Hadronic Contributions to the Muon $g-2$ in Improved Holographic QCD Models

1. Problem Statement

The anomalous magnetic moment of the muon, $a_\mu = (g-2)_\mu/2$, is a critical precision observable in particle physics. Recent experimental results from Fermilab (2025) have achieved unprecedented precision (127 ppb), revealing a persistent tension with Standard Model (SM) predictions. The dominant source of theoretical uncertainty lies in the hadronic contributions, specifically:

• Hadronic Vacuum Polarization (HVP): The leading-order contribution.
• Hadronic Light-by-Light (HLbL) scattering: A higher-order contribution.

Current determinations rely on two main approaches:

1. Data-driven methods: Using dispersion relations and $e^+e^- \to \text{hadrons}$ cross-sections. However, recent discrepancies in $e^+e^- \to \pi^+\pi^-$ data have created tensions between different datasets.
2. Lattice QCD: First-principles calculations that are independent of experimental cross-sections but computationally intensive.

Holographic QCD (AdS/QCD) offers a complementary non-perturbative framework. However, standard models (Hard-Wall and original Soft-Wall) suffer from quantitative shortcomings: they fail to reproduce linear Regge trajectories simultaneously with correct decay constants, leading to significant model-dependent uncertainties in $a_\mu$ predictions. This paper addresses the need to evaluate $a_\mu$ within infrared (IR)-improved holographic models that better capture low-energy QCD dynamics.

2. Methodology

The authors perform a systematic study using three distinct infrared-improved Soft-Wall AdS/QCD models (SW1, SW2, and SW3) to calculate both the LO-HVP and the pseudoscalar-pole contribution to HLbL scattering.

• Models Used:

  • SW1: An IR-improved model incorporating a specific dilaton profile and a $z$-dependent quartic coupling to ensure correct UV/IR asymptotic behaviors and chiral symmetry breaking.
  • SW2: An improved model where the bulk scalar vacuum expectation value (VEV) is a genuine solution to the equation of motion (EOM) with a $z$-dependent bulk mass, rather than being imposed by hand.
  • SW3: A generalization of SW2 to the 2+1 flavor case, including the 't Hooft determinant to correctly handle the chiral transition and octet meson spectra.

• Parameter Fixing:
  Model parameters are constrained by fitting the ground-state $\rho$-meson mass ($m_\rho = 775$ MeV), pion mass ($m_\pi = 135$ MeV), and pion decay constant ($f_\pi = 92.4$ MeV) to experimental values.

• Calculations:

  1. LO-HVP: The hadronic vacuum polarization function $\Pi_{\text{had}}^{\text{em}}(Q^2)$ is derived from the vector-current two-point correlator. The contribution to $a_\mu$ is computed via the integral over spacelike momentum $Q^2$.
  2. HLbL (Pseudoscalar Pole): The authors compute the pion transition form factor (TFF), $F_{\pi^0\gamma^*\gamma^*}(Q_1^2, Q_2^2)$, using the Chern-Simons action. This form factor is then inserted into the standard integral for the pseudoscalar-pole contribution to $a_\mu^{\text{HLbL}}$.
  3. Extrapolation: The $\eta$ and $\eta'$ contributions are estimated by rescaling the pion TFF normalization based on experimental ratios.

3. Key Contributions

• Unified Framework: The first unified calculation of both LO-HVP and pseudoscalar-pole HLbL contributions within a single set of IR-improved holographic models.
• Model Comparison: A systematic comparison of three different IR-improvement strategies (SW1, SW2, SW3) to assess the stability of holographic predictions against variations in the infrared sector.
• Diagnosis of Discrepancies: Identification of the specific origin of deviations between holographic predictions and dispersive/lattice results, linking them directly to the reproduction of vector meson decay constants.
• TFF Analysis: Detailed analysis of the pion transition form factor across different momentum transfers, highlighting how low-momentum behavior in holographic models affects the HLbL integral.

4. Key Results

A. Meson Spectra and Decay Constants

• All three models successfully reproduce the ground-state $\rho$ and $\pi$ masses and the pion decay constant $f_\pi$.
• Limitation: Despite improvements, all three models systematically underestimate the $\rho$-meson decay constant ($F_\rho$) compared to the experimental value ($\approx 348$ MeV).
  • SW1 predicts $F_\rho^{1/2} \approx 329$ MeV.
  • SW2 predicts $F_\rho^{1/2} \approx 260$ MeV.
  • SW3 predicts $F_\rho^{1/2} \approx 308$ MeV.

B. Leading-Order HVP Contribution ($a_\mu^{\text{LO-HVP}}$)

• Values (for $N_f=2$):
  • SW1: $482.5 \times 10^{-10}$
  • SW2: $276.4 \times 10^{-10}$
  • SW3: $404.0 \times 10^{-10}$
• Comparison: These values are generally lower than dispersive determinations ($\approx 558 \times 10^{-10}$).
• Correlation: The authors demonstrate a strong correlation between the underestimation of $a_\mu^{\text{LO-HVP}}$ and the underestimation of $F_\rho$.
• Validation: When the SW1 parameters are artificially adjusted to match the experimental $F_\rho$ (while keeping $m_\rho$ fixed), the predicted $a_\mu^{\text{LO-HVP}}$ rises to $573.5 \times 10^{-10}$, aligning well with dispersive estimates. This confirms that the vector sector's decay constant is the primary source of deviation.

C. Hadronic Light-by-Light (Pseudoscalar Pole)

• Pion Pole ($a_\mu^{\pi^0}$):
  • SW1: $55.5 \times 10^{-11}$
  • SW2: $69.2 \times 10^{-11}$
  • SW3: $56.7 \times 10^{-11}$
• Variation: Significant differences exist between models, particularly in the intermediate momentum region ($Q^2$) which dominates the HLbL integral.
  • SW2 yields a notably higher value, exceeding the White Paper estimate ($63.6 \times 10^{-11}$).
  • SW1 and SW3 are closer to Lattice QCD results (BMW and Mainz groups).
• Form Factor Behavior: While all models reproduce the asymptotic behavior ($Q^2 F \to 2f_\pi$) and agree with data at $Q^2=0$, they diverge in the low-to-intermediate $Q^2$ region. This highlights that reproducing the mass spectrum is insufficient; the specific IR realization of the TFF is crucial.

5. Significance and Conclusion

• Model Dependence: The study reveals that while improved soft-wall models offer better phenomenological descriptions of the meson spectrum than original models, they still exhibit significant model dependence in $a_\mu$ predictions.
• Vector Sector Sensitivity: The LO-HVP contribution is highly sensitive to the vector meson decay constant. The discrepancy with data-driven methods is largely attributed to the models' inability to simultaneously fit $m_\rho$ and $F_\rho$ perfectly within the current parametrizations.
• HLbL Sensitivity: The HLbL contribution is sensitive to the detailed shape of the transition form factor at low momentum transfer. The spread in results among SW1, SW2, and SW3 provides an estimate of the intrinsic theoretical uncertainty in holographic approaches.
• Future Outlook: The authors conclude that infrared-improved holographic models are a valuable intermediate framework between data-driven approaches and lattice QCD. However, to reduce uncertainties, future models must incorporate:
  1. More rigorous constraints on short-distance behavior (QCD asymptotics).
  2. Explicit quark-mass effects.
  3. Better descriptions of the vector decay constants.

In summary, the paper establishes that while holographic QCD provides qualitative insights and analytic control over resonance dynamics, quantitative precision for the muon $g-2$ requires resolving the tension between reproducing mass spectra and decay constants simultaneously.