. Velizhanin, “Three-loop universal
anomalous dimension of the Wilson operators in N = 4 SUSY Yang-Mills model,” Phys.
Lett. B 595, 521 (2004) [Erratum-ibid. B 632, 754 (2006)] [arXiv:hep-th/0404092].
[12] S. Moch, J. A. M. Vermaseren and A. Vogt, “The three-loop splitting functions in QCD:
The non-singlet case”, Nucl. Phys. B 688, 101 (2004) [arXiv:hep-ph/0403192]; A. Vogt,
S. Moch and J. A. M. Vermaseren, “The three-loop splitting functions in QCD: The singlet
case”, Nucl. Phys. B 691, 129 (2004) [arXiv:hep-ph/0404111].
[13] Yu. L. Dokshitzer, G. Marchesini and G. P. Salam, “Revisiting parton evolution and the
large-x limit,” Phys. Lett. B 634, 504 (2006) [arXiv:hep-ph/0511302].
[14] G. Altarelli, “Partons In Quantum Chromodynamics,” Phys. Rept. 81, 1 (1982).
[15] B. Basso and G. P. Korchemsky, “Anomalous dimensions of high-spin operators beyond
the leading order,” Nucl. Phys. B 775, 1 (2007) [arXiv:hep-th/0612247]. G. Korchemsky,
“Anomalous dimensions of high-spin operators beyond the leading order”, talk at the 12th
Claude Itzykson Meeting, Saclay, June (2007).
[16] Yu. L. Dokshitzer and G. Marchesini, “N = 4 SUSY Yang-Mills: Three loops made sim-
ple(r),” Phys. Lett. B 646, 189 (2007) [arXiv:hep-th/0612248].
[17] A. V. Kotikov, L. N. Lipatov, A. Rej, M. Staudacher and V. N. Velizhanin, “Dressing and
Wrapping,” J. Stat. Mech. 0710, P10003 (2007) [arXiv:0704.3586 [hep-th]].
[18] M. Beccaria and V. Forini, “Reciprocity of gauge operators in N=4 SYM,” arXiv:0803.3768
40
[19] G. P. Korchemsky, “Asymptotics of the Altarelli-Parisi-Lipatov Evolution Kernels of Parton
Distributions,” Mod. Phys. Lett. A 4, 1257 (1989).
[20] L. F. Alday and R. Roiban, “Scattering Amplitudes, Wilson Loops and the String/Gauge
Theory Correspondence,” arXiv:0807.1889 [hep-th].
[21] M. Kruczenski, “A note on twist two operators in N = 4 SYM and Wilson loops in
Minkowski signature,” JHEP 0212, 024 (2002) [arXiv:hep-th/0210115].
[22] L. F. Alday and J. M. Maldacena, “Gluon scattering amplitudes at strong coupling,” JHEP
0706, 064 (2007) [arXiv:0705.0303 [hep-th]].
[23] M. Kruczenski, R. Roiban, A. Tirziu and A. A. Tseytlin, “Strong-coupling expansion of
cusp anomaly and gluon amplitudes from quantum open strings in AdS
5
xS
5
,” Nucl. Phys.
B 791, 93 (2008) [arXiv:0707.4254 [hep-th]].
[24] L. J. Dixon, L. Magnea and G. Sterman, “Universal structure of subleading infrared poles
in gauge theory amplitudes,” JHEP 0808, 022 (2008) [arXiv:0805.3515 [hep-ph]].
[25] L. F. Alday, talk given at “Strings 2008”, August 2008, CERN, Geneva.
[26] M. Beccaria, Yu. L. Dokshitzer and G. Marchesini, “Twist 3 of the sl(2) sector of N=4 SYM
and reciprocity respecting evolution,” Phys. Lett. B 652, 194 (2007) [arXiv:0705.2639 [hep-
th]].
[27] N. Dorey, “A Spin Chain from String Theory,” arXiv:0805.4387 [hep-th].
[28] V. M. Braun, S. E. Derkachov, G. P. Korchemsky and A. N. Manashov, “Baryon dis-
tribution amplitudes in QCD,” Nucl. Phys. B 553, 355 (1999) [arXiv:hep-ph/9902375].
S. E. Derkachov, G. P. Korchemsky and A. N. Manashov, “Evolution equations for quark
gluon distributions in multi-color QCD and open spin chains,” Nucl. Phys. B 566, 203
(2000) [arXiv:hep-ph/9909539].
[29] M. Beccaria and F. Catino, “Large spin expansion of the long-range Baxter equation in the
sl(2) sector of N=4 SYM,” JHEP 0801, 067 (2008) [arXiv:0710.1991 [hep-th]].
[30] V. M. Braun, G. P. Korchemsky and D. Mueller, “The uses of conformal symmetry in
QCD,” Prog. Part. Nucl. Phys. 51, 311 (2003) [arXiv:hep-ph/0306057].
[31] E. Laenen, L. Magnea and G. Stavenga, “On next-to-eikonal corrections to threshold re-
summation for the Drell-Yan and DIS cross sections,” arXiv:0807.4412 [hep-ph].
[32] N. Beisert and M. Staudacher, “Long-range PSU(2, 2|4) Bethe ansaetze for gauge theory
and strings,” Nucl. Phys. B 727, 1 (2005) [arXiv:hep-th/0504190].
[33] M. Beccaria, “Three loop anomalous dimensions of twist-3 gauge operators in N=4 SYM,”
JHEP 0709, 023 (2007) [arXiv:0707.1574 [hep-th]].
[34] M. Kruczenski, “Spiky strings and single trace operators in gauge theories,” JHEP 0508,
014 (2005) [hep-th/0410226].
[35] N. Beisert, S. Frolov , M. Staudacher and A. A. Tseytlin, “Precision spectroscopy of
AdS/CFT,” JHEP 0310, 037 (2003) [hep-th/0308117].
[36] S. Frolov, A. Tirziu and A. A. Tseytlin, “Logarithmic corrections to higher twist scaling at
strong coupling from AdS/CFT,” Nucl. Phys. B 766, 232 (2007) [hep-th/0611269].
41
[37] B. Basso, G. P. Korchemsky and J. Kotanski, “Cusp anomalous dimension in maximally
supersymmetric Yang-Mills theory at strong coupling,” Phys. Rev. Lett. 100, 091601 (2008)
[0708.3933 [hep-th]].
[38] R. Roiban, A. Tirziu and A. A. Tseytlin, “Two-loop world-sheet corrections in AdS
5
xS
5
superstring,” JHEP 0707, 056 (2007) [0704.3638 [hep-th]]. revised (v4) 08/07
[39] R. Roiban and A. A. Tseytlin, “Strong-coupling expansion of cusp anomaly from quantum
superstring,” JHEP 0711, 016 (2007) [0709.0681 [hep-th]].
[40] R. Roiban and A. A. Tseytlin, “Spinning superstrings at two loops: strong-coupling cor-
rections to dimensions of large-twist SYM operators,” Phys. Rev. D 77, 066006 (2008)
[arXiv:0712.2479 [hep-th]].
[41] P. Y. Casteill and C. Kristjansen, “The Strong Coupling Limit of the Scaling Function
from the Quantum String Bethe Ansatz,” Nucl. Phys. B 785, 1 (2007) [arXiv:0705.0890
[hep-th]].
[42] N. Gromov, “Generalized Scaling Function at Strong Coupling,” JHEP 0811, 085 (2008)
[arXiv:0805.4615 [hep-th]].
[43] N. Drukker, D. J. Gross and A. A. Tseytlin, “Green-Schwarz string in AdS(5) x S(5):
Semiclassical partition function,” JHEP 0004, 021 (2000) [arXiv:hep-th/0001204].
[44] S. Schafer-Nameki and M. Zamaklar, “Stringy sums and corrections to the quantum string
Bethe ansatz,” JHEP 0510, 044 (2005) [arXiv:hep-th/0509096]. S. Schafer-Nameki, “Exact
expressions for quantum corrections to spinning strings,” Phys. Lett. B 639, 571 (2006)
[arXiv:hep-th/0602214]. S. Schafer-Nameki, M. Zamaklar and K. Zarembo, “How accurate
is the quantum string Bethe ansatz?,” JHEP 0612, 020 (2006) [arXiv:hep-th/0610250].
[45] A. A. Tseytlin, “Spinning strings and AdS/CFT duality,” hep-th/0311139.
[46] F. A. Dolan and H. Osborn, “On short and semi-short representations for four dimensional
superconformal symmetry,” Annals Phys. 307, 41 (2003) [hep-th/0209056]. G. Arutyunov
and E. Sokatchev, “Conformal fields in the pp-wave limit,” JHEP 0208, 014 (2002) [hep-
th/0205270].
[47] M. Kruczenski and A. A. Tseytlin, “Spiky strings, light-like Wilson loops and pp-wave
anomaly,” Phys. Rev. D 77, 126005 (2008) [0802.2039 [hep-th]].
[48] L. F. Alday and J. M. Maldacena, “Comments on operators with large spin,” JHEP 0711,
019 (2007) [0708.0672 [hep-th]].
[49] R. R. Metsaev, C. B. Thorn and A. A. Tseytlin, “Light-cone superstring in AdS space-time,”
Nucl. Phys. B 596, 151 (2001) [arXiv:hep-th/0009171].
[50] G. Arutyunov and S. Frolov, “Uniform light-cone gauge for strings in AdS(5) x S5: Solving
su(1|1) sector,” JHEP 0601, 055 (2006) [arXiv:hep-th/0510208].
[51] M. Beccaria, “Anomalous dimensions at twist-3 in the sl(2) sector of N = 4 SYM,” JHEP
0706, 044 (2007) [arXiv:0704.3570 [hep-th]].
[52] M. Beccaria, “Universality of three gaugino anomalous dimensions in N = 4 SYM,” JHEP
0706, 054 (2007) [arXiv:0705.0663 [hep-th]].
42
[53] N. Gromov, S. Schafer-Nameki and P. Vieira, “Efficient precision quantization in
AdS/CFT,” arXiv:0807.4752 [hep-th].
N. Gromov and P. Vieira, “The AdS(5) x S5 superstring quantum spectrum from the alge-
braic curve,” Nucl. Phys. B 789 (2008) 175 [arXiv:hep-th/0703191].
[54] D. Bombardelli, D. Fioravanti and M. Rossi, “Large spin corrections in N = 4 SYM sl(2):
still a linear integral equation,” arXiv:0802.0027 [hep-th].
[55] R. Ishizeki, M. Kruczenski, A. Tirziu and A.A. Tseytlin, Spiky strings in AdS
3
× S
1
and
their AdS-pp-wave limits, ArXiv:0812.2431.
43

1

Introduction and summary

superstring theory which utilise their integrability property led to important insights into the structure of dependence of anomalous dimensions of gauge-invariant operators on the quantum numbers like spin and on the t’Hooft coupling. While there was a remarkable recent progress in understanding the asymptotic large spin limit in which the compactness of the spatial direction of the world sheet may be ignored [1, 2],1 it is important to study corrections to this limit. Here we shall consider the famous example [6] of folded spinning string in AdS5 dual to a minimal twist gauge theory operator like tr(ΦDS Φ). Starting with the classical string energy + for the solution of [7, 6] and expanding it in large semiclassical spin parameter S one finds (see [8] and below) E= √ λ E(S) , S S=√ , λ 1 = S + a0 ln S + ac + (a11 ln S + a10 ) S ln3 S 1 + 2 (a22 ln2 S + a21 ln S + a20 ) + O( 3 ) , S S
√ λ 2π

Recent advances in the study of duality between planar N = 4 SYM theory and free AdS5 × S 5

(1.1)

E(S)S≫1

(1.2)

1 1 with a0 = π , ac = π (ln 8π − 1), etc.2 That means that in the semiclassical string theory limit

in which one first takes the string tension

to be large for fixed S and then expands in large √ S S, the corresponding string energy can be written as ( λ ≫ 1, √ ≫ 1)
λ

E = S + f ln S + fc + + where f = √ λ a0 + .., fc =

1 [f11 ln S + f10 ] S (1.3)

√ λ ac + ..., etc. The subleading coefficients simplify if we absorb

1 ln3 S [f22 ln2 S + f21 ln S + f20 ] + O( 3 ) , S2 S

the constant fc into the ln S term, i.e. if we re-write (1.3) as 1 ′ ˜ ˜ E = S + f ln(S/fc ) + [f11 ln(S/fc ) + f10 ] S ln3 S 1 ′ ′ ˜ ˜ + 2 [f22 ln2 (S/fc ) + f21 ln(S/fc ) + f20 ] + O( 3 ) , S S
1 where to leading order in √ expansion λ √ √ λ e λ λ λ3/2 ′ ˜ , fc = , f11 = 2 , f10 = 0, f22 = − 3 , f= π 8π 2π 8π
1

(1.4)

′ f21 =

5 λ3/2 , 16π 3

′ f20 =

λ3/2 8π 3

Here we refer to the integral equations that describe the minimal anomalous dimension in the band [3, 4].

These equations were obtained from the all-loop Bethe ansatz by taking a special scaling limit [5, 3] which describes a condensation of magnons and holes at the origin. √ 2 Note that the small S behaviour of the energy is quite different [9]: E = 2S[h0 + h1 S + ...].

2

Following the analysis of quantum corrections to the folded string solution in [10], one may conclude that this structure of the large S expansion is preserved by the α′ ∼ with the coefficients f, fc , f11 , ... being promoted to power series in
1 √ , λ 1 √ λ

corrections,

i.e. fmk ∼

Indeed, as we shall find below, the 1-loop corrections for leading coefficients in √ 3 ln 2 λ 1 1 − √ + O( ) , f= π λ λ √ e λ 1 1 ˜ fc = 1 + √ (3 ln 2 − c) + O( ) , 8π λ λ 6 ln 2 λ 1 f11 = 2 1 − √ + O( ) , 2π λ λ λ 0 1 ′ f10 = 2 0 − √ + O( ) . 2π λ λ

bmk,n √ n ( λ)n . (1.4) are3

(1.5) (1.6) (1.7) (1.8)

Here c is a constant that we were not able to determine with the method for evaluation of 1-loop string correction we used below. Equivalently, in (1.3) we get the same f , f11 and √ 8π 1 1 8π λ 1 = ln √ − 1 + √ − 3 ln 2 ln √ + c + O( ) , fc = f ln ˜ π λ fc λ λ λ fc 1 ′ ′ = f10 + f11 f10 = f10 + f11 ln ˜ f fc 8π 1 1 λ 1 8π = ln √ − 1 + √ − 6 ln 2 ln √ − ] + c + O( ) . 2 2π λ λ λ λ 2 relation f10 −
fc f f11

(1.9) (1.10) (1.11)

′ The (at first surprising) vanishing of the first two terms in f10 in (1.8) is a consequence of the

0. As we shall see below, these relations are consequences of the “functional relation” and reciprocity at strong coupling. Note that these conditions thus determine 2 our of 4 coefficients
′ in the part of E up to order O( ln 2S ). We thus led to expect that f10 = 0 should be true in to S
2

1 In fact, since we also see that in (1.5),(1.7) f11 = 1 f 2 , this relation is equivalent to f10 − 2 fc f = 2

= 0 (here verified at first two leading orders, i.e. to order O(λ0 ) in f10 ).

all orders in the strong coupling expansion, suggesting the advantage of the form of E in (1.4) ˜ over (1.3) and the importance of the function fc . Reversing the usual logic, we may then conjecture that structurally same large spin expansion should appear also at weak coupling, i.e. in the perturbative expressions for the corresponding gauge theory anomalous dimensions. This is not, a priori, guaranteed since the limit taken on the gauge theory side is different from the above string-theory limit: there one first expands the anomalous dimension in small λ at fixed S and then takes S large in each of the λn coefficients. Yet, remarkably, expanding in large S the known 2-, 3- and 4-loop perturbative anomalous dimensions of twist 2 and twist 3 operators in SYM theory one does find [11, 12, 8, 13, 15,
3

The leading correction to f was found in [10].

3

16, 17, 18] the expression of the form (1.3) with the coefficients given by power series in λ, i.e. fmk ∼
n 4 n amk,n λ .

Assuming that the expansion (1.3) or5 E−S =
∞ m=0

em (λ, ln S) , Sm

em (λ, ln S) =
k

fmk (λ) lnk S

(1.12)

applies for any λ, and given the important role of the universal scaling function or cusp anomalous dimension f (λ) [19], one may raise the question about the interpretation of other “interpolating” functions fmk (λ) in (1.3). From the gauge theory point of view, the function f (λ) appears in the asymptotics of anomalous dimensions of gauge invariant operators as well as (for twist 2) in the IR asymptotics of gluon scattering amplitudes related to UV cusp anomaly of light-like Wilson loops (for a review and references see, e.g., [20]). On string side that corresponds, respectively, to the closed string [6] and the open string [21, 22] sectors. They are connected in the strict large S limit since then the ends of the folded spinning string reach the boundary of AdS5 and thus the associated world surface has a Wilson line interpretation [23]. This open string sector interpretation should not be expected to apply to other subleading coefficients fmk (λ) since for finite S the end points of the folded string no longer touch the boundary (cf., however, [24, 25] where the gauge theory interpretation of the constant fc is discussed). In fact, many of the fmk coefficients in (1.3),(1.12) are not actually independent, as was first observed at few leading orders in weak coupling expansion and then given a general interpretation in [15, 16]. According to [15], these coefficients are constrained by (i) the so called “functional relation” suggested by the conformal invariance (which relates the leading fmm functions to powers of the scaling function f and thus implies their universality) and also by (ii) the “parity preserving relation” or “reciprocity” [12, 15, 16, 26, 18] (which relates some subleading nonuniversal coefficients, e.g., f10 to fc , f32 to f21 , etc.).

Our aim here will be to investigate the presence of such relations at strong coupling, i.e. in the semiclassical string theory expansion for the spinning string states, extending earlier observations made in [15].6
4

The 4-loop prediction for twist 2 and 3 anomalous dimension at finite S [17] so far was not based on direct

gauge theory computation. 5 Here, as in (1.2),(1.3), we suppress the dependence on finite twist J. 6 Let us note also that the fine structure of the constant term fc in (1.3) for “non-minimal” operators and the dual string states was studied in [3, 4] and in [27].

4

.. J. in the string-theory semiclassical expansion. 10 Here we assume that Φ in the operator tr(DS ΦJ ) is a scalar field (as is the case in the sl(2) sector of SYM + theory). Let us now review in more detail what is known at weak coupling (see [15] and references there). Let us mention also that coefficients in large S expansion beyond cusp anomaly one may be also controlled by integral equations like the BES [1] one (see in this connection [54]). that then leads to the following “functional relation” for γ 1 γ(S. 8 We thank A. Rej for a discussion of the structure of these terms at 1-loop order in sl(2) sector (see also [3]). The results of explicit higher-loop planar gauge-theory computations of anomalous dimensions γ(S. For a general method to derive higher order terms in 1/S expansion at fixed J see [29]. k2 = k23 J3 + k22 J ln S k2 (J) ln2 S Jk lnm S . Thus the dependence on J and S is different at strong and weak coupling and to relate the two expansions one needs a non-trivial resummation rather than simple interpolation of coefficients in λ [40. 2 argue that the anomalous dimension γ = ∆ − S − J should be a function of S only through its (1. Observing that such Wilson-type operators can be classified according to representations of the collinear SL(2. where for k1 = J2 + k21 J + k20 and k2 vanishes for J = 2.) ln S. i. We thank G.. While for low twists J = 2.. i.. Namely. since. Rej. one cannot distinguish between finite values of J and J = 0. 4) conformal group [30] which are labeled by the conformal spin s = 2 (S + ∆) one may dependence on the conformal spin s. private communication). this relation is nothing more than a change of variable. e0 = (k11 j + k23 j 3 + . J → C and j → s. non-minimal dimensions even for low twists. (J 9 There is a numerical evidence for the presence of k22 S) term for J ≥ 4 from the analysis of the corresponding ln 5 . J) = f S + 2 J + 1 γ(S.14) Without further information.13) (1. 3 in weak-coupling expansion the powers of ln S in em in the anomalous dimension (1.9 At strong coupling.) ln S. J . the terms appear [28] in 1-loop Baxter equation (A. J) = f(s.e.8 The presence of such terms was first observed in [3] (in the 1-loop approximation in the sl(2) sector) in the limit likely to be present also for finite J with S ≫ 1. 2]. Korchemsky for this clarification. for large J one finds (see [3] and section 2. 3. it is always possible to compute the function f in terms of the 7 J3 ln2 S Here we have in mind the minimal anomalous dimension in the band [3]. The relation between the notation used in [15] and ours is: N → S.2 below) that in a similar limit of large S with ℓ ≡ √ J λ ln S of large J and S with j ≡ fixed and small. J). Since the scaling dimension ∆ is10 ∆ = S + J + γ(S.7 for J > 3 one finds also terms with negative powers like J > 3 the leading term in E − S − J appears to be e0 = k1 (J) + k11 J + k10 . for + . R) subgroup of the 1 SO(2.e. L → J..12) appear to be positive.Before proceeding let us add an important clarification. and they are fixed e0 = (n1 ℓ2 + n2 ℓ4 + .. at least in perturbation theory. J) . λ) of operators like tr(DS ΦJ ) (S is the Lorentz spin and J is the twist) were + interpreted in [15] in the following way (see also [16]).

.. 24 (1. 4] f 2 ln S f 3 ln2 S f 4 ln3 S − + + .15) expanding the coefficients γn in large S (for fixed J) one finds that for all explicitly known perturbative gauge-theory results one gets the same (1.. which recently received a nice confirmation in [31]. it does not contain lnm S/S m terms [16. the lower bound of which is the minimal dimension for given S and J. Remarkably.. = f ln S + The universality (i. 8 f33 = 1 4 f . 2 S 8 S2 24 S 3 (1.3) γ(S)S≫1 = f ln S + fc + + f33 f11 ln S + f10 f22 ln2 S + f21 ln S + f20 + + S S2 ln4 S ln3 S + f32 ln2 S + f31 ln S + f30 +O . Let us note that anomalous dimensions of operators with twist higher than two occupy a band [3].14) [13]. the function f for twist 2 operators is closely related to a special reformulation of the parton distribution functions evolution equation which aims at treating symmetrically the space-like channel of deep inelastic scattering and the time-like crossed channel describing e+ e− annihilation [14]. this conclusion is also expected on the basis of the QCD origin of the functional relation (1. for twist 2 case.16) as they are simply proportional to f m+1 .14) turns out to be predictive within the large S expansion (1. In particular. this relation follows from a suitable modification of the evolution equations governing the renormalization of the twist operators. 1 f11 = f 2 . are power series in λ.17) 1 γ(S) = f ln S + 2 f ln S + . 26]. Nevertheless.e. Suppressing the dependence on J in γ and f we may write this functional relation simply as12 γ(S) = f S + 1 γ(S) . fc . twist and flavor independence) of the scaling function or cusp anomalous dimension f thus implies the universality of all of the coefficients fmm in (1. 16].16) where the coefficients f. f11 . the structure of this expansion turns out to be perfectly consistent with the functional relation (1.12) because the function f happens to be simpler than γ and.. .. J). in particular.. i.. 15.anomalous dimension γ(S.. The relation 11 1 f22 = − f 3 .. 2 power series in λ. 2 At weak coupling γ(S) = expansion as in (1. S3 S4 ∞ n n=1 γn (S)λ . + . (1. In that context.e. 12 In the approach of [13. 26. . the functional relation (1. 6 . . .18) These should be understood as relations between the functions fmm (λ) and f (λ) defined as Apart from conformal invariance. has no coefficients of the leading lnm S Sm lnn S Sn terms in it) so that the terms are all determined by the scaling function f [15. the above reasoning suggests that f could be more fundamental than γ.15): the function f starts with a logarithmic term (and is “simpler” than f .11 This is what we shall assume below when referring to the functional relation.

15 ˜ Note that using this expression for f10 and defining fc = e−fc /f to put (1.14 2 1 C 2 = (S + 1 J)(S + 2 J − 1) . it was observed on many examples that the function f should satisfy a “parity preserving relation” or reciprocity property. S controlled by special properties of the function f in the functional relation (1.32)). for twist 2 case. f55 . as was found at weak coupling.(1.17). For twist J = 2 we get simply 1 f10 = f (fc + 1) . [15] f(S) = an (ln C) .20) The reciprocity property (1.e.19).4) one finds √ ′ 1 that f10 = 2 f (−1 + J) (which. For twist 2. .. 16 (1. their coefficients are. Indeed. 2 13 f32 = 1 f [f 3 − 2f 2 (fc + 1) − 16f21 ] . in terms of R1 F (x) = −x F (1/x). k < m) terms in (1. at strong coupling. J = 2.. where f(S) = 0 dx xS−1 F (x). namely. S + 1 J → S + ℓJ.22) the Mellin transform: preserving relation” was suggested in [15] as a more appropriate name. Here for simplicity we shall not make this distinction and will often refer to reciprocity when implying the “parity invariance” property in (1. 1 + O( S ). .3 (see (2.17) is expected to apply for the minimal dimension in the band. C 2n n=0 ∞ (1. so that C = S + 1 2 The name reciprocity has its origin in the the formulation of this property.(1.20) below. at least partially. This is also what we shall see at strong coupling on the example of the spiky string in section 2. 1. 2 (1. 2 2 3 2 7 . s0 = 2 (S + ∆0 ) = S + 1 J . i. spinor or vector cases [30].16): f10 = 1 f (fc − 1 + J) .(1.16). These results are indeed consistent with (1. 13 k This property implies that the large S expansion of f(S) should run in the inverse even powers of the quadratic Casimir of the collinear SL(2. For twist larger than 2 the “parity for a scalar. f30 .. .16). f50 . As for the subleading ( ln mS . i.15 Again.20). is subleading to f11 term unless J ∼ λ).e. . R) group. 2 f32 = 1 16 f [f 3 − 2f 2 (fc − 1 + J) − 16f21 ] . one then has C 2 = S(S + 1). (1.19) where C is the “bare” quadratic Casimir defined in terms of the “canonical” value of the con1 formal spin s0 as C 2 ≡ s0 (s0 − 1)..18) and thus with the universality of the fmm coefficients in (1.15).16) into the form (1. these equations relate functions fmk (λ) defined as power series in λ. For generic flavour one is to replace. as obtained from the asymptotic Bethe ansatz of [32]. Interestingly.19) of the function f in the relation (1. etc. In Appendix F below we will summarize the known perturbative expansions for the minimal anomalous dimensions of twist 2 and twist 3 operators of various flavors. where ℓ = 1 . in (1. a similar relation also holds for the excited trajectories [4]..15) then imposes constraints on some of the coefficients of the subleading terms in the expansion (1.21) where dots stand for similar expressions for f31 . 14 Here we again consider the operator built of scalar fields...

The reason for the reciprocity property on the string theory side is even far less clear. they should then also be visible at strong coupling [15].15) for the anomalous dimensions of Wilson-type operators on the gauge theory side was argued [15] to follow from the invariance under the collinear SL(2. 4]. A four-loop test for the twist 3 anomalous dimension in the sl(2) sector was performed Three-loop tests of reciprocity for QCD and for the universal twist 2 supermultiplet in N = 4 SYM were in [26]. it could then be checked also at strong coupling on the example of the spiky string solution of [34]. terms in E in large S. as we will review in Appendix A. in the large spin expansion of the corresponding semiclassical string energies. 16 discussed in [15. However.These so called MVV [12] relations were first observed for twist 2 QCD anomalous dimensions up to 3 loops. while the semiclassical string-theory limit is to expand in large λ with fixed S = S √ λ and then expand the √1 ( λ)n limits commute (which so far appears to be verified only for the leading universal ln S term) the (1. The agreement in the structure of the large S expansion found in perturbative gauge theory and in perturbative string theory is already quite remarkable. If that were the case. with the reciprocity property of the function f. Since the planar perturbation theory should be convergent. the realization of the conformal group on states represented by classical spinning string solutions in global AdS5 coordinates is a priori different from the one used on the gauge-theory side (which is based on the collinear subgroup). so that the direct connection is not obvious.15) and. 16 It is natural to expect that the functional relation and the reciprocity property should hold also at higher orders in small λ expansion. one may think that it should then apply also on the string theory side. The case of twist 3 gauge field strength operators was analyzed in [33] (at three loops) and in [18] (at four loops). R) subgroup of the conformal SO(2.e. 8 . 4) group. i.e. moreover.19) is obscure on the semiclassical string theory side. 16]. i. of the reciprocity property The functional relation (1. In the latter paper it was also proved that even the wrapping-affected four loop result for the twist two operators [17] is reciprocity respecting. This agreement is non-trivial since. The large S expansions for the known twist 2 and twist 3 SYM anomalous dimensions that we will present in Appendix F are indeed consistent with these relations. Given that this argument is based on the conformal symmetry. the gauge-theory and string-theory perturbative expansions are organized differently: the gauge-theory limit is to expand in small λ at fixed S and then expand the λn coefficients in large S. One may also wonder if the reciprocity property may apply to higher twist operators above the lower bound of the band [3. Even assuming these reason for the validity of the functional relation (1. as was already mentioned.

3) of the large spin expansion and. In section 2 we shall first consider the large spin expansion of the classical energy of folded spinning string in AdS5 and show that the large spin expansion has the structure (1.2) the dependence on the angular momentum J in S 5 in the “long string” limit (J ≪ S). In Appendix F we shall summarize the known weak coupling planar SYM results for the large spin expansion of twist 2 and 3 anomalous dimensions up to 4-loop order in the ‘t Hooft coupling.If one identifies the energy E and the spin S of a string rotating in a plane in global AdS5 with dimension and Lorentz spin of the gauge theory operator like tr(DS ΦJ ). In Appendix B and C we shall review the folded spinning string solution and discuss long-string or large-spin expansions used in the 1-loop computation in section 3. the fact that the corresponding operator has higher than minimal dimension for a given spin. E − S − J = f(E + S. J) string considered in section 2. pointing out a subtlety in the definition of the latter in the semiclassical string expansion. (1.e. 15]). the functional + 1 relation (1. assume that J is negligible compared to S) and compute the 1-loop correction to the energy expanded in large S. In section 3 we shall return to the case of the folded spinning string in AdS5 (i. moreover. 9 in this case we shall find that the reciprocity condition is violated which should be related to . we shall verify that the string 1-loop corrections preserve the structure (1. In Appendix E we shall discuss some consequences of the functional relation and the reciprocity at strong coupling.23) As we shall discuss below (extending earlier observations in [8.2) and the functional and reciprocity relations between the coefficients are satisfied. not only the structure of the large spin expansion on the string theory side happens to be the same as on the gauge theory side but also its coefficients are indeed consistent with the functional relation and the reciprocity for the minimal dimension case represented by the folded spinning string. that the reciprocity condition is satisfied beyond the string tree level. The rest of this paper is organized as follows. determining corrections to several leading coefficients. We shall also show that the functional relation but not the reciprocity appears to apply also to the case of the classical spiky string solution. This will be demonstrated at the classical as well as 1-loop string theory level. i. In section 2. In Appendix D we shall give details of large spin expansions for (S.3 we shall study the same large spin expansion for a spiky string in AdS5 . J) . As result. We shall then include (in section 2.e.15) would then imply that γ = E − S − J should be a function of s = 2 (E + S). In Appendix A we shall make some comments on relation between different realizations of conformal group.

1 Large spin expansion: classical string theory Folded spinning string with J = 0 We shall start with a discussion of the limit when the S 5 momentum J of the string state can be ignored.. For such long string one has η → 0.2) (or from the form of the exact solution in global AdS5 The functional relation (1. (1.1) (2.23) implies that E − S should be a function of E + S. (2. η √ . ... + . with S the leading term in the function f being simply the logarithm (cf. Solving (2.. we shall concentrate only on the AdS5 spin S dependence of the string energy. We review the folded spinning string solution [6] in Appendix B. one can express the conserved charges in terms of the hypergeometric functions as in Appendix B. The large S expansion of the function f (its leading term in the strongcoupling limit) is much simpler than that of the anomalous dimension E −S in (2. which can be expressed in terms of the elliptic functions E and K of an auxiliary variable η E −S = S = 2 π 2 π 17 1+η 1 E − η η 1 1+η E − η η 1 1 −1 +K − η 1+η 1 −K − . Still.19).(1.. π 2 π coordinates) why such a relation should be natural for any value of S. indeed. This is the limit when the twist of the gauge theory operator is sufficiently small compared to the Lorentz spin.e.2)..3) as was already claimed in (1. The integrals of motion are √ √ the energy E = λ E and the spin S = λ S.1) and contains 17 Furthermore. 10 .1) one finds for E as a function of S E =S+ to the boundary of AdS5 ). Equivalently.4) property (1.17)) √ √ λ 1 λ E−S = ln S + ln S + . It the leading ( ln S )m terms in (2. Here we are interested in the large spin expansion which corresponds to the long string limit (when the string ends are close substituting it into (2. (2.2) for small η and ¯ ¯ ¯ ¯ ln S − 1 ln S − 1 2 ln2 S − 9 ln S + 5 + − 2S 3S 2 π 2π 16 π 3 ¯ 2 ¯ ¯ − 14 2 ln S − 18 ln S + 33 ln S + + .15). the coefficients of (2.22). i. . to be consistent with such a relation. it is possible to verify that the expansion of E − S also satisfies the reciprocity Here we follow the notation of [35].3) happen. 48 π 4 S 3 ¯ S ≡ 8πS .(1.2 2. is not immediately obvious from (2.2) To find the energy in terms of the spin one is to solve for η.

one can find the parameter ˜ in (2. concluding that it is given by a power series in odd negative powers of The definition of reciprocity condition in string semiclassical expansion is discussed in Appendix E.only even powers of C −1 ∼ S −1 (see (1. 8πη 2π 2π (2. ω − S + 1 γ (ω) 2˜ (2. we find that the MVV-like relations (1.6) with the expression (2. π π π (2.5) Equivalently. This relation also arises by expanding the denominator k=1 k! − 2 ∂S 4 11 .8) while the expression for the conformal spin runs in odd powers of η s(η) = ˜ 1 11 + 2 ln η 877 + 92 ln η 3 + η− η + . λ (2. Then defining x = S − 1 ˜ 19 2 2 18 From the equation for the pole of the integrand in (2.19 It is natural to replace the variable ω in (2. .. ˜ 20 This choice is not unique.6) in small γ and integrating the resulting series. An analogous transformation was used in [15]..18 A more systematic analysis of the reciprocity (parity invariance) property of the function f is possible with the help of an integral representation for it. s − S = 0. γ (S) = E − S. The expression that multiplies γ in the integrand has residue 1. The integral then gives the function γ evaluated at the zero of the ˜ denominator.2) for the semiclassical spin S(η) 1 ˜ f(S) = 2π i 1 where s(η) ≡ S(η) + 2 γ (η) = ˜ ˜ 1 2 (E dη γ (η) ˜ Γ s′ (η) ˜ . this is the same as the statement that the anomalous dimension as a function of ˜ the Lorentz spin is. so that the integral is γ evaluated at the ˜ ˜ ˜ f(S) we have 2S − 2x = γ which coincides with the equation for ˜ pole ω = S − 1 γ . f ¯ 1 ln S + 1 ˜ ¯ f(S) = ln S − 1 + +O π 16π 2 S 2 1 S4 1 + O( √ ) .. where S ′ = S + 1 γ (S).. s(η) − S ˜ (2.9) η in terms of the spin S. one can formally reconstruct it from γ using [15] f(S) = P∞ 1 ` 1 ´k−1 1 [γ(S)]k = γ − 1 (γ 2 )′ + 24 (γ 3 )′′ + · · · .7) it is useful to redefine the variable η as20 η → −1 + 16η + expressions. Using that (1.22) are satisfied. effectively.20)) f= √ λ˜. and renaming S ′ → S we have f(S ˜ ˜ γ S ˜ 2 2 1 ˜ f(S) = 2π i dω γ (ω) ˜ Γ 1 + 1 γ ′ (ω) 2˜ . One finds that γ (η) is a series in even powers of η ˜ γ (η) = − ˜ 1 + 256 η 2 and examine the large S or small η limit of the 1 + ln η 4(ln η + 12) 2 6(62 ln η + 777) 4 + η − η + . Note that assuming f exists.7) semiclassical quantities. the pole with x = ω.15) implies ˜ ′ ) = f(S ′ − 1 ˜ ′ ) .6) where the contour Γ encircles the pole of the integrand and prime stands for derivative. + S) is the “conformal spin” expressed in terms of the To verify the reciprocity property of the function ˜ f(S) in (2. . a function of the conformal spin s.7).

3)) E −S −J ≈ 1 1 π3 J 4 πJ2 ¯ (2.20))..e.2 Folded spinning string with J = 0 Let us now consider the case when the S 5 angular momentum of the string is not negligible compared to S. π S S ¯ 8π S . ω ω2 − J 2 ≡1+η . will apply (see also [10. going in the inverse powers of S with the coefficients being polynomials In the large S ≫ J or long string limit.. 15]).16). ¯ S π 2 ln S 4 ln S 4 1 3 1 2 ¯ ¯ − ¯2 (2 ln2 S − 9 ln S + 5) + π J 2 1 + ¯ − ln2 S − ln3 S + ..S. which is same as γ (η) evaluated at the pole. In this case it is possible to show again that the large S expansion is consistent with the 2. (2..19). one should distinguish between “small” J ln S We will be interested in large S expansion with S ≫ J since only in this case the expansions in ln S.11) (2. should also run only in even ˜ negative powers of S or C = C √ λ (cf. 35]: E =κ+ κ2 − J 2 = κ S . (1. ¯ ¯ π 2 ln S 4 1 2 3π 3 J 4 πJ2 ¯ + ¯ (ln S − 1) + − 2 ¯ 4 ¯ 1 − 3 ln S + . The leading terms in the large S expansion of the energy of a string with m folds are (see m 4 4 ¯ ¯ ¯ ¯ ln S − 1 + ¯ (ln S − 1) − ¯2 (2 ln2 S − 9 ln S + 5) + . π η Here κ and ω (or η) are parameters of the classical solution which should we eliminated to find E as a function of S and J . ˜ f(S). (2. i.. . ¯ ¯ S π 2 ln S 12 . when η ≪ 1.. or “large” J cases [10.. The corresponding charges are the energy E = λ E and the two angular momenta √ √ S = λ S and J = λ J [10.(1.12) 2 1 √ K −η . The above discussion has a straightforward generalization to the multifolded spinning string Appendix D) E −S = case.13) − (ln S − 1) + ¯ 8 ln3 S 1 − ln S + .. In the “slow long string” approximation (corresponding to taking S to be large with ℓ ≡ fixed and then expanding in powers of ℓ) the leading terms in the semiclassical energy read (cf. κ2 − J 2 √ 2π ηω 1 1 S=√ E −η − K −η κ2 − J 2 (2. like (1. when the string state is dual to an operator with large spin S and large √ twist J. As a result. i.e. 36].10) S≡ m reciprocity property.

14) − − 3 ln2 S + ln S + 1 + ˆ ln2 S ˆ ˆ 8 ln S S 11 2 13 ˆ ˆ ˆ + − + .15) ≪ 1 plays the role of an expansion parameter. 21 J2 ln S Note that the leading terms in expression of the previous subsection (2.. . in the fast long string case (2. while the “semiclassical” value of the Casimir operator in (1.12)... also its discussion in Appendix D) is somewhat outside To study the properties of the subleading corrections. Then the s′ (η) ˜ .14) we get higher powers of ln S not our main theme here. and so this case (cf. After a redefinition of η one can then show that the expansion of f in large C runs only in even long strings.. ˜ 2 2˜ structure as in (1. (2.. ln S − ln S + ˆ ˆ ˆ ln2 S ˆ 2 S S2 ln S ˆ 2 1 ln4 S 1 ˆ ˆ + + . Here the “conformal spin” is s = 1 (S + E) = S + 2 J + 1 γ .7) can be written as 1 ˜ f(C) = 2π i dη γ (η) ˜ Γ C √ λ 1 ≈ S + 2 J . Notice that in contrast to the slow long string case where the expansion (2. this parity invariance property was already demonstrated 2. The discussion will apply to both the “slow” 1 and the “fast” long string limits. π2 J (2. S ≪ 13 . The leading terms here can be summed up as [3] E −S = where ln S J ˆ 1 2 ˆ ˆ ˆ 4 ln S + 4 − 2 ln S + 1 + 3 + 2 + . Some details are given in Appendix D.3) dominate in the limit when ln S . the corrections to the energy read E −S −J ≈ ˆ ˆ where S ≡ 8S = 8S ≫ 1...3 Large spin expansion of energy of a spiky string in AdS5 Let us now consider the spiky spinning string in AdS5 [34]. s(η) − C ˜ s(η) = S(η) + 1 γ (η) . 1 2J π 1 + 4 3 π J 21 In the case of “fast long string”. + .. one may again make use of the integral representation for the functional relation as in (2. In the kinematic region of “fast” in a closely related way in [15]. when ln S ≪ J ≪ S..¯ where S ≡ 8πS.16) negative powers of C..13) has the same suppressed by S.20) is C = integral in (2. corrections in J J ˆ ˆ 1/ ln S can be added in the round brackets and terms like ln(ln S) have been neglected. with 1 ≪ ln S ≪ J ≪ S. 2 ln3 S − 19 ln2 S + 11 ln S + 13 + ˆ ln2 S ˆ ˆ2 ln S S J2 + 1 2 8S ln + . Dots in the square brackets indicate corrections in 1/S.. and dots stand for higher order corrections depending on J . ˜ 2˜ (2.. and find corrections to the leading ln S term in its large spin expansion.6). + .

. .19) by the number m of the folds. m 23 For n = 2 we have ∆θ = π (i. with ρ0 and ρ1 as its minimal and maximal values (positions of the bottom of the valley between the spikes and the spikes themselves). 2 14 . one is simply to substitute n → n m. to the case when the ends of the spikes y = e−2ρ1 . ∆θ = csc π and n E −S =− i. . and thus ρ0 = 0 or u0 = 1. E −S = n = 2.. p) Π( .e. 4π y (2.23) This is the result already found in [34]. which reduces to the case of the folded string when Expanding further near y ≃ 0 one gets S = ∆θ = 22 u0 + 1 n 1 u0 + .17) S=√ n cosh ρ1 √ 2 π u1 + u0 u2 − 1 π u1 − u0 0 Π( . Then.. ω = coth ρ1 . p) − Π( . and expand in y → 0. n 2 u1 − 1 2 u1 + 1 2 sinh ρ1 u1 + u0 − (1 + u0 )K(p) + (u1 + u0 )E(p) − .18) √ n u1 + u0 √ E − ωS = [K(p) − E(p)] . p= u1 − u0 . and use that in this case ∆θ = nπ . 2π n (2.e.20) The string is rigidly rotating with the radial coordinate being ρ = ρ(σ). 23 π n (2.The integrals of motion here are the difference between the position of the spike and of the middle of the valley between the two spikes.22) n 16 π S ln + . . + ln y + 1 − 2 u2 − 1 arccos + ln 0 4π y 2 u0 4 1 u0 + 1 π = arcsin + y 2 arcsin − π + . ρ0 and ρ1 are related by the condition (2. .25) In the case of the multiply folded string with n spikes multiplying formulas one should multiply (2. u1 = cosh 2ρ1 . . i. n u0 2 u0 (2. the spin and the energy [34] ∆θ = 22 π sinh 2ρ0 π u1 − u0 1 π u1 − u0 √ =√ . u1 + 1 2 u1 + 1 (2. n).. Solving for the remaining free parameter gives E = E(S. the angle between spikes is π). (2.21) 1 = arcsin u0 +O(y) implies u0 = cosh 2ρ0 = n ln y + O(y) . p) .19) (2.24) (2..17). u1 + u0 (2. 2 π sinh ρ1 where n is the number of the spikes and u0 = cosh 2ρ0 .. at leading order. Let us set The large spin limit corresponds to ρ1 → ∞. As a result.e. approach the boundary of AdS5 . 2π S= n 1 + O(ln y) .

+ ..26) only the dominant contributions at each order of the above expansion we obtain E −S = n n2 n3 n4 ln S + ln S − ln2 S + ln3 S + .5) has the following expansion ˜ f(S) = where n 1 1 q2 ¯ ¯ ¯ ln S + q1 + ¯ + ¯2 (q3 ln S + q4 ) + ¯3 (q5 ln S + q6 )...3) for the folded string in AdS5 when n = 2. 15 . the analog of the function ˜ f(S) in (2.. p6 = 33 + O(n − 2) . Eliminating then y in favor of S. the reciprocity property is not respected in this case. q2 = cot . 2π 8 π2 S 64 π 3 S 2 384 π 4 S 3 √ √ 1 λn λn E−S = ln S + ln S + . + ln csc n n n n n n = −18 + O(n − 2) .30) It is easy to check that (2. we have from (2.. q3 = 4 csc2 .32) implying that the functional relation is satisfied (cf.29) (2.27) (2.31) This may be rewritten as (2.19) E −S = n 4 4 ¯ ¯ ¯ ¯ ln S + p1 + ¯ ln S + p2 − ¯2 2 ln2 S + p3 ln S + p4 2π S S 32 ¯ ¯ ¯ + ¯3 2 ln3 S + p5 ln2 S + p6 ln S + p7 + . (2.. . .35) (2. n n n n π 2 π 2 π 2 n−2 2 . Indeed. 2π S S S (2.26) where p1 p3 p4 p5 ¯ 16 π S S= n π π π(n − 2) π = −1 + ln sin .4)). n n n n n π π 2 (n − 2)2 π π π 2 (n − 2)2 4π(n − 2) − cot + cot2 +1 = 6 − csc2 + n 2n2 n n n n2 π 2π(n − 2) π π π π 2 cot2 − cot − csc2 + 2 ln csc + 10 . (2.28) (2. . (2.. p7 = −14 + O(n − 2) .33) q1 = −1 + ln sin q4 q5 24 π π 2π(n − 2) π .26) coincides with the energy (2. 2π 2 2π (2. .24 Retaining in (2..where the second equation can be used to fix u0 in terms of y and the number of spikes n.. (1 − 2 csc ) + 4 ln sin csc = 4 + 2π n n n n = O(n − 2) . n n 2n n 2π(n − 2) π π π π = −10 + cot − 2 cot2 − 4 ln csc + csc2 . p2 = −1 + ln sin + cot .36) Note also that the form of p1 is consistent with the interpretation of the subleading term in the energy in [27]. However. 3S (2. q6 = O(n − 2) ..34) (2.

The fluctuation action in the conformal gauge expanded to quadratic order in fluctuations 25 This feature of the ˜ f-function is in a marked contrast with the anomalous dimension.3) remains the same also with the 1-loop corrections included. to the energy (2.32) is still satisfied.where q5 . 16 . a generalization to non-zero J was considered in [36.33) coincides with (2. while the reciprocity was checked at weak coupling only for the minimal anomalous dimensions. We shall follow the general approach for computation of quantum string corrections developed in [10] where the 1-loop shift of the ln S term was found. even if considerably simpler compared powers of ln S appear to cancel in the subleading terms in (2. and (ii) the constraints on the coefficients imposed by the functional relation and the reciprocity remain to be satisfied at the 1-loop order.26). 18. 4].27)) √ is proportional. 39]. This reduction of singularity of the large S expansion of ˜ was observed also at weak coupling [26. to 3.3) in the large S expansion. for n = 3. 3 Large spin expansion of folded string energy: 1-loop order Let us now go back to the folded spinning string case of section 2. (2.33). anomalous dimensions of operators of twist higher than two with trajectories close to the upper boundary of the band also do not respect the reciprocity as was seen recently in the twist three case at weak coupling in [4].5). Korchemsky for this observation. It is interesting that our strong-coupling result (2. whose large S expansion includes growing powers of ln S in the coefficients of 1/S n terms.33) has close similary with weakcoupling one found for n = 3 in [4]: the functional relation (2. The expansion (2. where indeed (2.25 The parity invariance is restored in the case of the folded string when n = 2. It is interesting though that higher This breakdown of parity invariance for a string with n > 2 spikes is not totally surprising. this coefficient should be a function of λ interpolating from weak to strong coupling but its dependence on n might be the same for any λ. 40] (parallel results from the Bethe ansatz were found in [41.33).3) by applying a perturbative procedure similar to the one used in [9] in the small spin expansion case.26 In general.33) (cf. the same factor appearing also in the corresponding expression at weak coupling [4]. suprisingly.(2. q6 are non-zero for n = 2.27 We shall find the 1-loop corrections to the subleading terms in (2.32). and the parity invariance is broken at level 1/S.1 and compute the leading 1-loop corrections to its energy (2. which is. Interestingly. The 2-loop correction to the scaling function was found in [38. Our aim will be to verify that: (i) the structure of the large spin expansion (2. the 1/S coefficient ∼ nq2 in (2. is not parity invariant under S → −S. 42]). Indeed. as such spiky string should correspond to an operator with non-minimal anomalous dimension for a given spin. f 26 27 We thank G.

17 . We shall assume this when writing the fermionic contribution below. κT T ≡ dτ → ∞ . φ µ2 = 2ρ′2 − w2 − κ2 . we may compute the relevant 2d functional determinants by reducing them to 1d functional determinants using 2 2 det[−∂1 − ∂0 + m2 ] = T dω 2 det[−∂1 + ω 2 + m2 ] . (3.3) and can be interpreted as describing a system of 4+4 2d Majorana fermions with σ-dependent mass ρ′ . 4) are fluctuations in S 5 .2) Here βu (u = 1. ρ µ2 = 2ρ′2 . (3. ζs (s = 1.4) half of fermions has +ρ′′ and half −ρ′′ in their E1 = (3. thus.λ ¯ near the folded spinning string solution S = − 4π √ dτ 2π 0 ¯ dσ L has the following bosonic part (see [10] and Appendix B) ˜ ˜ ˜ ¯ ˜ ˜ ˜ LB = − ∂a t∂ a t − µ2 t2 + ∂a φ∂ a φ + µ2 φ2 t φ 2 + ∂a βu ∂ a βu + µ2 βu + ∂a ϕ∂ a ϕ + ∂a ζs ∂ a ζs . after squaring the corresponding “Dirac” operator.12).5) Since the spinning string solution is stationary. t µ2 = 2ρ′2 − w2 . (B. The fermionic part of the quadratic fluctuation Lagrangian can be put into the form [10] ˜ ¯ ¯ LF = 2i(Ψγ a ∂a Ψ − µF ΨΓ234 Ψ) . as in [9]. µ F = ρ′ . β (3. the fermionic contribution to the 1-loop partition function can be represented as − 1 4 ln det∆F + + 4 ln det∆F − . from the 2d effective action Switching to euclidean signature (τ → iτ ). 3. ρ′′ = O(η) and since according to (3. while ϕ. 2. 2 ∆F ± ≡ −∂ a ∂a ± ρ′′ + ρ′2 . the 1-loop correction to the energy can be found Γ1 .4) In the leading-order computation in the long-string limit that we are going to discuss below the term ±ρ′′ in the effective fermionic mass squared term in (3. the leading O(η) contribution to the partition function can come only from the ρ′2 term in ∆F ± .6) where m2 is a generic mass term which may depend on σ.4) can be ignored: as follows from mass term. As explained in [9]. both the bosonic and the fermionic fluctuation Lagrangians do not depend on τ . β ¯ ¯ + 4˜(κ sinh ρ ∂0 t − w cosh ρ ∂0 φ) + ∂a ρ∂ a ρ + µ2 ρ2 ρ ˜ ˜ ρ˜ (3. 2) are the two AdS5 fluctuations transverse to the AdS3 subspace in which the string is moving.1) where µ2 = 2ρ′2 − κ2 . 2π (3.

i. 18 . Some relations needed below can be found in Appendix B.. ˜ ˜ ˜ the bosonic fluctuation Lagrangian becomes LB = L0 + η L1 + . and.5) we have from (B.Given that ρ(σ) is a complicated function (see (B.6) κ = κ0 − η (πκ0 − 2) + O(η 2 ) . π η (3. χ): ˜ ˜ ξ = −t sinh ρ + φ cosh ρ. we will resort to perturbation theory in η determining the maximal string length (see Appendix B and C). Then the fluctuation Lagrangian takes the form ¯ LB = −∂a χ∂ a χ + (µ2 sinh2 ρ − µ2 cosh2 ρ + ρ′2 )χ2 + ∂a ξ∂ a ξ + (µ2 cosh2 ρ − µ2 sinh2 ρ − ρ′2 )ξ 2 t t φ φ 2 +2˜χ(κ − w) sinh 2ρ + ∂a βu ∂ a βu + µ2 βu + ∂a ϕ∂ a ϕ + ∂a ζs ∂ a ζs ρ˙ β ˜ ˜ χ = −φ sinh ρ + t cosh ρ . we are unable to determine the fluctuation spectrum exactly..8) +4¯(κ sinh2 ρ − w cosh2 ρ)∂0 ξ + ∂a ρ∂ a ρ + µ2 ρ2 + 2ρ′ (χξ ′ − ξχ′ ) + χξ(µ2 − µ2 ) sinh 2ρ ρ ¯ ¯ t ρ¯ φ (3.11) + (χξ ′ − ξχ′ )[ − π 2 π As already mentioned.9)) with the coefficients To proceed.6)). the 1-loop effective action can be expressed in terms of 1d functional determinants (with ∂0 → iω. ˜˙ ˜˙ (3. ˜ ˜ 0 (3. Let us first perform (as in [36]) ¯ ¯ the following rotation (t. (see (B. φ) → (ξ.4)). (3. will have constant coefficients as at 28 Expanding the solution for ρ(σ) and the parameters κ and w in small η (see Appendix B). see (3. as in [9].7) Γ1 will also be expected to have expansion in powers of η ∼ containg powers of ln η. 36].e. we shall ignore the contribution of the turning points at σ = and and will treat the fluctuation problem separately on each “quarter-string” interval.10) and κ0 2 ˜ L1 = −κ2 cosh(2κ0 σ)ξ 2 − κ2 cosh(2κ0 σ)˜2 − κ2 sinh(2κ0 σ) ξχ − [κ0 π cosh(2κ0 σ) − 2]βu ρ 0 0 0 π κ0 2 1 cosh(2κ0 σ)] − ρχκ0 sinh(2κ0 σ) − ρξ[ + κ0 cosh(2κ0 σ)] . 4π κ0 ≡ 1 S 1 S or in parameter 1 16 ln . where ˜ L0 = − ∂a χ∂ a χ + ∂a ξ∂ a ξ + 2κ0 χξ ′ − 2κ0 χ′ ξ − 4κ0 ρξ ˜˙ 2 + ∂a ρ∂ a ρ + ∂a βu ∂ a βu + 2κ2 βu + ∂a ϕ∂ a ϕ + ∂a ζs ∂ a ζs . We shall denote the quadratic fluctuation operator in 28 π 2 3π 2 As discussed below. we need to expand the fluctuation Lagrangian in small η corresponding to large S. Lagrangian at order O(η 0 ) will become σ-independent. In (3..9) The reason for this rotation is that in the subsequent small η expansion the bosonic fluctuation the leading order in long-string expansion considered in [10.

(1) (3.  (3. and we performed the Fourier transform in σ..18) for fluctuation fields which are 2π periodic in σ.19) As in [9] the first. .e. replaced ∂1 → in. Q23 = −ω[ + cosh(2κ0 σ)] 2 π 2 (3.12) Q(0) =  −2ωκ0  ..13) Also.The 1-loop correction to the effective action is then Γ1 = T 4π ∞ −∞ ˜ the coupled (χ.. i. n = 0.. where Qω is the next to leading order coupled operator from (3.11)  0 Q12 Q13  (0)  Pω =   (1)  2 −(−∂1 + ω 2 ) 0 2 −∂1 + ω 2 0 0 2 −∂1 0  ω2 0 (1) 0 +   . 2 π 2 κ0 ω 1 κ0 =− sinh(2κ0 σ).13) can be computed to order O(η) using that ln det[O(0) + ηO(1) ] = η Tr[(O(0) )−1 O(1) ] + O(η 2 ) . ρ) sector as Qω . 2 π 2 κ2 1 κ0 = − 0 sinh(2κ0 σ) − in[ − cosh(2κ0 σ)] . Qω =  Q21 −κ2 cosh(2κ0 σ) Q23 0   2 cosh(2κ σ) −Q13 −Q23 −κ0 0 Q12 = − Q21 Q13 1 κ0 κ2 0 sinh(2κ0 σ) + in[ − cosh(2κ0 σ)] .15) (3. −2κ0 ∂1 −∂1 + ω 2 ω   2 + ω2 0 2ωκ0 −∂1 dω − 8 ln 2 2 det[−∂1 + ω 2 + 2ρ′2 ] det[−∂1 + ω 2 + ρ′2 ] + 2 ln 2 2 det[−∂1 + ω 2 + κ2 ] det[−∂1 + ω 2 + 2κ2 ] 0 0 − ln where 2 det8 [−∂1 + ω 2 + κ2 ] detQω detPω 0 + ln − ln 2 2 + ω 2 + 2κ2 ] det6 [−∂ 2 + ω 2 ] (0) (0) det [−∂1 detQω detQω 0 1 .17) (3. second and fourth terms in (3. Since L0 has constant coefficients. ξ. det O(0) 19 (3. (3.20) .16) (3. ±1.. as appropriate Our aim will be to determine the 1-loop correction to string energy to order η by computing Γ1 = Γ1 + Γ1 + O(η 2 ) . Qω = Qω + ηQω + . the leading part of the ˜ ˜ fluctuation operator coming from L0 can be written as   2 2κ0 ∂1 0 −(−∂1 + ω 2 )   2 (3. (0) (1) Γ1 = O(η) .14) Here   (1)  ..

(3. here we shall assume that one can ignore the contributions from the turning points.11) of the solution ρ(σ) in small η. it was proven in general for any string solution in [43]. n2 + ω 2 + κ2 0 n=−∞ 29 The mass sum rule implies the 1-loop UV finitness of the superstring. π ).g.2). the order η contribution of the decoupled boson βu in (3.29 here this leading term is no longer zero as in the present case the expansion is around a nontrivial string background with different propagators for different string fluctuations.21) Considering the first interval (0. We shall split the integral over σ as follows 2π 0 dσ 2π → 1 2π π 2 π dσ + 0 π 2 dσ + π 3π 2 2π dσ + 3π 2 dσ . for the fermionic contribution (the first term in (3. In (3.11) used in (3. 30 Note that for the second and the fourth σ intervals where ρ decreases we need to use the minus sign in (B. the expansion must be periodic in 0 ≤ σ ≤ 2π.22) Similarly.While in [9] a similar contribution to the effective action happened to vanish since it was proportional to the sum of squares of fluctuation masses. n2 + ω 2 + 2κ2 0 n=−∞ ∞ 0 dσ [πκ0 cosh(2κ0 σ) − 2] 2π (3. As discussed in Appendices B and C. 20 .11) is defined for 0 ≤ σ ≤ π 30 2 ).11) we used the expansion (B. We shall parametrize our current lack of control of such terms by including the possible contribution with an arbitrary coefficient in the final result. As in the computation of the leading order in [10]. this expansion breaks down at the turning points where subleading terms are of the same order as the leading term.23) sinh(πκ0 ) − 2 . (B.1) 2 and (3.13)) we get 2 det[−∂1 + ω 2 + ρ′2 ] ln 2 det[−∂1 + ω 2 + κ2 ] 0 (1) ηκ0 = − 2π = − ηκ0 8π 1 2 + ω 2 + κ2 n 0 n=−∞ ∞ ∞ π 2 0 dσ [πκ0 cosh(2κ0 σ) − 2] 2π (3. Direct implementation of this may effectively bring back the turning-point contributions. (3. The closed string fluctuations by definition We shall assume that we can treat the problem “piece-wise” also at the fluctuation level..10). The classical folded string solution is built out of four parts making up the closed string (e.11) can be obtained as ln 2 det[−∂1 + ω 2 + 2ρ′2 ] 2 2 det[−∂1 + ω 2 + 2κ0 ] (1) ∞ π 2 = − = − ηκ0 π ηκ0 4π 1 n2 + ω 2 + 2κ2 0 n=−∞ sinh(πκ0 ) − 2 .

(n2 + ω 2 )2 (n2 + ω 2 + 4κ2 ) 0 n=−∞ (3.e. then using the residue theorem in the integral over ω we arrive back at the same sum as below.33 is UV finite but formally has This term should be an artifact of our computational procedure related to the problem with expansion in η in (B. see also the discussion in Appendix C).19) coming from (3. Collecting the above results we observe that the final expression for the order η term in the effective action Γ1 is UV finite.31 Explicitly. and this is what we will do below. such terms should resum away. Moreover. Then the coefficient of Cn 21 . ξ) (time-like and longitudinal) that appear in the coupled part of the fluctuation Lagrangian cancel. Let us mention that if we perform the sum over n first.34 We believe that in a more systematic treatment that consistently treats the turning point contributions such terms will be automatically absent (equivalently. in our present form of the expansion.22)-(3. the part that does not contain the sinh(πκ0 ) factor is IR finite. 31 32 Their flat-space contribution is cancelled against the conformal gauge ghost contribution. ǫ → +0.24) The contributions of the other three intervals of σ are the same.11) near the turning points (see Appendices B and C). That can be done by shifting the upper limit of the integration over σ in (3. (3.27) κ0 3n + − 2n 4κ0 4κ0 n2 + κ2 0 3n2 4κ0 n Cn n2 + 2κ2 0 n2 + 4κ2 0 The coefficient of the part proportional to sinh(πκ0 ) given by an IR singular contribution. n=1 34 To justify the expansion in (B. the non-trivial potentially IR divergent contributions of the two unphysical AdS5 massless modes (χ. − ǫ.26) (3. 33 The IR singular contribution goes away if one separates it before doing the integral over ω.For the coupled part one finds ln detQω (0) detQω (1) = η 0 π 2 dσ Tr[(Q(0) )−1 Q(1) ] ω ω 2π ∞ = ηκ0 π 2 (n2 + ω 2 )2 − n2 (n2 + ω 2 + κ0 ) sinh(πκ0 ) . (3. Then we get for P P 3 the large κ0 behaviour of n Cn : 2 ∞ Cn = κ0 (ln κ0 − 6 ln 2 − 2 + γE ) + O((κ0 )0 ). .20) as Γ1 = − where An = Cn = (1) (1) Tη 4π ∞ n=−∞ An + Cn sinh(πκ0 ) . i.11) we need to omit the turning point contribution. integrating first over ω we obtain32 the order η contribution to the 1-loop effective action (3.25) 8κ0 n2 + κ2 0 − 4κ0 n2 + + 2κ2 0 2κ0 − 4κ0 n2 − + 4κ2 0 . Insisting on omitting the turning point contributions means that we should drop this IR singular ∼ sinh(πκ0 ) term.24) to will become sinh[(π − 2ǫ)κ0 ] ∼ η 2 −1+ π ǫ π 2 and thus is subleading compared to the contribution of order O(η 0 ).

25). (3. of course.28) we can extract its large κ0 behaviour ∞ n=−∞ An = 12κ0 ln 2 + O(e−2πκ0 ) . 12 (3.31) can be easily computed as a product over integer n of a matrix determinant (after ∂1 → in).26) is minus the derivative over κ0 of the sum in (3. the formal sum (3. in agreement with the result of [10] and also.12) has no σ dependence.19) coming from the third and fifth terms in (3. 2 2 2 det2 (−∂1 + ω 2 + 2κ2 ) det5 (−∂1 + ω 2 ) det(−∂1 + ω 2 + 4κ2 ) 0 0 ∞ 2 n=−∞ ln(n (3. with ref. which explains why the e−2πκ0 exponential correction [44] is determined by the mass of the lightest mode – in the present case of the √ fermionic mode. its functional determinant 2 2 detQ(0) = −det2 (−∂1 + ω 2 ) det(−∂1 + ω 2 + 4κ2 ) ω 0 (3.26) using the Euler-MacLaurin ∞ k=1 f (n) = f (1) + f (∞) dn f (n) + + 2 B2k (2k−1) [f (∞) − f (2k−1) (1)] .33) This turns out to be a direct generalization of the leading-order result of [10] where κ should be replaced by κ0 in the fluctuation mass terms (but not in the overall 1 κ 1 κ factor due to t = κτ ). (3.34) This is.29) The contribution of this term in Γ1 in (3. the expression in [44] √ contained an extra (minus “zero mode”) term 3 − 2. term was omitted in [10] since there it was subleading in the infinite κ limit).5) is then (using (B.13) as Γ1 = − (0) T 4π ∞ −∞ dω ln 2 det8 (−∂1 + ω 2 + κ2 ) 0 . for the subleading term. π κ (3.[44] considered.13). As a result. [44].32) with κ0 → κ and with the n = 0 term omitted (this 22 . (2k)! (3.32) 2 Since ln det(−∂1 + ω 2 + κ2 ) = + ω 2 + κ2 ).9)) E1 = − (0) (1) (1) 3 ln 2 κ0 η. Note that the coefficient in the exponent of the leading corresponding coefficients are closely related.Computing the remaining formula ∞ n=1 ∞ 1 n An contribution in (3.30) Let us now include the O(η 0 ) contribution to Γ1 (3. Using again the Euler-MacLaurin formula to transform the sum into an integral we find35 E1 = (0) − 3 ln 2 κ2 − 0 5 + O(e−2πκ0 ) . 2 Since detPω = −det3 (−∂1 + ω 2 ) we may write the relevant contribution from (3. following [10]. 36 Ref.33). The exponential term in the square bracket has also a prefactor of κ0 .(3.25) to the energy (3.36 35 Note that the sum of (3. Since Qω in (3. doing the integral over ω we finally √ n2 + 4κ2 + 5 n2 − 8 0 obtain the 1-loop correction to the string energy to order O(η) as (0) E1 1 = 2κ n=−∞ ∞ 2 n2 + 2κ2 + 0 n2 + κ2 0 .

S S 2 (3. . π b10 = (2) b10 = − 1 c..9) to order O(η) we get η= which. π b(0) = − c 3 ln 2 . π b11 = − (0) 3 ln 2 .. π bc = b10 1 − 3 ln 2 ln 8π + c .41) (3.39) 2 π 2 (1) in (3.36) κ0 = κ0 1 = 1+ +....5) we need to keep κ to order O(η) to get the correction in 23 . . π 2π 2 S κ2 0 κ (3. plugged into (3. πS π2 S 2 ln(8πS) ln(8πS) − 3 + +. π κ (3.3) E1 = b0 ln S + bc + The contribution from E1 b0 = − while E1 (0) (0) Thus we find that the 1-loop correction to the folded string energy to order O( ln 2S ) can be S ln2 S b11 ln S + b10 + O( 2 ) .As we shall argue in Appendix C.40) Thus finally we obtain for the full 1-loop coefficients b0 = − b11 37 3 ln 2 . π 2π S ∼ √ 1 ln S S 2 ln(8πS) − 3 2 − + .35) where c is an undetermined constant (we included factor of π for convenience). 2π 2 (3.37) κ 2πS 1 ln S and This means that the dominant term in (3.38 written in the same form as the classical energy (2. Inverting the relation between S and η in (B..e. i.. .30) and E1 (1) (2) in (3. . π2 b10 = − (0) 3 ln 2 5 ln 8π − . π E1 = (2) c κ0 .7). as in (1. π 1 1 = 2 − 6 ln 2 ln 8π − +c . these terms are expected to cancel out in a more systematic treatment. π 3 ln 2 =− 2 .3). π2 b(2) = c 1 c.2)..(1.38) in (3.42) 1 E1 also to order η ∼ S .34) is the −3 ln 2 √ κ0 −2πκ0 κ e should be subleading and should be ignored in the approximation we 5 one: the other terms − 12κ ∼ considered above where we dropped terms of higher order in η at earlier stages. gives37 κ= 1 ln(8πS) − 2 +. 2π 2 (3. an additional contribution that may come from near turning point regions can be parametrized as follows: Γ1 = (2) c κ0 T . Note that in the expression for the energy in (3.34) to the 1-loop coefficients is 3 ln 2 ln 8π. (3. (3. 38 The role of these subleading terms and their possible resummation remains to be studied in more detail.35) contribute as 6 ln 2 .

McLoughlin. Rej. Arutyunov. λ λ λ bc b1k ¯ ¯ fc = ac + √ + ..18). Gromov. f1k = a1k + √ + . were participants of the GGI Florence workshop “Non-Perturbative Methods in Strongly Coupled Gauge Theories” and we thank the Galileo Galilei Institute for Theoretical Physics for the hospitality and the INFN for partial support. λ λ (3. the functional and the reciprocity relations appear to apply also including string 1-loop corrections.38) should obey (see (E. A.6). fc ≡ √ . Acknowledgments We are grateful to F. b10 = 1 (a0 bc + b0 ac ) .T. it would be interesting to generalize the 1-loop computation of this section and the check of reciprocity to the case of non-zero J and to attempt to relate the strong-coupling version of the reciprocity discussed in Appendix E to its weak-coupling finite twist one in (1. Needless to say... . Note that this is true for any value of teh undetermined coefficient c.46) we see that the relations (3. private communication) an independent way of evaluating the 1-loop correction to the folded string energy based on the algebraic curve approach to extracting 24 . i. 2 b0 ¯ f = a0 + √ + . and A.(1. R.7)) b11 = a0 b0 .43) (3.. N. S. Note Added As we have learned (N.T. π ac = 1 (ln 8π − 1) . Korchemsky.F. Kruczenski. Korchemsky and R. L. λ fc f f1k ¯ ¯ ¯ f ≡ √ . Dixon. was supported in part by NSF under grant PHY-0653357.45) are indeed satisfied by the expressions in (3. M.e.The functional and reciprocity relations in (1.Y.T. G.22) at strong coupling are (see discussion in Appendix E) 1 ¯ ¯ f11 = 2 f 2 .. Forini is supported by the SFB 647 ‘Space-Time-Matter’ grant and by the Alexander von Humboldt foundation.A. π (3. Part of this work was done while V.21). A. f1k ≡ .. A.45) Recalling the values of the leading coefficients at the classical level in (2. Theisen for useful discussions. Seminara and S. Alday. 2 (3. . D. The research of V. G. We thank G. Roiban. ¯ ¯¯ f10 = 1 f fc . Park for a collaboration on this topic back in 2005. T.A.(E.41). Basso. . thanks I. Frolov.44) They imply that the coefficients in (3. B. Roiban for very helpful comments on the draft. Gromov.3) a0 = 1 .

(3. the global AdS5 energy. 1) subgroup of SO(2.3 that creates the dilatation operator D = Σ54 generates an SO(1. (A. the coupling redefinition λ → λ + 3 ln 2 (suggested to us by G.9).6). The AdS5 energy E = Σ05 or conformal Hamiltonian generates an SO(2) subgroup while After the Their eigenvalues happen to be the same since the two representations (the unitary one classified by SO(4) × e.(1. At the same time. 1 Then the standard spin is S = Σ12 = M12 . 1. fc = √ λ 8π √ π [ln λ − 1] + d π 1 + O( √λ ). n = 0. on the embedding coordinates Y M can be related to the standard boundary conformal group Σm5 = 1 (Km + Pm ) . just as it did in the cusp anomaly coefficient in [37]. the energy E of a string state in global AdS5 space with boundary R × S 3 should it. Appendix A: Comments on conformal algebra realizations Starting with a conformal theory in R1.42). 2.g. 4). 5 with the signature − + + + +−) of SO(2.11).5)–(1. It is interesting to note that for c = 6 ln 2+d. 4) similarity transformation (see. 4) linearly realised generators as (see.(1. 25 . SO(2) and the one classified by SO(4) × SO(1. and (and analytic continuation) this state may be associated to a local operator in R1. Korchemsky) removes all ln 2 terms from the leading 1-loop coefficients in (1. Namely. L0 ≡ (D + M−+ ). [46]). 2. 1. the conformal spin is S ′ = Σ34 = 2 (K3 − P3 ).(3. then we get f= √ λ π 1 + O( √λ ). 3) one may define the collinear SL(2. [45]) Σmn = Mmn . in the R2.fluctuation frequencies [53] leads to the value c = 6 ln 2 + π for the undetermined coefficient in (1. i. R) subgroup as generated by the following light-cone components [30]: i i L− ≡ K− . N = 0. [L+ .4 embedding representation of the global AdS5 space the gener- If the eigenvalue of D is dimension ∆ and the eigenvalue of M−+ – the collinear projection of the ators ΣM N (M. Mmn . 1 Σm4 = 2 (Km − Pm ) .41).39 39 be equal to the energy of the corresponding SYM state on R × S 3 . 4. Through radial quantization In general. then the eigenvalue of L0 is the conformal spin s = 1 (∆+S).(3. e.11). Km .35).6) from 3 ln 2 to −3 ln 2− π. f10 = 8π λ √ 2π 2 [ln λ − 1] + √ d λ 2π 2 + O(λ0 ). This c contribution changes the 1-loop coefficient in (1. (A.3) the conformal energy is the rotation generator in the 05 plane.1) (A.g. 1 E = Σ05 = 2 (K0 + P0 ).e. D (m.2) Lorentz spin S.3 with the standard SO(2. 3. where d is a constant not √ √ involving ln 2. L− ] = −2L0 . 2 2 [L0 . 1)) are related by a global SO(2... L+ ≡ −iP+ . 2 Σ54 = D . The corresponding 2 quadratic Casimir operator is C 2 = s(s − 1). L± ] = ±L± . 4) conformal group generators Pm .

unless one is interested in the large spin expansion (see in this connection [47. S).4) √ 1 1 . replacing the operator tr(ΦDS Φ) with tr(ΦD∗ Φ). Appendix B: Review of folded string solution with J = 0 In this Appendix we review the folded spinning string solution in AdS3 [7. 2) or string states in AdS3 are naturally labeled by (E. − η η 40 The formal relation can be achieved by a continuation to euclid: by replacing null direction like x0 + x3 with S a complex one x1 + ix2 . 55]).2) w2 ≡1+η . sinh ρ = √ sn κ η σ. φ = wτ. 47]). the representations of SO(2. [49.3) with the initial condition ρ(0) = 0 is 0≤σ≤ π . The solution is given by t = κτ. κ2 (B. (B. 2) subgroup of SO(2.1) where ρ(σ) satisfies ρ′ = ±κ 1 − η sinh2 ρ . 2 (B.e.23) or E − S = f(E + S) is reminiscent of a light-cone gauge expression. Since different choices are formally related via SO(2.Euclidean continuation of the embedding coordinate Y0 → iY0E (to allow for the mapping from 1 E = Σ05 = 2 (P0 + K0 ). where D∗ = D1 + iD2 . That relation may possibly be made more explicit by choosing a different set of coordinates in global AdS5 in which the boundary is not R × S 3 but AdS3 × S 1 (see [48] where such coordinates in the boundary theory where used to explain the leading E ∼ ln S behaviour). ds2 = − cosh2 ρ dt2 + dρ2 + sinh2 ρ dφ2 . where f would be a light-cone Hamiltonian (cf. 2) which is a symmetry of global AdS3 subspace of AdS5 where the folded spinning string is moving to the collinear SL(2. 6] and consider its large spin expansion (see also [10]). ρ = ρ(σ) . + 26 . Here ρ varies from 0 to its maximal value ρ0 related to the parameter η by coth2 ρ0 = The solution in the interval 0 ≤ σ ≤ π 2 (B. i. 50. 2) labels 2 (E+S) does not appear to be natural. R) subgroup classifying the operators like tr(ΦDS Φ) one is also to perform an additional analytic continuation that + interchanges the euclidean (12) plane with the hyberbolic (+−) plane. Let us mention also that the relation (1. 4) transformations and a re-identification of the generators one may expect that the two representations should be equivalent. R × S 3 to R4 ) one may exchange Y0E with Y4 which exchanges the generator Σ54 = D with To relate the SO(1.40 Still. 1 and the relation to SO(1.

Indeed. 2.6) η 2 2 η η 2 2 η 2η η 2 2 η In this paper we are interested in the large spin or long string limit.7) (B. η η (B. Hence at the turning point the order we have sinh(2κ0 σ) = sinh(πκ0 ) ≈ Let us note that these expansions were found using pre-Mathematica 6 versions of Mathematica (Mathematica 6 apparently has some bug leading to inconsistent expansions for some elliptic and hypergeometric functions). κ0 → ∞ the string touches the boundary of AdS5 (ρ0 = ∞) π 2 ∼ η −1 and similarly cosh(κ0 σ) = sinh( π κ0 ) 2 ∼ η −1/2 . 4π 41 1 S we will only need expansions to κ0 ≡ 1 16 ln .4) in small η we obtain.The condition satisfied at the turning point ρ0 at σ = intervals to cover the full 0 ≤ σ ≤ 2π interval. 1. 8 (B. small η expansion. − ). − ) (B.10) and the solution can be approximated (away from the turning points) by simply ρ = κ0 σ. − . E. To construct the full 2 π 2 (2π periodic) folded closed string solution one should glue together four such functions on < σ < π we have π ≤σ≤π . S in small η we obtain κ = κ0 − E= S= η (πκ0 − 2) + O(η 2 ) .7).and 2-loop string corrections [10. the energy and the spin in terms of η are √ 1 1 1 3 1 1 1 1 1 1+η 1 1 κ = √ 2 F1 ( .e. Notice that it goes in powers of η with coefficients containing ln η and is not.g. − ).11) η sinh(2κσ) − 2κσ cosh(κσ) + O(η 2 ) . 39]. S = √ 2 F1 ( . At the next order in small η expansion the “ends” (turning points) of the string are close to the boundary but no longer touch it.10) or (B.11). sinh ρ = sinh(κ0 σ) − η 4 sinh(2κ0 σ) − π σ cosh(κ0 σ) + O(η 2 ) . for 0 < σ < π ..9) 2 πκ0 + 1 η + − (2πκ0 − 3) + O(η 2 ) . valid close enough to the turning points. 1.5) The expressions for the parameter κ. π η (B. 8 (B. 27 . i. E = √ 2 F1 (− . 38.8) (B. for π 2 π 2 is ρ′ ( π ) = 0. πη 2π 32π Expanding the solution (B. . Expanding κ. e. πη 2π 32π πκ0 − 3 η 2 − − (2πκ0 + 13) + O(η 2 ) . . We should add a word of caution about the use of the formal expansion in (B. This limiting case proved to be a useful framework for computing 1. . using (B. 36. 2 sinh ρ = sinh(κσ) − or. 2 1 1 √ sinh ρ = √ sn κ η (π − σ). Since in section 3 we compute 1-loop correction only to order order O(η). for σ = 1 πκ0 2e 41 To leading order when η → 0. strictly speaking.

12) (B. Thus the correction to the leading result of [10] should mainly come from the “internal” parts of the σ-interval.e. η 4κ0 σ sinh(κ0 σ) w cosh ρ = κ0 cosh(κ0 σ) + 8π −[πκ0 cosh(2κ0 σ) − 3πκ0 − 4] cosh(κ0 σ) + O(η 2 ) .11) to compute it. as will be discussed in Appendix C.g. i. where the expansion (B. π ηκ0 µ2 = κ2 − πκ0 cosh(2κ0 σ) + 1 πκ0 − 1 + O(η 2 ) . We shall comment more on this point in Appendix C.e. i. 2 In section 3 we ignored the regions near the turning points and thus uses the formal expansion (B.16) (B. 0 φ 2 π 2 2 2 µρ = −ηκ0 cosh(2κ0 σ) + O(η ) .11) goes actually as η −1/2 .14) Notice again that these expansions are formally invalid at the turning point (but are justified away from it). e.11) is justified. is of the same order as the leading term sinh(κ0 σ). What happens is that at σ = κ2 0 proportional to κ2 (see Appendix C). µ2 = κ2 − t 0 (B. are Appendix C: Resummation of “long string” expansion near the turning points Continuing the discussion of the previous Appendix B here we will show that it is possible to resum systematically the terms ∼ e2 n κ0 σ appearing in the formal η expansion of ρ(σ) in 28 . The resummation at the level of the string profile ρ(σ) is completely equivalent to its expansion near the turning point σ = π .η term in (B. A reason behind near the turning points where ρ′ = 0.15) (B.. Similar assumption was made in [10] in the computation of the 1-loop shift this assumption is that the masses of string fluctuations in (3.2) depend on ρ′2 which is small of the coefficient of the ln S term in the energy (where it was indeed justified).17) π 2 the leading term gets cancelled agianst the sum of subleading terms which all are of the same order. ρ′2 = κ2 − 0 (B. If σ is slightly away from the turning point the subleading terms are smaller than the leading term but then the expansion and the contributions to the energy need to be resummed. In section 3 we included a possible term which we may thus miss with an aribitrary coefficient yet to be deterined.12) where the leading term does not vanish at the turning point where one should have ρ′ = 0. 2π η κ sinh ρ = κ0 sinh(κ0 σ) + 4κ0 σ cosh(κ0 σ) 8π −[πκ0 cosh(2κ0 σ) + 3πκ0 − 4] sinh(κ0 σ) + O(η 2 ) . in (B. This is evident. For the computation of the 1-loop correction in section 3 we will need the following expansions η κ0 [πκ0 cosh(2κ0 σ) − 2] + O(η 2 ) . 0 The masses appearing in the fluctuation Lagrangian in section 3 are then expanded as follows ηκ0 1 πκ0 cosh(2κ0 σ) − 2 πκ0 − 1 + O(η 2 ) .13) (B.

4 (C. The next-to-leading (NLO) result which is correct at O(η) can be written rNLO (t) = ln t − + ηt2 η 2 t4 η 3 t6 η 4 t8 + − + + ··· 16 512 12288 262144 η 2 t4 η 3 t6 ln t η 3 t6 ηt2 η 2 t4 ηt2 − + + ··· + + − + ··· 1− 32 512 8192 2πκ0 8 128 2048 π 2 (C. The dangerous terms can be easily identified by setting t = eκ0 σ . . These terms are potentially dangerous since they scale at the turning point σ = π/2 as en π κ0 = (16/η)n and spoil the perturbative expansion.(B. Introducing h(t) ≡ r(t) − ln t we get 1 + t h′ (t) = 1−η t eh − t−1 e−h 2 2 . (C. h(t) = r(t) − ln t . (C.2) and neglecting exponentially suppressed terms in the above expansion.1) + + − 2 16π 256π 128π 192π 1 cosh (4κ0 σ) sinh (2κ0 σ) 125 sinh (2σκ0 ) 3 η + O η4 . the function ρ(σ) obeys the differential the η expansion. Its formal expansion in powers of η (treating κ0 in (B..10)) σ 1 − sinh (2κ0 σ) η 2π 8 cosh (2κ0 σ) σ 13σ 1 1 + − − + sinh (2κ0 σ) + sinh (4κ0 σ) η 2 8π 64π 16 256 23σ sinh (2κ0 σ) σ 2 29 cosh (2κ0 σ) σ cosh (4κ0 σ) σ + + (C. This can be accomplished as follows.4) Taking the large t limit we arrive at the following equation for the leading order term hLO 1 + t h′ (t) ≃ LO which can be integrated and gives hLO (t) = − ln 1 + As a check we can reexpand and find indeed hLO (t) = − η 3 t6 η 4 t8 η 5 t10 ηt2 η 2 t4 + − + − + O t11 16 512 12288 262144 5242880 29 (C.3) η + ..10). − cosh (2κ0 σ) sinh (2κ0 σ) − − 128 3072 3072 ρ(σ) = κ0 σ + η 1 Since here.7) η t2 16 . (B.5) .6) 1 − 1 ηt2 e2 hLO (t) .7) as a constant parameter) reads (cf. in fact. All terms (η t2 )k are O(1) at σ = κ κ0 and the above infinite series need resummation. equation (B.2) with plus sign and ρ(0) = 0. (C. κ0 = − π ln 16 the hyperbolic functions potentially reduce the true order in In the first “quarter-string” interval 0 ≤ σ ≤ π 2. r(t) ≡ ρ ln t κ0 .

The NLO approximation hNLO (t) is simply obtained by including an extra piece in the square root and taking into account that the ratio κ/κ0 has a non trivial expansion in η. In terms of the string profile ρ(σ) = r(eκ0 σ ) = h(eκ0 σ ) − κ0 σ this result can be written as ρNLO (σ) = κ0 σ − ln 1 + η 16 η 1−2σ/π 1−2σ/π 1−2σ/π 1 σ 16 + +η 2π 2 1 + η 16 1− 2σ π . 2 4 32 96 (C. of course. we immediately reproduce that ρ′ ( π ) = 0. and neglecting all NNLO terms we then get hNLO (t) = − ln 1 + η t2 16 +η 1 η 2 t2 ln t + 2π κ0 32 1 + η t2 16 1− 2 ln t π κ0 . 4 2 (C.9) in powers of η gives hNLO (t) = ln t t2 ln t t2 t2 t4 − − + η+ η2 2πκ0 16 512 16πκ0 32 t6 t4 ln t t6 ln t t4 t6 t8 + − + − − + η3 + 12288 256πκ0 512 262144 4096πκ0 8192 8 ln t 8 10 t t t + − η5 + · · · . + − 5242880 65536πκ0 131072 η4 (C.which are the leading terms in r(t). (C.10) which agrees indeed with the expansion of rNLO in (C. This is true since 2 π ρNLO ( ) = 2 ln 2 − 1 η ln η + + O η 2 2 4 This expression is not. as an additional check. This expression resums at NLO order the contributions near the boundary point σ = value of ρ( π ) with order O(η) included.13) Also.12) is in agreement with the exact value of ρ( π ) which is 2 1 π ρ( ) = arcsinh √ = 2 η ln 2 − η 3η 2 5η 3 1 ln η + − + + O η4 .8) NLO Integrating this equation. but it must reproduce the exact (C. (C.3).11) π 2.9) The expansion of (C. 1 + t h′ (t) = κ κ0 1−η t2 2 h η e NLO (t) + . substituting the necessary terms in the expansion of κ/κ0 . expected to be correct near σ = 0. NLO 2 30 .

(C. ˆ η (C.16) 1 − η sinh2 ρ(x) . Plugging the expansion of ρ around σ = find Qω = Qω. This is not. Expanding it consistently we get the final result ˆ ρ(x) = ˆ 1 1 sech x πκ0 + ln − x tanh x η + 2 2 4 4 1 1 5 13 + − x2 sech2 x + cosh(2x) + x tanh x − 32 128 64 128 π 2 η 2 + O(η 3 ) (C.In order to obtain the resummation in a systematic way and to show that it comes from the behaviour of the string profile around the turning point we can work out the expansion of the differential equation for ρ(σ) around σ = π . where  (0) Qω. 2 V1 =  −n2 − ω 2 −V1 −V2  (0) (1) π 2 in the Qω operator in (3.2)) ρ′ (x) = − ˆ perturbatively in η. − σ in this expression. To this aim.11). i.13) and denoting the resulting terms with label “fold” to indicate the expansion point. let us define 2 x=κ π −σ . 2 ρ(x) ≡ ρ ˆ π x .14) and solve the corresponding equation (cf.18) n2 + ω 2 − 2κ0 ω V2 2ωκ0 + V2 −2ωκ0 − V2 n2 + ω 2 − 2κ0 ω V2  .19) κ2 0 + 2inκ0 tanh (κ0 z) .11) plus a new η2 term which is beyond the order of accuracy of (C.fold + · · · .e.20) 31 . however.15) Here the value ρ(0) was left unexpanded. cosh (κ0 z) 2 V2 = ωκ0 . we get precisely the NLO One may wonder if this systematic resummation of ρ(σ) can be used to resum the associated contributions in the 1-loop correction to string energy discussed in section 3.fold + η Qω. (B. ˆ 1 ρ(0) = arcsinh √ .17) If we now use the definition of x = κ resummation in Eq. cosh2 (κ0 z) (C. − 2 κ (C.fold =  V1  V2 z≡ π −σ . we (C. immediately clear. 1 η ρ(x) = arcsinh √ + ln sech x − x tanh x ˆ η 4 + η2 −1 + cosh(2x) − 4 x2 sech2 x + 10 x tanh x + O(η 3 ) 128 (C.  (C.

In the large κ0 limit. this is precisely what is required by the reciprocity conditions in (3. the small η expansions for the “anomalous” 32 .  (C. In principle. δb10 = (C. (C. higher order terms coming from this term can be estimated to have the same order of magnitude and thus must be resummed.46). ω (0) (0) (0) ′ ∞ 0 dz δ+ (z) = 1) (C. In the “slow long strings” regime (S ≫ 1. Unfortunately.38) must obey δb0 = δb11 = 0 . J ≪ S). while the new piece is  0 2 i n ln 2 − κ0 −ω −2κ0 −ω   (0) ′ Qω.22) where Qω is the same operator (3. Appendix D: Details of large spin expansion for folded (S.45). we can make the following replacements ( κ0 sech2 (κ0 z) −→ δ+ (z) . a task which we (0) leave for the future. it is encouraging to note that a possible non-zero contribution from the near-turningpoint region is expected to change the one-loop energy by a term proportional to κ0 1 1 1 =1+ − + ··· . J) spinning string In this section we collect some details on large spin expansions used in Section 2.(3.23) This is a O(κ0 ) perturbation over Qω whose matrix elements are O(κ2 ) (since n. ω ∼ κ0 in the 0 combined sum and integral like in (3. the contribution from Qω.24) This means that the induced modification of the coefficients in appearing in (3.fold =  −2 i n ln 2 − κ0 −2κ0  ω ω (0)   .fold δ+ (z).fold must be treated exactly and separately.12) we found in section 3 in the expansion valid near σ = 0. Still.24)). After this substitution we can write Qω.fold = Q(0) + Qω. 2π (C. η + ··· = 1 + κ 4 2 π κ0 2π S 1 δbc .21) κ0 tanh(κ0 z) −→ κ0 − ln(2) δ+ (z) .25) Remarkably.2.

. ...part of the energy and the conformal spin read γJ ≪1 = κ + ˜ 42 sJ ≪1 ˜ κ 1 + ln η 4(ln η + 12) 2 S −S −J ≈ − + η + O(η 4 ) − J ω π π 20 44 10 (1 − ln η) + 2 − 3 − η2 + O(η 4 ) + . + 16 + ˆ ˆ ln C ln C S J (D. + O(1/C 4 ) − ln C + ˆ2 ˆ ln2 C ˆ π2 J 2 16 C ln C ˆ 1 1 ln4 C ˆ ˆ ˆ − 4 3 4 ln3 C + 5 ln2 C + 9 ln C + ˆ π J 8 32 C 2 34 24 ˆ + 2 + . + πJ2 2 ln η ln η ln η 2 ln η 1 2 ln η + 11 1 1 ˜ +η + O(η 3 ) = S + 2J + 2γ ≈ 8πη 2π 3 11 1 13 + O(η 3 ) + .4) = + 1 2 ˆ and dots inside round brackets indicate corrections in 1/ ln C..2.14) coincides with the anomalous dimension evaluated at zero of the denominator in (2.12) and (1. 2 ˜≈ f ˆ where C = already noted in section 2... for slow long strings ˜ ≈ f ln 8πC − 1 ln 8πC + 1 + + O(1/C 4 ) − J π 16 π 3 C 2 1 3 +π J 2 − + O(1/C 4 ) + . it ˜ ˜ has expansion in even negative powers of the semiclassical Casimir C. the expansion in the case of the fast long strings is not of the same type as in (1.. (D.. . with ln S ≪ J ≪ S) one finds γln S≪J ≪S ≈ ˜ + 1 π4J 3 − 1 1 ln η 1 + ln η + 44 η 2 π2 J 2 1 2 ln η − ln η − 1 + O(η 4 ) 11 sln S≪J ≪S ˜ + 1 ln4 η + 4 ln3 η + 2 η 2 (−5 ln4 η + 5 ln3 η + 33 ln2 η) + O(η 4 ) + ... 2 C 2 ln 8πC 2 ln 8πC 32π (D..12) after the redefinition η → −1 + 16η + p 1 + 256 η 2 . .2) 1 ln η −η − π2J 2 16 η f √ λ Since the function ˜ = f in (1. .11) and (2.. 42 C J ˆ 1 ln2 C 3 7 1 ˆ ˆ ˆ 4 ln C + 3 + + + .1) For the “fast long strings” (S ≫ 1. + πJ2 2 − η 2 ln η − 2 − 16 η ln η 4 ln η 2 ln3 η (D.19) assumed in the main part of this paper.16) in both cases we get an equation expressing the parameter η in terms of only the odd powers of the Casimir C. As was The expansions are obtained from (2. follows that the function f Explicitly. even in S. From the power series expressions for γ .. 8 1 11 + 12 ln η + O(η 3 ) ≈J − +η 1− 8 η ln η 2 ln2 η 3 ln2 η 11 + 4 ln η − 2 4 + O(η 3 ) + . 33 .3) where C = S + 1 J and dots indicate corrections in J . the first few corrections read. + O(1/C 4 ) + .. .. .. For fast long strings.

one is then able to compute the coefficients fmm of in terms of the strong-coupling expansion coefficients in the scaling function f ..3) √ fmk ≡ ( λ)m+1 lnm S Sm Assuming the functional relation or (1. This implies the condition dρ (κ2 − J 2 ) cosh2 ρ − (ω 2 − J 2 ) sinh2 ρ √ . one should find non-trivial relations between strong-coupling expansion coefficients in (E.22) which here should be understood in terms of power series in 1 √ . b0 c0 ¯ f = a0 + √ + √ + .. .2) (E. The latter are known up to 2-loop order directly from the string-theory computations [10. λ As a result. (D. 34 .5) 2π = 0 dσ = 4 m which leads to a factor of m in front of the relevant expressions for E. π m. arbitrarily high) order from the analytic strong coupling solution [37] of the BES [1] equation for the function f .17).4) and also to a high (in principle... As we have seen. the large spin expansion of anomalous dimensions at strong coupling appears to have the same structure as at weak coupling (1.. fmk . (E.1) (E.2) and (E.19) should lead to additional constraints on the subleading coefficients like (1.(1. parameter m enters only in combination with the string tension √ λ 2π . Assuming the validity of the reciprocity condition (1. S. . Once E is expressed in terms of S and J .. The large Appendix E: Higher order relations from reciprocity at strong coupling The evidence for the functional relation and reciprocity (1.into 4m segments: for 0 < σ < then decreases to zero for 2π π 2m Let us mention also that in the case of the m-folded string the interval 0 ≤ σ < 2π is split π 2m the function ρ(σ) increases reaching its maximal value ρ0 . π 1 c0 = − K . π b0 = − 1 3 ln 2 . ≤σ≤ ρ0 0 etc... λ ( λ)2 cmk bmk ¯ ¯ + . √ ¯ λ fc .3).19) at weak coupling suggests that the corresponding constraints should hold also in strong-coupling expansion. the κ2 − J 2 .. π . .16) where now f≡ fc ≡ √ ¯ λf . λ ( λ)2 bc cc ¯ fc = ac + √ + √ + . This means that fmm are then effectively determined if the functional relation applies. spin expansion is then similar to the m = 1 case. 39] a0 = 1 . fmk = amk + √ + √ λ ( λ)2 (E.14).

1 3 b11 = a0 b0 . J) string and then resum the series for its energy (both in J and in λ) so that the limit of finite J would make sense. 43 In the case of twist 3 operators. i. 3. see footnote 9 in the Introduction) are fixed accordingly. and ignore (E.3) we then find that some of the 1-loop coefficients can be expressed in terms of the tree-level coefficients and the coefficients in f .e.21) once J (and the flavor index ℓ.. the anomalous dimensions we will consider here are the minimal in the band. 43 They are derived from the closed expressions in terms of the harmonic sums that were obtained (mainly exploiting the maximum transcendentality principle and the asymptotic Bethe ansatz). etc..19) as C = λ C. The expansions are indeed of the generic form (1. All these expansions were proven to satisfy the reciprocity property.22). respectively.21) by noting that at strong coupling f ∼ fc ∼ λ and thus terms of order 1 or J ≪ sqrtλ can be ignored. b22 = − a2 b0 2 8 0 1 3 1 3 b33 = a0 b0 . Multiplying the series in (E.5) √ Equivalently. up to four loops in the gauge coupling and up to 1/S 3 order. however.7) Appendix F: Large S expansions for twist 2 and twist 3 anomalous dimensions at weak coupling Here we shall collect the coefficients of large spin expansion of anomalous dimensions for planar SYM operators of twist 2 and 3. f10 = f fc . in [51.19) at subleading order in strong √ coupling at J = 0 one may simply take the Casimir C in (1. . . b32 = a3 (2b0 − bc ) − a21 b0 − a2 ac b0 − a0 b21 . 35 . It is usually assumed that the folded string in AdS5 with zero angular momentum in S 5 describes an operator of small twist. in [17] (at four loops) for the twist two scalar sector. but that can be J = 2 or J = 3. For example. can not be distinguished from the formal case of J = 0. these relations follow from (1. 0 6 8 8 0 We have verified the validity of these relations for b11 and b10 in section 3.There is. Explicitly.21) or (1. a subtlety in formulating the reciprocity condition in the context of large spin expansion at strong coupling as defined by string semiclassical perturbation theory where all non-zero charges are automatically large at large λ. getting 1 ¯ ¯ 1 ¯¯ ¯ ¯ ¯ ¯ ¯ f32 = f (f 3 − 2f 2 fc − 16f21 ) . the shifts in brackets in (1.16) where the coefficients satisfy the relations (1. b10 = (a0 bc + ac b0 ) . 17] for the twist three scalar sector and in [33] (at three loops) and [18] (at four loops) for the “gauge” sector. 2 16 (E. Here we shall assume that in checking the reciprocity (1...6) (E. C = S. .. for a review see [18]. the case of finite twist J = 2.. .1)–(E. To establish a relation to the definition of reciprocity in weakly coupled gauge theory expansion with finite twist one would need to consider the case √ of semiclassical (S.

1) ˆ ¯ where S = eγE S and the coefficients will be power series in λ = 44 λ .44 As far as these leading terms are concerned. In [52] it was proved that the anomalous dimension ϕ their anomalous dimension is expressed in terms of a universal function with shifted arguments γJ =2 (S) = for twist three operators built out of gauginos is related to the one of the twist two universal supermultiplet as ψ ϕ γJ =3 (S) = γJ =2 (S + 2). γJ =2 (S) = γuniv (S + 1) .16) as 2 ¯ ¯ ¯ ¯ ¯ ¯ ¯ ¯ f11 ln S + f10 + f22 ln S + f21 ln S + f20 + γ(S)S≫1 = f ln S + fc + 2 S S 3 ¯ ¯ ln2 S + f31 ln S + f30 ¯ ¯ ¯ ¯ ¯ f33 ln S + f32 ln4 S + . there is no need to explicitly write down the results for the twist two gaugino and gauge sectors. the closed formulas for their anomalous dimensions can be deduced from the one for the twist two scalar case by just shifting the argument of the harmonic sums45 but such shifts do not affect the coefficients of the leading lnm S/S m terms.The coefficients of the leading lnm S/S m terms below are manifestly universal in twist and flavor. γJ =2 (S) = γuniv (S + 2). and ψ A γuniv (S) . is a nontrivial consequence of the functional relation (1. However. At weak coupling it is useful to rewrite (1. +O S3 S4 (F. a well-known feature of the leading ln S coefficient (or cusp anomaly). 16π 2 Then one finds: The only exception being the four loop coefficient of the term ln2 S/S 2 in the case of twist two scalar operators. 36 . Indeed. as was noticed in [33] and emphasized in [4].14). it seems reasonable to relate this exception to the wrapping-induced breakdown of the Bethe equations at four loops for twist two operators. and the twist three gaugino sector. It is worth stressing again that this universality. 45 It is well known that in N = 4 SYM all twist two operators belong to the same supermultiplet.

315 64 π 2 ˆ 3 96 π 4 ˆ 4 ˆ λ + λ . ˆ ˆ ˆ f = 8λ − 3 3 45 315 16 56 80 ¯ ˆ ˆ ˆ fc = −24ζ3 λ2 + ( π 2 ζ3 + 160ζ5 )λ3 + (− π 4 ζ3 − π 2 ζ5 − 1400ζ7 )λ4 3 15 3 64 π 2 ˆ 3 96 π 4 ˆ 4 ˆ f11 = 32λ2 − λ + λ . 3 15 32π 2 22π 4 2π 2 ˆ 2 2ˆ ˆ ¯ )λ − ( + − 48ζ3 )λ3 f20 = − λ + (24 + 3 9 3 135 2 136π 4 146π 6 32 16ζ3 ˆ +( + − 384ζ3 − π 2 ζ3 + − 320ζ5 )λ4 15 945 3 3 512 ˆ 4 λ . (F. 3 5 6 4 2 ˆ ˆ ˆ ¯ ˆ 4π λ2 + ( 44π − 96ζ3 )λ3 + (− 292π + 160 π 2 ζ3 − 32ζ 2 + 640ζ5 )λ4 . 64π 2 96π 4 ¯ ˆ ˆ ˆ f21 = −32λ2 + (128 + + 128 ln 2)λ3 + (−256 − 128π 2 − − 128π 2 ln 2 + 256ζ3 )λ4 . f11 = 32λ2 − 3 5 88π 4 64 8π 2 ˆ ˆ ¯ ˆ − 32 ln 2)λ2 + ( + π 2 ln 2 − 32ζ3 )λ3 f10 = 8λ + (− 3 45 3 8 2 ˆ − (73π 6 + 756π 4 ln 2 − 840π 2 ζ3 + 2520ζ3 + 1260ζ5 )λ4 . 3 5 37 .2) 2 4 16 π ˆ 3 32 π ˆ ¯ ˆ f21 = −16λ2 + (128 + )λ + (−128π 2 − + 448ζ3 )λ4 . f31 = λ2 + (−256 + 3 9 3 15 56 ˆ 224π 2 32π 4 64 320ζ5 ˆ 4 40π 2 ¯ ˆ f30 = − λ2 + (96 + − 16ζ3 )λ3 − ( + − 800ζ3 + π 2 ζ3 − )λ 3 9 3 15 9 3 Twist three scalar sector: ˆ f = 8λ − 8π 2 ˆ 2 88π 4 ˆ 3 584π 6 2 ˆ + 64ζ3 )λ4 .Twist two scalar sector: 2 4 6 ˆ 8π λ2 + 88π λ3 − ( 584π + 64ζ 2 )λ4 . f10 = 4λ − 3 3 45 315 3 ˆ ˆ f22 = −64λ3 + (64π 2 − 128ζ3 )λ4 . 315 ˆ ˆ f22 = −64λ3 + 64π 2 λ4 . f32 = 64 λ3 + (−768 − 3 512π 2 64π 4 16π 2 ˆ 3 16 ˆ ˆ ¯ )λ + (512 + − − 576ζ3 )λ4 . f33 = 3 64 π 2 ˆ ¯ ˆ + 128ζ3 )λ4 . λ + λ −( 3 45 315 88 8 8 ˆ ˆ ¯ ˆ fc = −8 ln 2λ + ( π 2 ln 2 − 8ζ3 )λ2 + (− π 4 ln 2 + π 2 ζ3 − 8ζ5 )λ3 3 45 3 8 2 ˆ + (73π 6 ln 2 − 84π 4 ζ3 + 2520 ln 2ζ3 + 105π 2 ζ5 + 17325ζ7 )λ4 .

(F.3) 3 ¯ ˆ ˆ f32 = 128λ3 − (768 + 128π 2 + 512 ln 2)λ4 . 3 9 3 Twist three “gauge” sector: 4 6 2 ˆ ˆ ˆ ˆ 8π λ2 + 88π λ3 − ( 584π + 64ζ 2 )λ4 . f = 8λ − 3 3 45 315 8 ¯ ˆ 8 ˆ fc = 8(1 − ln 2)λ + (−12 − π 2 + π 2 ln 2 − 3ζ3 )λ2 − (−1440 − 60π 2 − 11π 4 3 45 8 4 2 3 ˆ +11π ln 2 − 15π ζ3 + 45ζ5 )λ + (−100800 − 3360π 2 − 336π 4 315 2 2 ˆ −73π 6 + 73π 6 ln 2 − 84π 4 ζ3 − 2520ζ3 + 2520 ln 2ζ3 + 105π 2 ζ5 + 17325ζ7 )λ4 64 π 2 ˆ 3 96 π 4 ˆ 4 ˆ f11 = 32λ2 − λ + λ . +(768 + 3 5 224 64 ln 2 ˆ 2 352π 2 128 2 64ζ3 ˆ 3 ¯ f30 = −( + )λ + (128 + + 512 ln 2 + π ln 2 + 128 ln2 2 − )λ 3 3 9 9 3 512π 4 1408 2 64 448π 2 − − 768 ln 2 − π ln 2 − π 4 ln 2 − 768 ln2 2 +(896 − 3 15 3 5 512 ln3 2 128 2 64ζ5 ˆ 4 −128π 2 ln2 2 − + 640ζ3 + π ζ3 + 512 ln 2ζ3 − )λ . 256π 2 384π 4 ¯ ˆ ˆ ˆ f21 = −128λ2 + ( + 128 ln 2)λ3 + (256 − − 128π 2 ln 2)λ4 3 5 16π 2 440π 4 256 2 200π 2 200 ˆ ˆ ¯ + 128 ln 2)λ2 + (480 − − − π ln 2 λ + (16 + f20 = − 3 9 3 27 3 1120π 2 64π 4 2920π 6 ˆ −64 ln2 2 + 128ζ3 )λ3 + (−2816 − + + − 256 ln 2 3 15 189 2 256 2 1600ζ3 384 4 ˆ π ln 2 + 64π 2 ln2 2 − 128ζ3 − π ζ3 + + 128ζ5 )λ4 + 5 3 3 38 . 3 5 32 32π 2 ˆ ˆ ¯ ˆ − 32 ln 2)λ2 + (−180 − 30π 2 + 11π 4 + 30π 2 ln 2 − 45ζ3 )λ3 f10 = 32λ + (32 − 3 45 32 2 ˆ − (−10080 − 840π 2 − 189π 4 + 73π 6 + 189π 4 ln 2 − 210π 2 ζ3 + 2520ζ3 + 315ζ5 )λ4 315 ˆ ˆ f22 = −64λ3 + 64π 2 λ4 . +64π 2 ln2 2 − 128ζ3 − π ζ3 − 256 ln 2ζ3 + 3 3 512 ˆ 4 f33 = λ . 64 ˆ 128π 2 ˆ ¯ − 256 ln 2)λ3 f31 = λ2 + (−512 − 3 9 64π 4 1408π 2 ˆ + + 1536 ln 2 + 256π 2 ln 2 + 512 ln2 2 − 512ζ3 )λ4 .8λ 8π 2 ¯ ˆ f20 = − + (48 + + 32 ln 2)λ2 3 9 64 80π 2 88π 4 ˆ − − 128 ln 2 − π 2 ln 2 − 64 ln2 2 + 32ζ3 )λ3 +(32 − 3 135 3 32π 2 352π 4 584π 6 96 +(−512 − + + + 256 ln 2 + 128π 2 ln 2 + π 4 ln 2 3 15 945 5 2 64ζ3 64 2 ˆ + 32ζ5 )λ4 .

4) 3 ¯ ˆ ˆ f32 = 512λ3 + (−256 − 512π 2 − 512 ln 2)λ4 1600 ˆ 2 3200π 2 ¯ ˆ f31 = − 1024 ln 2)λ3 λ + (−640 − 3 9 1792π 2 ˆ + 320π 4 + 512 ln 2 + 1024π 2 ln 2 + 512 ln2 2)λ4 +(−1792 + 3 1120 1600 ln 2 ˆ 2 5824 1856π 2 704π 4 ¯ ˆ f30 = 192λ + (− − 64π 2 − )λ + (− + + + 640 ln 2 3 3 3 9 15 1600ζ3 ˆ 3 25984 15488π 2 544π 4 4672π 6 3200 2 π ln 2 + 512 ln2 2 − )λ + ( + − − + 9 3 3 9 3 105 1792 2 512 ln3 2 +1792 ln 2 − π ln 2 − 320π 4 ln 2 − 256 ln2 2 − 512π 2 ln2 2 − + 1152ζ3 3 3 3200 2 1600ζ5 ˆ 4 2 + π ζ3 − 1536ζ3 − )λ 9 3 f33 = 39 . (F.512 ˆ 4 λ .

L. A. P. “Quasiclassical QCD pomeron. V.” JHEP 0206. M. “Three-loop universal anomalous dimension of the Wilson operators in N = 4 SUSY Yang-Mills model. Dokshitzer and G. S. B 646. Saclay. “Quantum corrections to energy of short spinning string in AdS5. [4] A. 504 (2006) [arXiv:hep-ph/0511302]. N. Rej. Frolov and A. Vogt. [16] Yu.4758 [hep-th]. “Transcendentality and crossing. Tseytlin. Moch and J. Lett. Vogt. Nucl. Logarithmic scaling in gauge / string correspondence. Freyhult. Phys. Korchemsky and R. B 462. A. Lett. Staudacher and V. Altarelli. “A Generalized Scaling Function for AdS/CFT. [15] B. Eden and M. “Fine structure of anomalous dimensions in N=4 super Yang-Mills theory. “Revisiting parton evolution and the large-x limit. P. G. Dokshitzer. V. 0701. P.References [1] N. Mech.” arXiv:0806. P021 (2007) [arXiv:hep-th/0610251]. D 54. Vermaseren and A. Korchemsky. [12] S. G.” Phys. S. N. Velizhanin. 007 (2002) [arXiv:hep-th/0204226]. Vermaseren.” Nucl. “The three-loop splitting functions in QCD: The non-singlet case”. [6] S.” Phys. Basso and G. Staudacher. J.” Phys. L. “Anomalous dimensions of high-spin operators beyond the leading order. [2] L. B 688. P.” Phys. N. B 634. Korchemsky.” arXiv:0712. Tseytlin. B 632. 1 (2007) [arXiv:hep-th/0612247]. Polyakov.2743 [hep-th]. 24 (2006) [arXiv:hep-th/0601112]. [17] A. P10003 (2007) [arXiv:0704. G.” arXiv:0806. Phys. Kotikov. Moch. “Semiclassical quantization of rotating superstring in AdS(5) x S(5). 101 (2004) [arXiv:hep-ph/0403192]. V.” J.3768 40 . [8] I. L. [18] M. V. Park and A. B 595. “Dressing and Wrapping. Tseytlin.3657 [hep-ph]. Lipatov.3586 [hep-th]]. Belitsky. Phys. Nucl. Stat. L. Phys. Belitsky. B 636. Pasechnik. Rej and M. 754 (2006)] [arXiv:hep-th/0404092].-Y. Beccaria and V. Staudacher. N. June (2007). Lett. S.” J. “A semi-classical limit of the gauge/string correspondence. Forini. [5] G. B 748. “Anomalous dimensions of high-spin operators beyond the leading order”. “Reciprocity of gauge operators in N=4 SYM. Beisert. Salam. A. 189 (2007) [arXiv:hep-th/0612248]. A. Korchemsky. “The three-loop splitting functions in QCD: The singlet case”. A. Mech.” Phys. Onishchenko and V. I. talk at the 12th Claude Itzykson Meeting. Phys. Klebanov and A. [11] A. 7513 (1996) [hep-th/9607056]. A. J. B. Korchemsky. I. B 775. “N = 4 SUSY Yang-Mills: Three loops made simple(r). Stat. Marchesini. Lipatov.A. [9] A. “Partons In Quantum Chromodynamics. P. [10] S. Rept. 129 (2004) [arXiv:hep-ph/0404111].” arXiv:0803. 521 (2004) [Erratum-ibid. [14] G. Egusquiza. A. M. Marchesini and G. S. B 691.” Nucl. Phys. Kotikov. M. L. Gorsky and G. Gubser. A. Tirziu and A. unpublished (2005). [7] H. A. “Planetoid String Solutions in 3 + 1 Axisymmetric Spacetimes. 1 (1982). [3] A. de Vega and I. 333 (1996) [arXiv:hep-th/9508025]. [13] Yu. Rev. 81. 0710.” Nucl. Velizhanin. M. 99 (2002) [hep-th/0204051]. Nucl. R.

4412 [hep-ph]. Frolov. B 553. Kruczenski. 194 (2007) [arXiv:0705.” JHEP 0310. “Gluon scattering amplitudes at strong coupling. Korchemsky and D.2639 [hepth]]. A. Manashov. CERN. B 766. Phys. G. 037 (2003) [hep-th/0308117]. “Asymptotics of the Altarelli-Parisi-Lipatov Evolution Kernels of Parton Distributions. “Large spin expansion of the long-range Baxter equation in the sl(2) sector of N=4 SYM. [31] E. “Precision spectroscopy of AdS/CFT. [26] M. 51. G. [32] N. Magnea and G.1889 [hep-th]. “Spiky strings and single trace operators in gauge theories. Stavenga. Korchemsky. “The uses of conformal symmetry in QCD.” Nucl. “Baryon distribution amplitudes in QCD. B 652. S. Dixon. [30] V.” arXiv:0807.1574 [hep-th]]. [23] M. Part. Beccaria. [35] N. [25] L. August 2008. 41 .” JHEP 0801. [28] V. Lett. Beisert and M.” Nucl.” JHEP 0212. Marchesini.” Nucl. M. Tirziu and A. R. [21] M. A. S. N. G. J.” Mod. B 566. 023 (2007) [arXiv:0707. [36] S.3515 [hep-ph]]. 1257 (1989). Kruczenski. F. [34] M. “A Spin Chain from String Theory.” JHEP 0706. 064 (2007) [arXiv:0705. Beisert. A. Phys. M. “Three loop anomalous dimensions of twist-3 gauge operators in N=4 SYM. Phys. “Evolution equations for quark gluon distributions in multi-color QCD and open spin chains. Roiban. Braun. Phys. Nucl. 355 (1999) [arXiv:hep-ph/9902375].” Nucl. Phys. 2|4) Bethe ansaetze for gauge theory and strings. [24] L. E. Manashov. Staudacher and A. Sterman. Dorey. Catino. 067 (2008) [arXiv:0710. P. Tseytlin. 024 (2002) [arXiv:hep-th/0210115]. B 727.” JHEP 0508.[19] G. L.” arXiv:0807. M. 93 (2008) [arXiv:0707. F. A 4. talk given at “Strings 2008”. 1 (2005) [arXiv:hep-th/0504190]. L.1991 [hep-th]]. Roiban. Derkachov. “Scattering Amplitudes. Tseytlin. Geneva.” Prog. P. Alday and R. Frolov . 022 (2008) [arXiv:0805.4387 [hep-th]. Yu. Magnea and G. Braun. Phys. “Logarithmic corrections to higher twist scaling at strong coupling from AdS/CFT. A. Staudacher. Alday and J. [22] L. A. 203 (2000) [arXiv:hep-ph/9909539]. Lett. [33] M. “Universal structure of subleading infrared poles in gauge theory amplitudes. Dokshitzer and G.” JHEP 0808. P. Tirziu and A. Mueller. [27] N. 014 (2005) [hep-th/0410226]. Beccaria. Kruczenski. Derkachov. B 791. Laenen.” Nucl. M.4254 [hep-th]]. “Strong-coupling expansion of cusp anomaly and gluon amplitudes from quantum open strings in AdS5 xS 5 . F. E. [29] M. Korchemsky and A. Korchemsky and A. “A note on twist two operators in N = 4 SYM and Wilson loops in Minkowski signature. P.” arXiv:0805. Phys.” JHEP 0709. 311 (2003) [arXiv:hep-ph/0306057]. “On next-to-eikonal corrections to threshold resummation for the Drell-Yan and DIS cross sections. Beccaria and F. Tseytlin. Maldacena. L. S. Alday. N. “Twist 3 of the sl(2) sector of N=4 SYM and reciprocity respecting evolution. “Long-range P SU (2. 232 (2007) [hep-th/0611269].” Phys. Wilson Loops and the String/Gauge Theory Correspondence. [20] L.0303 [hep-th]].

055 (2006) [arXiv:hep-th/0510208]. 014 (2002) [hepth/0205270]. Alday and J. “Generalized Scaling Function at Strong Coupling. Roiban and A.” Phys. Kristjansen.” Nucl.0672 [hep-th]]. Phys. 307.0681 [hep-th]]. [52] M. Zamaklar. Basso. Thorn and A. Korchemsky and J. Roiban. D. Maldacena. [38] R. Metsaev. “On short and semi-short representations for four dimensional superconformal symmetry.” JHEP 0004.” JHEP 0612. Tirziu and A. “Spiky strings. 085 (2008) [arXiv:0805.” JHEP 0601. [50] G.” JHEP 0706. “Green-Schwarz string in AdS(5) x S(5): Semiclassical partition function. Tseytlin. A. Sokatchev. Y.3933 [hep-th]]. S.” Phys.” Nucl. B 639. Rev. B 785. Tseytlin. S. D 77. C. Tseytlin. Lett. “Cusp anomalous dimension in maximally supersymmetric Yang-Mills theory at strong coupling. [41] P. [47] M. F. 126005 (2008) [0802. [42] N. A. 056 (2007) [0704. Roiban and A. A. Rev. [51] M. G. B 596. “Conformal fields in the pp-wave limit. “How accurate is the quantum string Bethe ansatz?. Rev. Tseytlin. [46] F. Tseytlin.” JHEP 0811.3570 [hep-th]]. Arutyunov and E. A. A. Zarembo.4615 [hep-th]]. J. Dolan and H. Schafer-Nameki. “Spinning superstrings at two loops: strong-coupling corrections to dimensions of large-twist SYM operators. 1 (2007) [arXiv:0705. Tseytlin. “Universality of three gaugino anomalous dimensions in N = 4 SYM. 016 (2007) [0709.” JHEP 0510. Schafer-Nameki. Zamaklar and K. Lett. 066006 (2008) [arXiv:0712. “Two-loop world-sheet corrections in AdS5 xS 5 superstring. Tseytlin. 020 (2006) [arXiv:hep-th/0610250]. P.[37] B. revised (v4) 08/07 [39] R. Phys. “Light-cone superstring in AdS space-time. Osborn. [48] L. Drukker. 42 .” Phys. A. B. Kotanski. “Stringy sums and corrections to the quantum string Bethe ansatz. “Spinning strings and AdS/CFT duality. [44] S.” JHEP 0706.2039 [hep-th]]. “Uniform light-cone gauge for strings in AdS(5) x S5: Solving su(1|1) sector. 021 (2000) [arXiv:hep-th/0001204]. 019 (2007) [0708.” hep-th/0311139. D 77. 151 (2001) [arXiv:hep-th/0009171]. “Strong-coupling expansion of cusp anomaly from quantum superstring. [45] A. G. 091601 (2008) [0708.0890 [hep-th]]. Gross and A. A. M. [40] R. “Exact expressions for quantum corrections to spinning strings. “The Strong Coupling Limit of the Scaling Function from the Quantum String Bethe Ansatz.3638 [hep-th]]. R. “Anomalous dimensions at twist-3 in the sl(2) sector of N = 4 SYM.” JHEP 0711. Frolov. A. M. 044 (2007) [arXiv:0704.2479 [hep-th]]. light-like Wilson loops and pp-wave anomaly. Casteill and C.” JHEP 0208. “Comments on operators with large spin. Schafer-Nameki and M. Arutyunov and S. Kruczenski and A.” Phys. [43] N. Beccaria. A. [49] R.” JHEP 0711. 100.” JHEP 0707. 054 (2007) [arXiv:0705. Beccaria. 571 (2006) [arXiv:hep-th/0602214].” Annals Phys. 41 (2003) [hep-th/0209056]. Gromov.0663 [hep-th]]. 044 (2005) [arXiv:hep-th/0509096].

4752 [hep-th]. Tseytlin. Vieira. Rossi. B 789 (2008) 175 [arXiv:hep-th/0703191].” arXiv:0807. ArXiv:0812.0027 [hep-th]. A. Phys.A. “Large spin corrections in N = 4 SYM sl(2): still a linear integral equation. [55] R. Ishizeki. Fioravanti and M. D. Kruczenski. Vieira. N. “The AdS(5) x S5 superstring quantum spectrum from the algebraic curve. Spiky strings in AdS3 × S 1 and their AdS-pp-wave limits. Tirziu and A.2431.” arXiv:0802.” Nucl. Gromov and P. [54] D. M. Bombardelli.[53] N. Gromov. 43 . Schafer-Nameki and P. S. “Efficient precision quantization in AdS/CFT.