A perturbative Liouville prescription for the celestial three-gluon amplitude
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: High Energy Physics: Theory. The day in high-energy theory: gravity, strings, fields and the questions left open.
Kai: We're looking at the paper: "A perturbative Liouville prescription for the celestial three-gluon amplitude".
Title and authors: Kai: So this is about "A perturbative Liouville prescription for the celestial three-gluon amplitude". It sounds like they are taking something called the Mellin Liouville formulation that Stieberger, Taylor, and Zhu used and trying to fix a big problem with it.
Mira: Exactly. The original formulation had this ambiguity in how you identify the variables, which was messing up their ability to calculate corrections beyond the simplest level. They're fixing that using global conformal covariance and checking it against how quantum Liouville theory behaves when the coupling is small.
Lev: From an engineering side, that sounds like they're defining the proper way to normalize these operators so they actually make sense for a calculation. If you get the normalization wrong, all those higher-order terms just go off into infinity or become meaningless noise.
Kai: Right. So they aren't just adding some arbitrary rules; they are using physics—conformal covariance and semiclassical expansions—to nail down the operator normalizations and the parameter dictionary for this entire setup. It sounds like a lot of work just to get a solid foundation for loop corrections.
Mira: It really is about creating that controlled perturbative expansion in powers of b squared, where b is related to the Liouville coupling. That control is what makes these finite-b corrections systematic instead of just guesswork.
Lev: If they can fix the map uniquely, then Lev can start thinking about whether this structure actually maps onto something we could build a quantum error correction scheme on for real hardware.
The paper's summary: Kai: The core idea is that they resolve the residual freedom in the Mellin Liouville map and operator normalizations by imposing those two physical requirements we just talked about, which fixes how finite-b corrections organize themselves. They determine exactly how these corrections are supposed to look and where they should come from.
Mira: They show that at leading order, when you take the inverse Mellin transform in the small total momentum limit, you get the standard tree-level Yang–Mills amplitude, which is what we expect from that STZ proposal. But then they move to subleading order, and this is where things get interesting.
Lev: They derive an explicit analytic expression for the first finite-b correction at this subleading level. That's a concrete result because it’s not just some abstract structure; it’s an actual formula for what the one-loop correction looks like in terms of those modified Bessel functions.
Kai: And they show that this one-loop amplitude can be written as the leading order term multiplied by some combination of constants and these Bessel functions. That gives them a clear separation between the geometric stuff and the infrared logarithmic part when you look at the soft limit.
Mira: They also show that when you take that ratio of the one-loop to the leading order amplitude, it has a finite soft limit, which is expressed as a function of those parameters s1 and s2.
Lev: That separation between universal infrared data and ultraviolet running in celestial Yang–Mills amplitudes is something Lev can actually use to predict what kind of corrections would be important if you were scaling up the system to larger momenta.
The paper's improvements: Kai: The improvement here is taking that one-loop integral, which was originally a complex mess, and showing that it can be evaluated exactly as a linear combination of modified Bessel functions K1 and K0 with explicit coefficient functions. That’s the closed-form expression for the one-loop amplitude.
Mira: That closed form is powerful because it lets them organize the entire finite-b expansion in powers of b squared, where higher orders are then fixed iteratively by using this structure from lower orders. It suggests a very clean recursive structure for all these corrections.
Lev: For Lev, that means if we want to go to the next order in the b squared expansion, we don't have to start from scratch; you just use the same machinery they developed here and apply it again. That saves a ton of work on real hardware simulations because you know exactly what kind of structure to look for.
Kai: And when they prove that operator representation, which is showing that the polynomial dependence on Mellin variables can actually be converted into differential operators acting only on the external energy variable, that really justifies why this calculation is tractable.
Mira: That proof shows they have successfully untangled the complexity of the Mellin integral and shown it has a much simpler mathematical structure than it first appeared. It confirms that their initial assumptions about how these variables interact were actually correct.
Lev: If you can prove that kind of operator representation, then Lev can start thinking about the actual computational cost. It means the overhead for calculating this correction on a real system is manageable because you have a clean differential operator to work with rather than just integrating an intractable polynomial.
Conclusion: Kai: So to wrap up, the main thing here is that they've proposed a concrete prescription using global conformal covariance and semiclassical expansions to fix the ambiguities in the Mellin Liouville map, which gives them a controlled perturbative expansion for these celestial three-gluon amplitudes.
Mira: They got closed-form expressions for both the zeroth and one-loop contributions in terms of modified Bessel functions, and they showed how these terms separate into a soft geometry part and an infrared logarithmic term.
Lev: For Lev, this means that the leading term matches the standard Yang–Mills tree amplitude, but it's now clear how to systematically calculate those finite-b corrections by treating them as a loop expansion in powers of b squared.
Kai: So what does this mean for us? It gives us a practical framework for computing these loop corrections, which is crucial if we want to move beyond just the tree level calculations in celestial holography.
Mira: It suggests that there's a clear division between universal infrared data and the ultraviolet running in these amplitudes, which is important context for any future work connecting this to other areas of quantum field theory.
Lev: I see it as a solid piece of machinery; if they can extend this structure to four-point amplitudes or formulate perturbative bootstrap equations at loop level, then we have a much better tool for the real hardware side.
Kai: It’s definitely a solid paper on fixing the ambiguities left open in the STZ conjecture and providing that practical path forward for calculating loop corrections.
Doctoral School University of Szczecin · Institute of Physics, University of Szczecin · Computer Science Department, WSB Merito University in Poznałn
hep-th
Submitted: 2026-04-30
Updated: 2026-11-17
Comments: 31 pages; version 2 has been substantially revised
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 79/100
The gist: The paper proposes a perturbative Liouville prescription for calculating finite-b corrections to celestial three-gluon amplitudes by resolving ambiguities in the Mellin–Liouville map through global…
Key concepts
- Mellin–Liouville Map
- This is a mathematical tool used to relate different variables (Mellin space and Liouville space) when calculating celestial amplitudes. The authors resolve ambiguities in this map by requiring it to respect global conformal covariance and match the semiclassical expansion of quantum Liouville theory, ensuring a consistent physical description.
- Global Conformal Covariance
- This is a physical requirement stating that the correlation function describing the three-gluon system must remain unchanged under transformations that preserve angles (conformal transformations). Applying this condition, including the canonical prefactor, uniquely determines how to normalize the celestial gluon operators and fix inconsistencies in previous formulations.
- Modified Bessel Functions
- These special functions appear as the closed-form result for the one-loop amplitude. They provide a compact mathematical representation of the complex loop integrals. The structure of these functions allows researchers to clearly separate different physical contributions, such as those related to soft geometry versus infrared logarithmic terms.
Terminology
Summary
The paper proposes a perturbative Liouville prescription for calculating finite-b corrections to celestial three-gluon amplitudes by resolving ambiguities in the Mellin–Liouville map through global conformal covariance and compatibility with semiclassical expansions The resulting framework systematically organizes the perturbative expansion in powers of b squared, leading to closed-form expressions for the one-loop amplitude in terms of modified Bessel functions
Resolving Ambiguities and Fixing the Map
The authors resolve ambiguities in the identification of Liouville and Mellin variables by imposing two physically motivated requirements: (i) global conformal covariance of the celestial correlator, including the canonical zij-dependent prefactor, and (ii) compatibility with the semiclassical (small-b) expansion of quantum Liouville theory These conditions uniquely fix the Mellin–Liouville map and the normalization of celestial gluon operators This process determines how finite-b corrections organize and removes inconsistencies present in the original formulation.
Leading-Order Structure and Correspondence
At leading order, the inverse Mellin transform in the small total momentum limit reproduces the standard tree-level Yang–Mills amplitude, as anticipated in the STZ proposal The zeroth-order result is given by A(0)3gluon(ωi, zi, z¯i) = 4πµ˜bM4f a1a2a3⟨12⟩ 3⟨23⟩⟨31⟩M K1 2Q/M. This leading term is directly compared with the known form of the tree-level three-gluon amplitude in Yang–Mills theory. p. 5 9)
yields a consistent expansion whose leading O(b
0
) term reproduces the tree-level Yang–
Mills amplitude, while higher-order coefficients do not acquire spurious dependence on b
2
,
unlike in the STZ construction. 2 Perturbative expansion
The problem posed in [14] is to construct a unique perturbative (loop) expansion of the
three-gluon amplitude (2.Where the paper says it
Subleading Structure and One-Loop Corrections
The subleading structure organizes into a transparent structure controlled by C1 and C0, separating the soft geometry-dependent contribution from the infrared logarithmic term The one-loop amplitude is written as A(1)3gluon(ωi, zi, z¯i) = A(0)3gluon(ωi, zi, z¯i)[Q C1(ωi; ρjk) + Q C0(ωi; ρjk) K0 2Q/M K1 2Q/M]. The ratio A(1)3gluon/A(0)3gluon has a finite soft limit, A(1)3gluon A(0)3gluon = F(ρij; s1, s2) + O Q squared ln Q. p. 8 1), applied to
πµγ(b
2
)b
−2b
2
1−
P
i
σi
A˜(0)
, (2. 29)
reproduces the STZ tree-level amplitude [14]:
A
(0)
3gluon(ωi
, zi
, z¯i) = π
b
µ˜
M2
f
a1a2a3
z
3
12
z23 z31
I
(0)(ω1, ω2, ω3), (2. 1), applied to
πµγ(b
2
)b
−2b
2
1−
P
i
σi
A˜(0)
, (2. 29)
reproduces the STZ tree-level amplitude [14]:
A
(0)
3gluon(ωi
, zi
, z¯i) = π
b
µ˜
M2
f
a1a2a3
z
3
12
z23 z31
I
(0)(ω1, ω2, ω3), (2.Where the paper says it
Where the paper says it
Operator Representation and Recursion
The subleading contribution is evaluated by acting with a linear differential operator in the variables ln ωi on the leading integral I(0), yielding I(1)(ωi) = Oρ,γE I(0)(ωi). This allows for an exact closed-form expression of the one-loop integral as a linear combination of modified Bessel functions K1 and K0 with explicit coefficient functions. p. 2 At subleading order, we derive an explicit analytic expression for the first finite-b correction. Remarkably, the result can be written in closed
form in terms of modified Bessel functions, yielding a compact and tractable representation
of the one-loop amplitude in celestial variables. This observation suggests that the full finite-b expansion of the three-gluon amplitude
is naturally organized as a loop expansion in powers of b
2
, with higher-order contributions
fixed iteratively by lower-order data. 3 Leading-order structure
3.Where the paper says it
Where the paper says it
Momentum-Space Kinematics and Soft Limit
The amplitude can be written entirely in terms of the invariants sij and the energies ωi, where Q2 = s12 + s13 + s23. In the soft limit Q → 0, the coefficient QC1(ωi; ρjk) is finite: Q C1 → F(ρij; s1, s2). p. 0 The one-loop correction can be written in closed form using modified Bessel functions, and
its soft limit exhibits a clear separation into geometric and logarithmic contributions. The
resulting framework extends the STZ proposal to finite-b corrections in a consistent and
computable way. The leading term reproduces the tree-level Yang–
Mills amplitude in the small total momentum limit, as anticipated in the STZ proposal. The one-loop correction can be written in closed form using modified Bessel functions, and
its soft limit exhibits a clear separation into geometric and logarithmic contributions.Where the paper says it
Where the paper says it
Future Directions
Several directions for future work are particularly important, including computing higher-order corrections in the b squared expansion analytically or numerically Extending the map to four-point celestial amplitudes and formulating perturbative bootstrap equations at loop level are also key areas. Developing an alternative spectral-geometric formulation based on a path-integral description of quantum Liouville theory is another interesting prospect. p. 2 to loop-level contributions
in the celestial amplitude. In this paper we resolve this ambiguity by imposing two physically motivated and mutually consistent requirements: (i) global conformal covariance of the celestial correlator,
including the canonical zij -dependent prefactor, and (ii) compatibility with the semiclassical (small-b) expansion of quantum Liouville theory.Where the paper says it
A Proof of the Operator Representation
The proof justifies the identity I(1)(ωi) = Oρ,γE I(0)(ωi), showing that the polynomial dependence on the Mellin variables ∆i in (4. p. 0 com
Abstract: We study the celestial three-gluon amplitude in a dilaton background through
the Mellin–Liouville formulation proposed by Stieberger, Taylor and Zhu (STZ). The original map contains an ambiguity in the identification of Liouville and Mellin variables; we
resolve it by requiring global conformal covariance and compatibility with the semiclassical expansion of Liouville theory. 3 Leading-order structure
3. 1 Inverse Mellin representation at leading order
We calculate the inverse Mellin integral
I
(0)(ω1, ω2, ω3) =
1
2πi3 Z c+i∞
c−i∞
d∆1d∆2d∆3 M∆1+∆2+∆3−1 ω
−∆1
1 ω
−∆2
2 ω
−∆3
3
– 3 Leading-order structure
3. 1 Inverse Mellin representation at leading order
We calculate the inverse Mellin integral
I
(0)(ω1, ω2, ω3) =
1
2πi3 Z c+i∞
c−i∞
d∆1d∆2d∆3 M∆1+∆2+∆3−1 ω
−∆1
1 ω
−∆2
2 ω
−∆3
3
– com
Abstract: We study the celestial three-gluon amplitude in a dilaton background through
the Mellin–Liouville formulation proposed by Stieberger, Taylor and Zhu (STZ). The original map contains an ambiguity in the identification of Liouville and Mellin variables; we
resolve it by requiring global conformal covariance and compatibility with the semiclassical expansion of Liouville theory.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
One-Loop Soft-Kinematics Calculation
The calculation of the soft-momentum limit of the one-loop amplitude shows that Q C0(ωi; ρjk) K0 2Q/M K1 2Q/M = O Q squared ln Q. This leads to the conclusion that the ratio A(1)3gluon/A(0)3gluon has a finite soft limit, A(1)3gluon A(0)3gluon = F(ρij; s1, s2) + O Q squared ln Q. p. 8 1), applied to
πµγ(b
2
)b
−2b
2
1−
P
i
σi
A˜(0)
, (2. 29)
reproduces the STZ tree-level amplitude [14]:
A
(0)
3gluon(ωi
, zi
, z¯i) = π
b
µ˜
M2
f
a1a2a3
z
3
12
z23 z31
I
(0)(ω1, ω2, ω3), (2. Within this framework, the celestial three-gluon amplitude admits a controlled perturbative expansion. At leading order, the inverse Mellin transform in the small total
momentum limit reproduces the standard tree-level Yang–Mills amplitude, in agreement
with the intuition underlying [14].Where the paper says it
Where the paper says it
Concluding Remarks
The paper proposes a concrete resolution of the ambiguity left open in the STZ conjecture and a practical framework for computing finite-b (loop) corrections to tree-level celestial three-gluon amplitudes A central result is that the zeroth- and oneloop contributions admit finite analytic expressions in terms of modified Bessel functions. The leading term reproduces the STZ tree-level ansatz and, equivalently, the standard Yang–Mills tree amplitude The subleading terms organize into a transparent structure controlled by C1 and C0, separating the soft geometry-dependent contribution from the infrared logarithmic term. p. 0 The leading term reproduces the tree-level Yang–
Mills amplitude in the small total momentum limit, as anticipated in the STZ proposal. The one-loop correction can be written in closed form using modified Bessel functions, and
its soft limit exhibits a clear separation into geometric and logarithmic contributions.Where the paper says it
In this appendix we justify the identity I(1)(ωi) = Oρ,γE I(0)(ωi), i.e. that the polynomial dependence on the Mellin variables ∆i in (4.3) may be converted into differential operators acting on the leading inverse Mellin integral I(0). The key observation is that Di acts only on the external variable ωi and leaves the Mellin variables ∆j untouched.
The paper proposes a concrete resolution of the ambiguity left open in the STZ conjecture and a practical framework for computing finite-b (loop) corrections to tree-level celestial three-gluon amplitudes A central result is that the zeroth- and oneloop contributions admit finite analytic expressions in terms of modified Bessel functions The leading term reproduces the STZ tree-level ansatz and, equivalently, the standard Yang–Mills tree amplitude The subleading terms organize into a transparent structure controlled by C1 and C0, separating the soft geometry-dependent contribution from the infrared logarithmic term.
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Improved AI can perform systematic derivation of finite-b corrections by utilizing global conformal covariance and compatibility with semiclassical expansions of Liouville theory to uniquely fix operator normalizations and parameter dictionaries, resolving ambiguities present in previous formulations.
-
Improved AI can compute the one-loop amplitude in closed form using modified Bessel functions, providing a compact representation that separates geometric and logarithmic contributions in the soft limit.
-
Improved AI can analyze the structure of higher-order corrections by recognizing the recursive structure suggested by Mellin integrals and polynomial dependence on sigma parameters, enabling iterative calculation of coefficients Omega n.
-
Improved AI can determine if a physical quantity exhibits a finite soft limit or an infrared logarithmic term by analyzing the small-Q expansion of modified Bessel functions and kinematic functions derived from the structure constants.
Abstract
We study the celestial three-point Mellin representation for three-gluon scattering in a dilaton background within the Mellin--Liouville formulation of Stieberger, Taylor and Zhu (STZ). We formulate an alternative perturbative prescription in which the light Liouville parameters are kept fixed in the small- b expansion. Requiring global conformal covariance, a controlled light-operator expansion of the Liouville DOZZ structure constant, and consistency with the standard celestial three-gluon representation fixes the corresponding Mellin-Liouville map and operator normalization. Starting from the full Liouville three-point function, we derive the leading and first subleading terms in the b squared expansion. The leading term reproduces the standard tree-level celestial Yang--Mills representation in the appropriate small-total-momentum limit, while the first subleading term is interpreted as the first coefficient of a controlled Liouville deformation of this tree-level construction. We develop an operator representation of the inverse Mellin transform, whereby polynomial dependence on the Mellin parameters is converted into differential operators acting on the leading Mellin integral. The resulting subleading contribution is obtained in closed form in terms of modified Bessel functions, with characteristic logarithmic and nonanalytic behaviour at small total momentum. Finally, we clarify the relation between the fixed- σ and STZ finite- b parametrizations by deriving the induced nonlinear reparametrization of the Mellin dimension and its action on the inverse Mellin representation. The result provides a systematic framework for studying finite- b Liouville deformations and their possible relation to perturbative structures in celestial Yang-Mills theory.
Sources
- Lectures on the Infrared Structure of Gravity and Gauge Theory
- Celestial Holography
- Lectures on Celestial Holography
- Celestial holography: An asymptotic symmetry perspective
- Lectures on Celestial Amplitudes
- Minimalist approach to the classification of symmetry protected topological phases
- Tree-level gluon amplitudes on the celestial sphere
- Gauge and Gravity Amplitudes on the Celestial Sphere
- MHV Graviton Scattering Amplitudes and Current Algebra on the Celestial Sphere
- A Celestial Dual for MHV Amplitudes
- Remarks on celestial amplitudes and Liouville theory
- MHV leaf amplitudes from parafermions
- Analytic Continuation of Liouville Theory
- Conformal field theory on the plane
- Liouville Perturbation Theory
- Small- b expansion of the DOZZ formula for light operators
Related papers
- The facts on quantum self-dual Einstein gravity and self-dual Yang-Mills theory
- Quantum Langlands correspondence for Painlev'e II
- Fortuitous BMN Operators from Multiple Cores
- Building far-from-equilibrium effective field theories using shift symmetries
- Inflation, Cosmic Perturbations and Non-Gaussianities
- Wavefunctions of AdS 3 Universes and T-deformed Torus Partition Functions