Infrared singularities of multileg amplitudes with a massive particle at three loops

Basic Information

• Paper ID: 2510.27567
• Title: Infrared singularities of multileg amplitudes with a massive particle at three loops
• Authors: Einan Gardi, Zehao Zhu (Higgs Centre for Theoretical Physics, University of Edinburgh)
• Categories: hep-ph (High Energy Physics - Phenomenology), hep-th (High Energy Physics - Theory)
• Submission Date: October 31, 2025
• Paper Link: https://arxiv.org/abs/2510.27567

Abstract

This paper determines the complete three-loop QCD soft anomalous dimensions for multileg amplitudes involving a single massive colored particle and an arbitrary number of massless particles. This is achieved through a novel strategy based on asymptotic light-cone expansions of semi-infinite Wilson line correlators using the Method of Regions (MoR). The resulting region integrals depend only on rescaling-invariant ratios that remain finite in the limit. The authors evaluate these integrals using differential equation techniques. The results are expressed in terms of uniform weight-five generalized polylogarithms (GPLs) in a twelve-letter alphabet of three variables, and are compatible with the massless limit as well as two-particle and three-particle collinear factorization.

Research Background and Motivation

Core Problem

The core problem addressed in this paper is the calculation of three-loop soft anomalous dimensions for QCD amplitudes involving one massive particle and an arbitrary number of massless particles. This is a key question in the study of infrared singularity structures in quantum chromodynamics (QCD).

Significance

1. Theoretical Significance: Infrared singularities possess a highly constrained universal structure, whose exponentiation is controlled by soft anomalous dimensions. This is a central universal quantity in amplitude research.
2. Phenomenological Applications: Important applications in collider physics, particularly:
   • Resummation of top quark production cross sections
   • Processes involving heavy quarks, light quarks, and gluons
   • Precise predictions require resummation-enhanced logarithmic corrections
3. Computational Challenges: Cancellation of infrared singularities involves summation over different multiplicities, requiring complex subtraction techniques

Limitations of Existing Methods

1. Massless Case: Soft anomalous dimensions for multileg massless scattering are known to three loops 35
2. Massive Case: Current techniques for cases involving massive particles extend only to two loops 48, 49, 50
3. Partial Results: Recently, only three-loop corrections for one massive particle interacting with two massless particles were obtained 60

Research Motivation

Although soft anomalous dimensions show significant simplification compared to finite corrections (spin-independent, rescaling symmetry), these simplifications are difficult to exploit in calculations because the multiplicative renormalizability of Wilson line correlators holds only when the Wilson lines are non-light-like. The novel strategy proposed in this paper directly addresses this issue.

Core Contributions

1. Complete Three-Loop Result: First complete three-loop soft anomalous dimensions for amplitudes with one massive particle and arbitrary massless particles, particularly the previously unknown four-point function $F_{1,3}^{(3)}$
2. Novel Computational Strategy: Proposes a new method for asymptotic light-cone expansion of timelike Wilson lines based on the Method of Regions (MoR), such that the resulting integrals depend only on rescaling-invariant ratios that remain finite in the limit
3. Analytic Results: Results expressed as uniform weight-five generalized polylogarithms (GPLs) in a twelve-letter alphabet of three variables {x, z, z̄}
4. Consistency Verification: Results are fully compatible with:
   • Massless limit (reproducing known $F_{0,4}^{(3)}$)
   • Two-particle and three-particle collinear factorization
   • First entry condition
   • Galois symmetry
5. Open-Source Implementation: Complete Mathematica implementation provided on GitHub

Detailed Methodology

Task Definition

Calculate the infrared singularity factor $Z$ in the renormalized amplitude $M = ZH$, where: $Z = P\exp\left\{-\int_0^\mu \frac{d\tau}{\tau}\Gamma(\tau, \alpha_s(\tau^2))\right\}$

The general form of soft anomalous dimension $\Gamma$ includes dipole and quadrupole contributions. The paper's goal is to determine the three-loop quadrupole function $F_{1,3}^{(3)}(r_{ijQ}, r_{jkQ}, r_{ikQ})$, which describes long-range interactions between one massive particle and three massless particles.

Core Methodological Architecture

1. Wilson Line Correlator Method

Starting from the correlator of four timelike Wilson lines: $\langle \Phi^{(m)}_{\beta_I}\Phi^{(m)}_{\beta_J}\Phi^{(m)}_{\beta_K}\Phi^{(m)}_{\beta_Q}\rangle = \exp\left[\sum_n \left(\frac{\alpha_s}{4\pi}\right)^n w^{(n)}\right]$

where $\beta_U^2 > 0$ denotes timelike velocities, and superscript $(m)$ denotes infrared regularization.

2. Light-Cone Expansion via Method of Regions

Key innovation: Using MoR for asymptotic light-cone expansion of timelike Wilson lines: $\beta_I^2 \sim \lambda_i, \quad \beta_J^2 \sim \lambda_j, \quad \beta_K^2 \sim \lambda_k$

Expansion form: $T_{\lambda_i,\lambda_j,\lambda_k}[Y^{(-1)}_{(IJ;KQ)}(\{\alpha_{VU}\})] = Y^{(-1)}_{(ij;kQ)}(r_{ikQ}, r_{jkQ}, r_{ijQ}; \lambda_i, \lambda_j, \lambda_k) + O(\lambda)$

The right-hand side depends only on three rescaling-invariant ratios $r_{uvQ}$ and logarithms of expansion parameters.

3. Region Integral Decomposition

The complete asymptotic expansion is given by the sum of region integrals: $T_{\lambda_i,\lambda_j,\lambda_k}[W_{1111}] = \sum_{[n_i,n_j,n_k]} \lambda_i^{n_i\epsilon}\lambda_j^{n_j\epsilon}\lambda_k^{n_k\epsilon} Y^{[n_i,n_j,n_k]}_{(ij;kQ)}(r_{ikQ}, r_{jkQ}, r_{ijQ})$

where regions are identified by scaling laws $\lambda_u^{n_u\epsilon}$ ($n_u$ integers).

For $W_{1111}$ (connected web), 56 regions are obtained. The most complex 8 regions (hard region and 7 collinear mode regions) depend on all three $r_{uvQ}$ ratios, represented by color-coded Feynman diagrams (blue=hard, green/orange/brown=collinear to i/j/k directions).

Technical Innovations

1. Rescaling-Invariant Variable Parametrization

Introduction of new variable parametrization {x, z, z̄}: $r_{ijQ} = \frac{x(x+z-\bar{z})}{(1-z)(1-\bar{z})}, \quad r_{jkQ} = \frac{x(x+z-\bar{z})}{z\bar{z}}, \quad r_{ikQ} = \frac{x(x+z-\bar{z})}{(1-z)(1-\bar{z})z\bar{z}}$

This parametrization rationalizes the alphabet, simplifying the extended alphabet (containing $\Delta_1, \Delta_2$) to the physical alphabet.

2. Canonical Form of Differential Equations

• Identification of uniformly transcendental integrals
• Each region individually possesses uniform transcendentality
• Use of Initial package to obtain canonical form
• Solution order-by-order using PolyLogTools

3. Simplification of Region Summation

Key observation: Although individual region integrals exhibit:

• High-order $\epsilon$ poles
• $\Delta_1, \Delta_2$ singularities

After summing all regions:

• Only a single ultraviolet $1/\epsilon$ pole remains
• $\Delta_1, \Delta_2$ singularities completely cancel
• Only weight-five functions with physical alphabet remain

4. Boundary Condition Determination Strategy

Boundary values are uniquely determined through the following physical constraints:

1. Bose Symmetry: $Y_{(ji;kQ)} = -Y_{(ij;kQ)}$
2. Single Pole Condition: Connected web has no sub-divergences, only single $1/\epsilon$ pole
3. Light-Cone Limit: $x \to 0$ reproduces known $F_{0,4}$ results
4. Collinear Limit: Satisfies strict collinear factorization requirements, $y$-independence
5. First Entry Condition: Symbol must have form $r_{ijQ} \otimes \cdots + r_{jkQ} \otimes \cdots + r_{ikQ} \otimes \cdots$

Contributions from Other Webs

W1112 (One Three-Gluon Vertex)

• Characteristic: Double poles, sensitive to infrared regularization
• Strategy: Unify three parameters $\lambda_i = \lambda_j = \lambda_k \equiv \lambda$ before expansion
• Use CANONICA to obtain canonical form
• New singularities $1+r_{uvQ}$ and $r_{uvQ}+r_{uwQ}$ appear but do not contribute to single pole

W1122 and W1113 (Multi-Gluon Exchange Webs)

• Already computed using GPLs 127, 128
• Direct expansion of known results

Three-Line and Two-Line Webs

• Obtained from known results 36
• Or derived using color charge conservation

Experimental Setup

Computational Toolchain

This is a pure theoretical calculation using the following main computational tools:

1. Region Determination:
   • pySecDec 112
   • AmpRed 74
   • Utilization of faces of Newton polytopes of graph polynomials to determine complete region sets
2. Differential Equation Derivation and Solution:
   • AmpRed 74: Derivation of differential equations
   • Kira 117-119: Integral reduction
   • LiteRed 120, 121: Integral reduction
   • Initial 123: Obtaining canonical form
   • CANONICA 126: Canonical form for W1112
   • PolyLogTools 124: Solution in terms of GPLs
3. Alphabet Rationalization:
   • RationalizeRoots 131

Kinematic Configuration

• Euclidean Region: $\beta_U \cdot \beta_V < 0$ for all velocity pairs, ensuring no hidden regions
• Seven disconnected Euclidean domains $R_i^E$ ($i=1,\ldots,7$)
• Five disconnected physical domains $R_i^P$ ($i=1,\ldots,5$)

Verification Strategy

Multi-level consistency checks:

1. Single pole structure
2. Massless limit ($x \to 0$)
3. Collinear limit ($\delta \to 0$)
4. Bose symmetry
5. Galois symmetry
6. First entry condition
7. Reality in Euclidean regions

Experimental Results

Main Results

Complete Three-Loop Soft Anomalous Dimension

Complete form of one-mass soft anomalous dimension (Equation 2): $\Gamma = \sum_{i<j} T_i \cdot T_j \gamma_{\text{cusp}}(\alpha_s)\log\left(-\frac{\tau^2}{2p_i \cdot p_j}\right) + \sum_i \gamma_i(\alpha_s)$$+ \sum_i T_i \cdot T_Q \gamma_{\text{cusp}}(\alpha_s)\log\left(-\frac{\tau\sqrt{p_Q^2}}{2p_i \cdot p_Q}\right) + \Omega_Q(\alpha_s)$$+ \text{Quadrupole Terms} + O(\alpha_s^4)$

where quadrupole terms include:

• $F_{0,4}$ (four massless particles, known)
• $F_{0,3}$ (three massless particles, known)
• $F_{1,2}$ (one massive two massless, known 60)
• $F_{1,3}$ (one massive three massless, new result in this paper)

Analytic Structure of $F_{1,3}^{(3)}$

Antisymmetric Form: $F_{1,3}^{(3)}(r_{ijQ}, r_{jkQ}, r_{ikQ}) = 16[F_{1,3}(x, 1-\bar{z}, 1-z) - F_{1,3}(x, z, \bar{z})]$

Alphabet (12 letters): $\{\omega_i\}_{\text{phy}} = \{z, \bar{z}, 1-z, 1-\bar{z}\} \cup \{x, \bar{z}-z-x\}$$\cup \{x+z, \bar{z}-x, 1-z-x, 1-\bar{z}+x, z+x-z\bar{z}, \bar{z}-x-z\bar{z}\}$

Physical Interpretation:

• First 4 letters: $r_{uvQ} \to \infty$ (same as $F_{0,4}$)
• Middle 2 letters: $r_{uvQ} \to 0$
• Last 6 letters: $r_{uvQ} \to 1$

Galois Symmetry: $F_{1,3}(x, z, \bar{z}) = F_{1,3}(-x, \bar{z}, z) = F_{1,3}(-x-z+\bar{z}, z, \bar{z})$

Key Verification Results

1. Collinear Limit

$F_{1,3}^{(3)}(0, r, r) = -F_{1,3}^{(3)}(r, 0, r) = 2F_{0,3}^{(3)}$ Equivalently: $F_{1,3}(0, 1, 1) = 4C, \quad F_{1,3}(0, 0, 0) = 0$ where $C = 2\zeta(2)\zeta(3) + \zeta(5)$

2. Light-Cone Limit (Massless Limit)

$F_{1,3}(0, z, \bar{z}) = F_{0,4}(z, \bar{z})$ Perfectly reproduces the known result from reference 35, also representing the three-particle collinear limit.

3. First Entry of Symbol

$S[Y_{(ij;kQ)}^{(-1)}] = r_{ijQ} \otimes \cdots + r_{jkQ} \otimes \cdots + r_{ikQ} \otimes \cdots + \lambda_i \otimes \cdots + \lambda_j \otimes \cdots + \lambda_k \otimes \cdots$

Only physical channels (Mandelstam invariants) have branch points on the principal sheet.

4. Simplification of Region Summation

• Individual region integrals: High-order $\epsilon$ poles, containing $\Delta_1, \Delta_2$ singularities
• After summing all regions:
  • Only single $1/\epsilon$ pole
  • $\Delta_1, \Delta_2$ completely vanish
  • Expansion parameters $\lambda$ completely cancel
  • Depend only on physical variables {x, z, z̄}

Numerical Characteristics

• Weight: Uniform weight five
• Number of Variables: Three (x, z, z̄)
• Alphabet Size: 12 (physical) + 2 (extended, appearing only in intermediate steps)
• Number of Regions: 56 (for each channel of W1111)
• Most Complex Regions: 8 (depending on all three kinematic variables)

Related Work

Theoretical Framework for Infrared Singularities

1. Establishment of Wilson Line Methods 1-11:
   • Polyakov, Arefeva et al. established multiplicative renormalizability of Wilson lines
   • Korchemsky-Radyushkin developed cusp anomalous dimension theory
2. Factorization and Exponentiation 12-18:
   • Catani, Sterman et al. proved exponentiation of singularities
   • Dixon, Gardi, Magnea et al. developed systematic theory of soft anomalous dimensions
3. Rescaling Symmetry 19-22, 61:
   • Gardi-Magnea, Becher-Neubert identified rescaling-invariant ratios
   • Dixon emphasized importance of this symmetry for computational simplification

Diagrammatic Exponentiation Method

4. Systematic Theory of Webs 26-33:
   • Gatheral, Frenkel-Taylor established diagrammatic exponentiation
   • Gardi et al. developed renormalization theory of multi-Wilson line vertices

Three-Loop Results for Massless Case

5. Completely Massless Scattering 35-37:
   • Almelid, Duhr, Gardi (2016) computed $F_{0,4}^{(3)}$
   • Using Mellin-Barnes technique with light-cone expansion

Progress in Massive Cases

6. Two-Loop Results 48, 49:
   • Ferroglia et al. (2009) provided two-loop one-mass soft anomalous dimensions
7. Three-Loop Partial Results 60:
   • Liu-Schalch (2022) computed $F_{1,2}^{(3)}$ (one massive two massless)

Development of Method of Regions

8. Theoretical Foundations of MoR 63-74:
   • Beneke-Smirnov (1998) established Method of Regions
   • Recent developments: Newton polytope method for determining complete region sets
   • Automated tools: AmpRed, pySecDec, etc.
9. Prior Work by Authors 62:
   • Gardi-Zhu proposed strategy for applying MoR to Wilson line correlators
   • Two-loop examples verified method effectiveness

Unique Contributions of This Paper

Compared to related work, this paper:

1. First Completion: Complete calculation of one-mass three-loop soft anomalous dimensions
2. Methodological Innovation: First systematic application of MoR to three-loop four-line Wilson line correlators
3. Technical Breakthrough: Handling complex integrals depending on three kinematic variables
4. Paradigm Establishment: Establishes paradigm for calculations at higher loops and with more massive particles

Conclusions and Discussion

Main Conclusions

1. Completeness: Determines complete three-loop soft anomalous dimensions for QCD amplitudes with one massive particle and arbitrary massless particles
2. Analytic Structure: Results expressed as uniform weight-five GPLs in a twelve-letter alphabet of three variables, with letters directly corresponding to physical collinear singularities $r_{uvQ} \to \{0, 1, \infty\}$
3. Consistency: Results satisfy all physical constraints:
   • Light-cone limit reproduces $F_{0,4}^{(3)}$
   • Two-particle and three-particle collinear factorization
   • First entry condition
   • Galois symmetry
   • Reality in Euclidean regions
4. Methodological Breakthrough: Demonstrates effectiveness of novel MoR-based strategy, where each region integral depends only on three finite rescaling-invariant ratios rather than the original six variables

Limitations

1. Particle Number Restriction: Currently handles only one massive particle; multiple massive particles (e.g., $t\bar{t}$ production) remain uncalculated
2. Loop Order Restriction: Three-loop result; four loops and beyond exhibit new structures 38-40
3. Regularization Dependence: Non-connected webs (e.g., W1112) are sensitive to infrared regularization, though final ultraviolet poles are unaffected
4. Computational Complexity:
   • 56 region integrals, 8 most complex
   • Requires coordination of multiple specialized software packages
   • Boundary condition determination requires multiple physical constraints
5. Analytic Complexity: Although results are uniformly transcendental, GPL expressions are quite complex (see GitHub implementation)

Future Directions

1. Generalization to Multiple Particles:
   • Two or more massive particles (e.g., $t\bar{t}$ production)
   • Combinations of different masses
2. Higher Loop Orders:
   • Four-loop one-mass soft anomalous dimensions
   • Requires handling new color structures 38-40
3. Phenomenological Applications:
   • N³LL resummation for top quark pair production
   • Precise predictions for heavy quark jets
   • Precise predictions for LHC and future colliders
4. Method Development:
   • Further optimization of Method of Regions
   • Development of more efficient boundary condition determination strategies
   • Automation of computational workflows
5. Mathematical Structure:
   • Deeper understanding of geometric origins of alphabets
   • Study of special properties of multivariate GPLs
   • Exploration of connections with string theory amplitudes

In-Depth Evaluation

Strengths

1. Methodological Innovation (★★★★★)

• Paradigm Shift: Systematic application of MoR to multi-line Wilson line correlators is major innovation
• Technical Breakthrough: Successfully handles three-loop four-line problem, reducing variables from 6 to 3
• Universality: Method generalizable to higher loops and more particles

2. Computational Rigor (★★★★★)

• Multi-Level Verification:
  • Single pole structure
  • Light-cone limit
  • Collinear limit
  • Symmetries
  • First entry condition
• Region Completeness: Newton polytope method ensures no hidden regions
• Boundary Conditions: Uniquely determined through physical constraints

3. Result Importance (★★★★★)

• Fills Gap: Completes last piece of one-mass three-loop soft anomalous dimensions
• Phenomenological Value: Direct application to precise predictions in top quark physics
• Theoretical Value: Verifies universal structure of soft anomalous dimensions

4. Technical Details (★★★★★)

• Complete Toolchain: From region determination to differential equation solution
• Alphabet Rationalization: Clever parametrization eliminates non-physical singularities
• Open-Source Implementation: Complete Mathematica code on GitHub

5. Clarity of Presentation (★★★★☆)

• Clear Structure: Main text concise, supplementary materials detailed
• Physical Intuition: Color-coded Feynman diagrams clearly show region structure
• Formula Standards: Clear symbol definitions, easy to understand

Weaknesses

1. Computational Complexity

• Requires multiple specialized software packages (pySecDec, Kira, AmpRed, etc.)
• Boundary condition determination requires physical insight, difficult to fully automate
• GPL expressions complex, numerical evaluation potentially unstable

2. Generalization Difficulty

• Method in principle generalizable, but computational cost increases dramatically for multiple particles
• Four loops introduce new color structures, requiring further method development

3. Numerical Implementation

• Paper lacks detailed discussion of numerical evaluation
• Complex GPLs potentially numerically unstable in certain kinematic regions
• Missing comparison with existing numerical methods

4. Physical Interpretation

• Limited discussion of geometric/physical origins of alphabet
• Physical meaning of $\Delta_1, \Delta_2$ singularities not deeply explored
• Could discuss more connections with other theoretical frameworks (e.g., SCET)

Impact Assessment

Contribution to Field (★★★★★)

1. Theoretical Completeness: Completes three-loop calculation for one-mass case, milestone in QCD precision calculations
2. Methodology: Establishes new paradigm for handling complex Wilson line correlators
3. Citation Potential: Expected to become standard reference in field

Practical Value (★★★★☆)

1. Direct Applications:
   • N³LL resummation for top quark production
   • Theoretical predictions for heavy quark jets
   • LHC precision physics
2. Indirect Impact:
   • Provides theoretical input for Monte Carlo event generators
   • Improves PDF extraction
   • Background predictions for new physics searches
3. Limitations: Currently only one massive particle; practical applications (e.g., $t\bar{t}$) require further generalization

Reproducibility (★★★★☆)

1. Advantages:
   • Detailed method description
   • Open-source code (GitHub)
   • Clear toolchain
2. Challenges:
   • Requires multiple specialized software packages
   • High computational resource requirements
   • Boundary condition determination requires specialized knowledge

Applicable Scenarios

Direct Applications

1. Top Quark Physics:
   • Single top production ($tqb$, $tW$)
   • Top quark + jets
   • Top quark decay
2. Heavy Quark Production:
   • Bottom quark + multiple jets
   • Heavy quark resonances + jets

Requiring Generalization

1. Top Quark Pair Production: Requires three-loop results for two massive particles
2. Higgs + Heavy Quark: May involve different mass scales
3. Beyond Standard Model: Production of new massive particles

Methodological Generalization

1. Higher Loop Orders: Four-loop one-mass, three-loop multiple masses
2. Other Gauge Theories: N=4 SYM, etc.
3. Gravitational Theories: Graviton scattering amplitudes

Summary of Technical Highlights

Core Technical Innovations

1. Systematic MoR Application: First systematic application in three-loop four-line problem
2. Variable Simplification: Reduction from 6 to 3 variables, dramatically reducing complexity
3. Alphabet Rationalization: Eliminates non-physical singularities $\Delta_1, \Delta_2$
4. Region Summation: Demonstrates cancellation of high-order poles and non-physical singularities

Computational Techniques

1. Region Determination: Newton polytope method ensures completeness
2. Differential Equations: Canonical form + uniform transcendentality
3. Boundary Conditions: Multiple physical constraints uniquely determine solution
4. Symbol Verification: First entry condition validation

Physical Insights

1. Rescaling Symmetry: Fully exploited to reduce variable count
2. Collinear Factorization: Provides strong constraints
3. Galois Symmetry: Reflects parametrization redundancy
4. Euclidean Reality: Verifies correctness of analytic continuation

Selected References

Key Methodological References

• 35 Ø. Almelid, C. Duhr, and E. Gardi, PRL 117, 172002 (2016) - Massless three-loop
• 62 E. Gardi and Z. Zhu, arXiv:2509.18017 - MoR strategy
• 63 M. Beneke and V. A. Smirnov, NPB 522, 321 (1998) - MoR foundations

Theoretical Framework

• 1-11 Wilson line renormalizability series
• 25 N. Agarwal et al., Phys. Rept. 994, 1 (2023) - Review

Related Results

• 60 Z. L. Liu and N. Schalch, PRL 129, 232001 (2022) - $F_{1,2}^{(3)}$
• 48, 49 Ferroglia et al., JHEP (2009) - Two-loop one-mass

Overall Assessment: This is an important work of milestone significance in the field of QCD precision calculations. The methodological innovations are substantial, the calculations rigorous, and the results complete, with important implications for both theory and phenomenology. While generalization to more complex scenarios remains challenging, the paradigm established in this paper will guide future research. Highly recommended for researchers working on QCD precision calculations, resummation theory, and collider physics.