pith. sign in

arxiv: 2506.21433 · v3 · submitted 2025-06-26 · ✦ hep-ph · hep-ex

Probing Neutral Triple Gauge Couplings via ZZ Production at e^+e^- Colliders with Machine Learning

Pith reviewed 2026-05-19 07:41 UTC · model grok-4.3

classification ✦ hep-ph hep-ex
keywords neutral triple gauge couplingsZZ productione+e- collidersmachine learningSMEFTdimension-8 operatorsnew physics scales
0
0 comments X p. Extension

The pith

Machine learning on angular distributions in ZZ production probes neutral triple gauge couplings up to multi-TeV scales at future e+e- colliders.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops form factors for neutral triple gauge couplings that are consistent with electroweak symmetry breaking and correspond to dimension-8 operators in the Standard Model Effective Field Theory. It examines ZZ boson pair production at high-energy e+e- colliders, considering both visible and invisible decays of the Z bosons. Machine learning is applied to the angular distributions of the decay fermions to better distinguish signal from background. This method, combined with beam polarization, significantly improves the ability to detect deviations caused by these couplings. As a result, the new physics scales associated with these couplings can be constrained up to the multi-TeV range at facilities like CEPC, FCC-ee, ILC, and CLIC.

Core claim

Neutral triple gauge couplings first appear in dimension-8 operators of the SMEFT. The work formulates the corresponding form factors for ZZV* vertices that respect the spontaneous breaking of the electroweak gauge symmetry and match the broken-phase operators. Through analysis of ZZ production with machine learning techniques handling the four-body final states, it demonstrates that future e+e- colliders can achieve sensitivity to these couplings that reaches new physics scales of several TeV. A specific dimension-8 operator is identified that affects only the pure ZZZ* coupling without impacting the ZZg* coupling, and correlations between the two are explored.

What carries the argument

The machine learning classifier that uses angular distributions of final-state fermions from ZZ decays to suppress Standard Model backgrounds, applied to dimension-8 nTGC form factors for ZZV* vertices.

Load-bearing premise

The nTGC form factors are formulated to be compatible with the spontaneous breaking of the SU(2) x U(1) electroweak gauge symmetry and to match the dimension-8 operators in the broken phase.

What would settle it

A precise measurement at one of the proposed colliders showing ZZ production rates and angular distributions consistent with Standard Model expectations up to center-of-mass energies of several TeV would rule out nTGC new physics scales below the multi-TeV range.

Figures

Figures reproduced from arXiv: 2506.21433 by Hong-Jian He, John Ellis, Rui-Qing Xiao, Shi-Ping Zeng.

Figure 1
Figure 1. Figure 1: Unitarity bounds on the nTGC form factors f γ 5 and f Z 5 in plot (a) and on the new physics scales ΛBW˜ and Λ3Z in plot (b). These bounds are derived from the p-wave amplitudes for the reaction e −e +→ZZ. which can be obtained from the unitarity bound (2.18) by the replacement c V L ↔ c V R . using these results we derive the following unitarity bounds on the nTGC form factors f γ 5 and f Z 5 [PITH_FULL_… view at source ↗
Figure 2
Figure 2. Figure 2: Feynman diagrams that contribute to the reaction e −e +→ZZ. Plot (a) shows the signal process containing the nTGC vertices ZZV ∗ , and plots (b) and (c) show the leading SM background contributions. As shown in [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Kinematic structure of the reaction e −e +→ ZZ followed by the fermionic decays Z→f ¯f , in the e −e + collision frame. We see that the SM contribution scales as s −1 and decreases with energy, whereas the interference term scales as s 0/Λ 4 and thus is insensitive to the energy. Finally, the last contribution (arising from the squared amplitude) scales as s 2/Λ 8 and increases with energy rapidly. For the… view at source ↗
Figure 4
Figure 4. Figure 4: Normalized angular distributions in the polar scattering angle θ for different collider energies, √ s = (0.25, 0.5, 1, 3) TeV, are shown in plots (a)-(d) respectively. In each plot, the black, red and blue curves denote the contributions from the SM, the O(Λ−4 ) term, and the O(Λ−8 ) term, respectively. where k= 1 2MZ holds since we treat the final-state fermions as effectively massless. Here the positive … view at source ↗
Figure 5
Figure 5. Figure 5: Normalized angular distribution in the polar angle θa in the Z decay frame for different collider energies, √ s = (0.25, 0.5, 1, 3) TeV, as shown in plots (a)-(d) respectively. In each plot, the black, red and blue curves denote the contributions from the SM, the O(Λ−4 ) interference, and the O(Λ−8 ) term respectively, where the red and blue curves exactly overlap. to θa and ϕa are similar to the correspon… view at source ↗
Figure 6
Figure 6. Figure 6: Normalized angular distributions in the azimuthal angle ϕa of down-type quarks for different collider energies, √ s = (0.25, 0.5, 1, 3) TeV, are shown in plots (a)-(d) respectively. In each plot, the black, red, green and blue curves denote the contributions from the SM, the O(Λ−4 ) interference with γ ∗ , the O(Λ−4 ) interference with Z ∗ , and the O(Λ−8 ) term, respectively, where the blue and black curv… view at source ↗
Figure 7
Figure 7. Figure 7 [PITH_FULL_IMAGE:figures/full_fig_p022_7.png] view at source ↗
Figure 8
Figure 8. Figure 8 [PITH_FULL_IMAGE:figures/full_fig_p023_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Sensitivity reaches for the nTGC form factors in plot (a) and for the nTGC new physics scales in plot (b) for various e −e + collision energies and integrated luminosities. In this figure, plot (a) presents the (2σ, 5σ) sensitivity reaches in (heavy, light) colors, whereas plot (b) shows the (2σ, 5σ) sensitivity reaches in (light, heavy) colors. across different collider energies. Here the smallest enhance… view at source ↗
Figure 10
Figure 10. Figure 10: Distribution of probabilities P(γ ∗ ) and P(Z ∗ ) of simulated events belonging to γ ∗ signal or Z ∗ signal category within the 4 regions of Eq.(4.12) for the case of e +e −→V ∗→ZZ→(ℓ ¯ℓ)(d ¯d) at a collider energy √ s = 250 GeV. We divide the phase space of the final states into 4 regions based on the signs of the angular 27 [PITH_FULL_IMAGE:figures/full_fig_p027_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Distribution of probabilities P(γ ∗ ) and P(Z ∗ ) of simulated events belonging to γ ∗ signal or Z ∗ signal category within the 4 regions of Eq.(4.12) for the case of e +e −→V ∗→ZZ→(ℓ ¯ℓ)(d ¯d) at a collider energy √ s = 3 TeV. distribution f 1 Ω in Eq. (4.2) or f 1 Ω′ in Eq.(4.5): R++ [PITH_FULL_IMAGE:figures/full_fig_p028_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Bounds on nTGC form factors correlations and on the nTGC new physics scales at the 2σ level, for the indicated collider energies and integrated luminosities. The plots (a)-(b) are shown for collision energy √ s = 250GeV and plots (c)-(d) are for collision energy √ s = 3TeV. The contours of plots (a) and (c) depict the correlations between the form factors (f γ 5 , f Z 5 ), whereas the contours of plots (b… view at source ↗
read the original abstract

Neutral triple gauge couplings (nTGCs) first arise from the dimension-8 operators of the Standard Model Effective Field Theory (SMEFT), rather than the dimension-4 SM Lagrangian and dimension-6 SMEFT operators, opening up a unique window for probing new physics at the dimension-8 level. In this work, we formulate the nTGC form factors of $ZZV^*$ ($V\!\!=\!Z,\gamma$) that are compatible with the spontaneous breaking of the SU(2)$\otimes$U(1) electroweak gauge symmetry and consistently match the dimension-8 nTGC operators in the broken phase. We study the sensitivities for probing both the $ZZV^*$ form factors and the corresponding new physics scales through $ZZ$ production (with visible/invisible fermionic $Z$ decays) at high energy $e^+e^-$ colliders including CEPC, FCC-ee, ILC and CLIC. In particular, we identify the dimension-8 operator that contributes to the pure triple $Z$ boson coupling $ZZZ^*$ alone, but not the mixed $ZZ\gamma^*$ coupling. We further study the correlations between probes of the $ZZZ^*$ and $ZZ\gamma^*$ couplings. Using machine learning, we show that angular distributions of the final-state fermions can play key roles in suppressing the SM backgrounds. The sensitivities can be further improved by using polarized $e^\mp$ beams. We demonstrate that machine learning is advantageous for handling the 4-body final states from $ZZ$ decays and improves significantly the sensitivity reaches of probes of nTGCs in $e^+e^-$ collisions. We find that nTGC new physics scales can be probed up to the multi-TeV scale at the proposed $e^+e^-$ colliders.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper claims to formulate neutral triple gauge couplings (nTGCs) from dimension-8 SMEFT operators in a manner compatible with electroweak symmetry breaking, matching the broken-phase operators for ZZV* (V=Z,γ) vertices. It then analyzes ZZ production (with visible and invisible fermionic decays) at future e+e- colliders including CEPC, FCC-ee, ILC and CLIC, using machine learning on angular distributions of final-state fermions to suppress SM backgrounds. The work identifies a dimension-8 operator contributing only to the pure ZZZ* coupling, studies correlations between ZZZ* and ZZγ* probes, incorporates beam polarization, and reports that ML improves sensitivity, allowing nTGC new physics scales to be probed up to the multi-TeV range.

Significance. If the central results hold after validation, the work would be significant as a timely phenomenological study of dimension-8 operators at proposed lepton colliders. The symmetry-consistent form-factor formulation and explicit identification of a pure ZZZ* operator provide a useful bridge between SMEFT and collider observables. The emphasis on angular distributions in 4-body final states and the multi-TeV reach claims, if substantiated, would strengthen motivation for high-energy e+e- programs and illustrate the potential of ML techniques in complex kinematics.

major comments (2)
  1. [Machine learning analysis and results] The central claim that machine learning is advantageous for handling 4-body final states from ZZ decays and significantly improves sensitivity reaches (abstract and results discussion) rests on the assertion of effective background suppression via angular distributions, but the manuscript provides no side-by-side comparison of the same Monte Carlo samples under the neural network classifier versus an optimized cut-based analysis or BDT using identical input variables (angles, energies, missing pT). Without this differential test, the reported improvement and multi-TeV reaches for both ZZZ* and ZZγ* operators could be an artifact of suboptimal traditional methods rather than an intrinsic ML advantage. This directly affects the headline sensitivity claims.
  2. [Formulation of nTGC form factors] The formulation of nTGC form factors for ZZV* that are compatible with SU(2)⊗U(1) breaking and consistently match the dimension-8 operators in the broken phase (abstract and formulation section) is load-bearing for all subsequent sensitivity projections; explicit matching equations or a table relating form-factor parameters to the operator coefficients should be provided to allow verification that no additional assumptions are introduced.
minor comments (2)
  1. [Results figures] The presentation of sensitivity contours for different colliders and polarization settings would benefit from more uniform axis scaling and explicit labeling of which curves correspond to visible versus invisible channels.
  2. [Notation and definitions] Notation for the nTGC form factors (e.g., definitions of f_{Zγ} and f_{ZZ}) should be cross-referenced to the operator basis in a single equation or table for clarity.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading of our manuscript and the constructive comments. We address each of the major comments below and outline the revisions we will make to strengthen the paper.

read point-by-point responses
  1. Referee: [Machine learning analysis and results] The central claim that machine learning is advantageous for handling 4-body final states from ZZ decays and significantly improves sensitivity reaches (abstract and results discussion) rests on the assertion of effective background suppression via angular distributions, but the manuscript provides no side-by-side comparison of the same Monte Carlo samples under the neural network classifier versus an optimized cut-based analysis or BDT using identical input variables (angles, energies, missing pT). Without this differential test, the reported improvement and multi-TeV reaches for both ZZZ* and ZZγ* operators could be an artifact of suboptimal traditional methods rather than an intrinsic ML advantage. This directly affects the headline sensitivity claims.

    Authors: We acknowledge that providing a direct comparison with traditional methods would better substantiate the advantages of the machine learning approach. In the revised version of the manuscript, we will add a section or subsection that presents a side-by-side comparison of the neural network classifier against an optimized cut-based analysis and a boosted decision tree (BDT), using the same Monte Carlo samples and identical input variables including angles, energies, and missing transverse momentum. This will allow readers to assess the improvement quantitatively and confirm that the reported sensitivity reaches are not due to suboptimal traditional analyses. revision: yes

  2. Referee: [Formulation of nTGC form factors] The formulation of nTGC form factors for ZZV* that are compatible with SU(2)⊗U(1) breaking and consistently match the dimension-8 operators in the broken phase (abstract and formulation section) is load-bearing for all subsequent sensitivity projections; explicit matching equations or a table relating form-factor parameters to the operator coefficients should be provided to allow verification that no additional assumptions are introduced.

    Authors: We agree that explicit matching is essential for transparency and verifiability. Although the formulation section describes the compatibility with electroweak symmetry breaking and the matching to dimension-8 operators, we will enhance it by adding a dedicated table that explicitly relates each nTGC form-factor parameter to the corresponding SMEFT operator coefficients. This will include the matching equations for the ZZV* vertices and confirm consistency in the broken phase without additional assumptions. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation relies on independent simulations and operator matching

full rationale

The paper defines nTGC form factors to match dimension-8 SMEFT operators under electroweak symmetry breaking, then computes sensitivities via Monte Carlo simulations of ZZ production at proposed colliders, applying machine learning to angular distributions for background suppression. No quoted step reduces a prediction to a fitted input by construction, nor does any central claim rest on a self-citation chain that itself assumes the target result. The ML improvement is presented as an empirical outcome from handling 4-body kinematics, not a tautological renaming or self-referential fit. External collider parameters and SMEFT operator bases supply independent content, keeping the chain self-contained.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claims rest on the SMEFT framework at dimension-8, electroweak symmetry breaking consistency, and standard collider simulation assumptions; no new particles or forces are postulated.

free parameters (1)
  • nTGC operator coefficients / form factor strengths
    These are the parameters whose values or bounds are determined from the ZZ production data; they are not fixed by prior literature.
axioms (1)
  • domain assumption Form factors must be compatible with spontaneous breaking of SU(2)⊗U(1) and match dimension-8 operators in the broken phase
    Invoked in the initial formulation step to ensure consistency with electroweak symmetry.

pith-pipeline@v0.9.0 · 5881 in / 1500 out tokens · 43299 ms · 2026-05-19T07:41:36.779696+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

25 extracted references · 25 canonical work pages · 11 internal anchors

  1. [1]

    Beyond Standard Model: From Theory to Experiment

    For reviews, John Ellis, [arXiv:2105.14942 [hep-ph]], in the conference proceedings of “Beyond Standard Model: From Theory to Experiment” (BSM-2021), Zewail City, Egypt, March29-31, 2021; I. Brivio and M. Trott, Phys. Rept. 793 (2019) 1 [arXiv:1706.08945 [hep-ph]]; and references therein

  2. [2]

    Probing CP-Violating Neutral Triple Gauge Couplings at Electron-Positron Colliders

    J. Ellis, H.-J. He, and R.-Q. Xiao, Science China (2025), in Press [arXiv:2504.13135 [hep-ph]]

  3. [3]

    Ellis, H.-J

    J. Ellis, H.-J. He, and R.-Q. Xiao, Phys. Rev. D108 (2023) L111704, no.11 [arXiv:2308.16887 [hep-ph]]

  4. [4]

    Ellis, H

    J. Ellis, H. J. He, R.-Q. Xiao, Phys. Rev. D 107 (2023) 035005, no.3 [arXiv:2206.11676 [hep-ph]]

  5. [5]

    Liu, R.-Q

    D. Liu, R.-Q. Xiao, S. Li, J. Ellis, H.-J. He, and R.Yuan, Front. Phys. 20 (2025) 015201, no.1 [arXiv:2404.15937 [hep-ph]]

  6. [6]

    Ellis, H.-J

    J. Ellis, H.-J. He, R.-Q. Xiao, Science China (Phys. Mech. Astron.) 64 (2021) 221062, no.2, [arXiv:2008.04298 [hep-ph]]

  7. [7]

    Ellis, S.F

    J. Ellis, S.F. Ge, H.-J. He, R.-Q. Xiao, Chin. Phys. C 44 (2020) 063106, no.6, [arXiv:1902.06631 [hep-ph]]

  8. [8]

    Ellis, H.-J

    J. Ellis, H.-J. He, R.-Q. Xiao, S.-P. Zeng, and J. Zheng, Phys. Rev. D 111 (2025) 015007, no.1, [arXiv:2408.12508 [hep-ph]]

  9. [9]

    Cepedello, F

    See also, R. Cepedello, F. Esser, M. Hirsch, and V. Sanz, JHEP 07 (2024) 275, [JHEP 12 (2024) 084 (erratum)], [arXiv:2402.04306 [hep-ph]]

  10. [10]

    G. J. Gounaris, J. Layssac, and F. M. Renard, Phys. Rev. D 61 (2000) 073013 [arXiv:hep-ph/ 9910395 [hep-ph]]; Phys. Rev. D 62 (2000) 073013 [arXiv:hep-ph/0003143 [hep-ph]]

  11. [11]

    A basis of dimension-eight operators for anomalous neutral triple gauge boson interactions

    C. Degrande, JHEP 1402 (2014) 101 [arXiv:1308.6323 [hep-ph]]

  12. [12]

    Dimension-Six Terms in the Standard Model Lagrangian

    W. Buchmuller and D. Wyler, Nucl. Phys. B268 (1986) 621; B. Grzadkowski, M. Iskrzynski, M. Misiak, and J. Rosiek, JHEP 10 (2010) 085 [arXiv:1008.4884 [hep-ph]]; and references therein

  13. [13]

    Complete Higgs Sector Constraints on Dimension-6 Operators

    E.g., J. Ellis, V. Sanz, and T. You, JHEP 1407 (2014) 036 [Xiv:1404.3667 [hep-ph]] and JHEP 1503 (2015) 157 [arXiv:1410.7703 [hep-ph]]; H. J. He, J. Ren, and W. Yao, Phys. Rev. D 93 (2016) 015003 [arXiv:1506.03302 [hep-ph]]; J. Ellis and T. You, JHEP 03 (2016) 089 [arXiv:1510.04561 [hep-ph]; S. F. Ge, H. J. He, and R. Q. Xiao, JHEP 10 (2016) 007 [arXiv:16...

  14. [14]

    Measurement of the Z gamma to nu nu-bar gamma production cross section in pp collisions at sqrt(s) = 8 TeV and limits on anomalous Z-Z-gamma and Z-gamma-gamma trilinear gauge boson couplings

    V. Khachatryanet al., [CMS Collaboration], “Measurement of theZγ → ν¯νγ production cross section in pp collisions at √s = 8TeV and limits on anomalousZZγ and Zγγ trilinear gauge boson couplings”, Phys. Lett. B 760 (2016) 448-468 [arXiv: 1602.07152 [hep-ex]]

  15. [15]

    Measurement of the $Z\gamma\rightarrow\nu\bar{\nu}\gamma$ production cross section in $pp$ collisions at $\sqrt{s}$ = 13 TeV with the ATLAS detector and limits on anomalous triple gauge-boson couplings

    M. Aaboudet al.,[ATLAS Collaboration], “Measurement of theZ → ν¯ν production cross section in pp collisions at √s =13TeV with the ATLAS detector and limits on anomalous triple gauge- boson couplings”, JHEP 12 (2018) 010 [arXiv:1810.04995 [hep-ex]]

  16. [16]

    Measurements of Zγ differential cross sections and search for neutral triple gauge couplings inpp collisions at √s =13 TeVwith the ATLAS detector

    ATLAS Collaboration, “Measurements of Zγ differential cross sections and search for neutral triple gauge couplings inpp collisions at √s =13 TeVwith the ATLAS detector”, ATLAS-CONF- 2025-001, April, 2025

  17. [17]

    Jahedi, JHEP (2023) 31 [arXiv:2305.11266 [hep-ph]]; S

    E.g., S. Jahedi, JHEP (2023) 31 [arXiv:2305.11266 [hep-ph]]; S. Spor, E. Gurkanli and M. Köksal, arXiv:2302.08245 [hep-ph]; S. Jahedi and J. Lahiri, JHEP 04 (2023) 085 [arXiv:2212.05121 [hep- ph]]; S. Spor, Nucl. Phys. B991 (2023) 116198 [arXiv:2207.11585 [hep-ph]]; A. Senol, S. Spor, E. Gurkanli, V. Cetinkaya, H. Denizli, and M. Köksal, Eur. Phys. J. Plu...

  18. [18]

    Guo, C.J

    Y.C. Guo, C.J. Pan, M.Q. Ruan and J.C. Yang, [arXiv:2411.04804 [hep-ph]]

  19. [19]

    Scales of Fermion Mass Generation and Electroweak Symmetry Breaking

    D.A. Dicus and H.-J. He, Phys. Rev. D 71 (2005) 093009 [arXiv:hep-ph/0409131 [hep-ph]]; Phys. Rev. Lett. 94 (2005) 221802 [arXiv:hep-ph/0502178 [hep-ph]]

  20. [20]

    Jacob and G

    M. Jacob and G. C. Wick, Ann. Phys. (N.Y.) 7 (1959) 404, no.7

  21. [21]

    Global Analysis for Probing Electroweak Symmetry Breaking Mechanism at High Energy Colliders

    For a comprehensive review, H. J. He, Y.P.Kuang and C.P.Yuan, [arXiv:hep-ph/9704276 [hep- ph]] and DESY-97-056, in the proceedings of the workshop on “Physics at the TeV Energy Scale”, 35 vol.72, p.119, 1996. See also, H. J. He and W.B. Kilgore, Phys. Rev. D 55 (1997) 1515; H. J. He, Y.P. Kuang, C.P. Yuan, Phys. Rev. D 51 (1995) 6463; Phys. Rev. D 55 (199...

  22. [22]

    Englert and R

    F. Englert and R. Brout, Phys. Rev. Lett. 13 (1964) 321; P. W. Higgs, Phys. Rev. Lett. 13 (1964) 508; Phys. Lett. 12 (1964) 132; G. S. Guralnik, C. R. Hagen and T. Kibble, Phys. Rev. Lett. 13 (1964) 585; T. Kibble, Phys. Rev. 155 (1967) 1554

  23. [23]

    The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulations

    J. Alwall, R. Frederix, S. Frixione, V. Hirschi, F. Maltoni, O. Mattelaer, H. S. Shao, T. Stelzer, P. Torrielli and M. Zaro, JHEP07, 079 (2014) doi:10.1007/JHEP07(2014)079 [arXiv:1405.0301 [hep-ph]]

  24. [24]

    Plehn, A

    T. Plehn, A. Butter, B. Dillon, T. Heimel, C. Krause and R. Winterhalder, [arXiv:2211.01421 [hep-ph]]

  25. [25]

    The role of positron polarization for the inital $250$ GeV stage of the International Linear Collider

    K. Fujii, C. Grojean, and M. E. Peskinet al.,[arXiv:1801.02840 [hep-ph]]. 36