Corrections to the conformal hyperscaling relation beyond the ladder Schwinger-Dyson approximation, and reliable extraction of the infrared-fixed-point anomalous dimension from lattice data

The finite-size hyperscaling relation, $M_H = L^{-1} f(L\,m_0^{1/(1+\gamma_m^*)})$, is the standard tool for extracting the mass anomalous dimension $\gamma_m^*$ at the infrared fixed point (IRFP) of near-conformal gauge theories from lattice spectra. A 2012 ladder Schwinger-Dyson (SD) study by the LatKMI collaboration showed, within that model, that the explicit fermion mass generates a leading correction of relative order $(m_P/\Lambda)^{2-2\gamma_m^*}$ to the hyperscaling relation, that fitting the pure hyperscaling form then returns an effective anomalous dimension well below the true fixed-point value (e.g. $\gamma_m^{\rm eff}\approx 0.5$-$0.6$ where the model has $\gamma_m^*\approx 0.8$), that finite-volume corrections are comparatively negligible, and that a mass-corrected fit recovers the model's true $\gamma_m^*$. The open problem is to establish this correction structure beyond the ladder approximation with its simplified running coupling and linearization, and to determine data-analysis strategies, validated on known-answer synthetic data and applied to real simulations, that recover the true IRFP properties ($\gamma_m^*$ and the conformal-versus-chirally-broken classification) from noisy finite-size lattice data. Later work applied the SD-inspired corrected fit to twelve-flavor data, exposed its failure modes in the walking eight-flavor theory, and documented method-dependent spreads in $\gamma^*$ determinations, but has not closed this core.

Background

Asymptotically free SU(3) gauge theory coupled to $N_f$ massless fermions in the fundamental representation is expected to change character as $N_f$ grows: below a critical flavor number the theory confines with spontaneous chiral symmetry breaking (a mass gap forms, spontaneously breaking the approximate symmetry between left- and right-handed fermions), while above it the two-loop beta function (the quantity governing how the gauge coupling changes with the energy scale, computed here to second order in perturbation theory) develops an infrared fixed point (IRFP, a zero of the running coupling at which the long-distance theory becomes scale invariant); the range of $N_f$ with such a fixed point is called the conformal window (for SU(3) with fundamental fermions, two-loop running places it within $9\le N_f\le 16$). This structure underlies walking-technicolor and composite-Higgs models (scenarios for physics beyond the Standard Model in which the Higgs sector arises from a new strongly interacting gauge theory), in which a large mass anomalous dimension $\gamma_m^*$ at (or near) the IRFP is phenomenologically desirable; here $\gamma_m^*$ governs how a fermion mass grows under renormalization toward the infrared.

The mass anomalous dimension can be probed by deforming the massless theory with a small explicit fermion mass $m_0$. Renormalization-group arguments (Del Debbio and Zwicky, 2010) imply that in the mass-deformed conformal theory any low-energy hadronic mass $M_H$ obeys the hyperscaling relation $M_H \sim m_0^{1/(1+\gamma_m^*)}$, and in a finite space-time box of linear size $L$ the finite-size hyperscaling relation $M_H = L^{-1} f(x)$ with scaling variable $x = L\, m_0^{1/(1+\gamma_m^*)}$, where $f$ is a scaling function. Lattice practice measures spectra for many combinations of $L$ and $m_0$ and fits these forms to extract $\gamma_m^*$ and to judge whether the theory is infrared conformal.

Two concerns motivated the source paper. First, the probe mass itself breaks infrared conformality, so hyperscaling holds only up to corrections whose form at the masses actually simulated was unknown. Second, in a chirally broken but near-conformal (walking) theory an approximate hyperscaling can appear as a remnant of approximate scale invariance, so the analysis can misclassify the theory in either direction.

Aoki et al. (LatKMI Collaboration, the lattice gauge-theory group of the Kobayashi-Maskawa Institute at Nagoya University; 2012) studied both questions in a controlled dynamical model: the ladder Schwinger-Dyson (SD) equation, i.e. the gap equation for the fermion propagator closed by replacing the full gauge-boson propagator and fermion-gauge-boson vertex by tree-level objects dressed with a running coupling (the improved ladder approximation), simplified further by a step-function running coupling frozen at its IRFP value below the ultraviolet scale $\Lambda$ (the scale at which infrared conformality ends in this model, used as the unit for all dimensionful quantities), a simplified momentum argument for that coupling, and a linearization of the resulting differential equation. The analytical solution expresses the bare mass $m_0$ in terms of the fermion pole mass $m_P$ (the mass at which the fermion propagator has a pole) as a leading hyperscaling term plus a correction of relative order $(m_P/\Lambda)^{2-2\gamma_m^*}$. In finite volume (periodic boundary conditions, $L\Lambda = 12$-$30$, data restricted to pole masses above the minimum box momentum, $m_P^2 > (2\pi/L)^2$) they found: (a) the mass correction can be sizable at the masses used in typical lattice studies, so that fitting with the pure hyperscaling form returns an effective anomalous dimension $\gamma_m^{\rm eff}$ substantially below the true $\gamma_m^*$; (b) finite-volume ($1/L$) corrections are comparatively negligible in the studied range; (c) in the chirally broken phase close to the critical point, the finite-size hyperscaling relation remains approximately operative; and (d) a modified fit function including the mass-correction term,

$$m_0(L, m_P) = \left[A\, m_P\left(1 + B\, m_P^{2-2\gamma}\right)^{1/(1+\gamma)} + \frac{C}{L}\right]^{1+\gamma},$$

with $A, B, C, \gamma$ free, recovers the true $\gamma_m^*$ on SD-generated data ($\gamma = 0.794$ versus the model value $\gamma_m^*\simeq 0.8$ for $N_f=12$), whereas fixing $B = 0$ returns the biased value $\gamma \approx 0.52$.

The paper itself flags what remains unestablished: all of these results were obtained within the ladder approximation with its specific simplifications; the correction power $2-2\gamma$ is stated to be specific to the ladder SD analysis; the studied volume and mass ranges were limited; which practical analysis strategy best recovers the true IRFP properties from noisy lattice data was left for systematic study (the SD equation, where the answer is known, being the suggested testbed); and extensions to other gauge groups and fermion representations were noted as straightforward in principle within the ladder framework but were not carried out. Treatments beyond this approximation form a broader class that includes non-ladder Schwinger-Dyson treatments, functional renormalization group methods (continuum approaches that evolve an effective action from ultraviolet to infrared scales), and effective-field-theory arguments; none of these lines has yet decided whether the ladder correction structure is a generic property of near-conformal dynamics or a model artifact.

Problem Statement

Establish quantitative reliability for hyperscaling-based extraction of infrared-fixed-point properties in mass-deformed SU(3) many-flavor gauge theories near the lower edge of the conformal window, building on the ladder Schwinger-Dyson analysis of Aoki et al. (2012). Three linked questions constitute the open core.

(1) Correction structure beyond the ladder model. Determine the structure of the finite-mass and finite-volume corrections to the (finite-size) hyperscaling relation in a dynamical treatment that goes beyond the improved ladder approximation with its step-function running coupling, simplified coupling argument, and linearized equation; in particular, whether the leading relative mass correction of order $(m_P/\Lambda)^{2-2\gamma_m^*}$ and the associated modified fit form are robust properties of near-conformal dynamics or artifacts specific to the ladder model.

(2) Confrontation with lattice data. Establish whether the mass-corrected finite-size hyperscaling fit, taken in any of these alternative forms (with the SD-predicted correction exponent, with the exponent left free, or with whatever generalization question (1) yields), recovers the true infrared-fixed-point mass anomalous dimension $\gamma_m^*$ and the correct conformal-versus-chirally-broken classification when applied to actual lattice simulation data of theories in or near the conformal window (for example SU(3) with $N_f=12$), over parameter ranges beyond the source study (for example larger $L\Lambda$, smaller fermion masses, and boundary conditions other than periodic).

(3) Validated analysis methodology. Determine which data-analysis strategies (choice of fit function, treatment of correction terms, fit-range selection, use of multiple observables) recover the true infrared-fixed-point properties within quantified uncertainties from noisy finite-size lattice data, validating them systematically against synthetic data with known answers, such as SD-generated or otherwise controlled ensembles, before application to real data; because this strategy comparison evaluates the mass-corrected fit of question (2) on the same lattice data, a complete answer to this question also discharges question (2).

The common objective uniting these deliverables is that a hyperscaling-based determination of $\gamma_m^*$ and of the phase classification should be demonstrably unbiased with controlled systematics; the theory side (correction structure) and the practice side (validated analysis) are the two halves of that objective, as framed by the source paper itself, whose SD model is the known-answer testbed and whose modified fit is the proposed tool. Extensions to other gauge groups and fermion representations, which within the ladder framework differ only through the group-theory factor multiplying the coupling in the SD equation, follow the same scheme and are peripheral.

References

1. Yasumichi Aoki, Tatsumi Aoyama, Masafumi Kurachi, Toshihide Maskawa, Kei-ichi Nagai, Hiroshi Ohki, Akihiro Shibata, Koichi Yamawaki, Takeshi Yamazaki (LatKMI Collaboration), 'Study of the conformal hyperscaling relation through the Schwinger-Dyson equation', Phys. Rev. D 85, 074502 (2012), DOI: 10.1103/PhysRevD.85.074502, arXiv:1201.4157, https://arxiv.org/abs/1201.4157
2. Luigi Del Debbio, Roman Zwicky, 'Hyperscaling relations in mass-deformed conformal gauge theories', Phys. Rev. D 82, 014502 (2010), DOI: 10.1103/PhysRevD.82.014502, arXiv:1005.2371, https://arxiv.org/abs/1005.2371
3. Yasumichi Aoki, Tatsumi Aoyama, Masafumi Kurachi, Toshihide Maskawa, Kei-ichi Nagai, Hiroshi Ohki, Akihiro Shibata, Koichi Yamawaki, Takeshi Yamazaki (LatKMI Collaboration), 'Lattice study of conformality in twelve-flavor QCD', Phys. Rev. D 86, 054506 (2012), DOI: 10.1103/PhysRevD.86.054506, arXiv:1207.3060, https://arxiv.org/abs/1207.3060
4. Yasumichi Aoki, Tatsumi Aoyama, Ed Bennett, Masafumi Kurachi, Toshihide Maskawa, Kohtaroh Miura, Kei-ichi Nagai, Hiroshi Ohki, Enrico Rinaldi, Akihiro Shibata, Koichi Yamawaki, Takeshi Yamazaki (LatKMI Collaboration), 'Light flavor-singlet scalars and walking signals in $N_f=8$ QCD on the lattice', Phys. Rev. D 96, 014508 (2017), DOI: 10.1103/PhysRevD.96.014508, arXiv:1610.07011, https://arxiv.org/abs/1610.07011
5. Kohtaroh Miura, Kei-ichi Nagai, Akihiro Shibata, 'Schwinger-Dyson Study for Walking/Conformal Dynamics with IR Cutoffs', in 'Origin of Mass and Strong Coupling Gauge Theories (SCGT15)', World Scientific, pp. 355-358 (2017), DOI: 10.1142/9789813231467_0052, arXiv:1510.03603, https://arxiv.org/abs/1510.03603
6. Etsuko Itou, Akio Tomiya, 'Determination of the mass anomalous dimension for $N_f=12$ and $N_f=9$ SU(3) gauge theories', proceedings of the 32nd International Symposium on Lattice Field Theory (Lattice 2014), Proc. Sci. PoS(LATTICE2014)252 (2015), DOI: 10.22323/1.214.0252, https://doi.org/10.22323/1.214.0252, arXiv:1411.1155, https://arxiv.org/abs/1411.1155
7. M. P. Lombardo, K. Miura, T. J. Nunes da Silva, E. Pallante, 'On the particle spectrum and the conformal window', JHEP 2014, 183 (2014), DOI: 10.1007/JHEP12(2014)183, https://doi.org/10.1007/JHEP12(2014)183
8. Maarten Golterman, Yigal Shamir, 'The large-mass regime of the dilaton-pion low-energy effective theory', Phys. Rev. D 98, 056025 (2018), DOI: 10.1103/PhysRevD.98.056025, arXiv:1805.00198, https://arxiv.org/abs/1805.00198
9. Maarten Golterman, Yigal Shamir, 'Fits of $SU(3)$ $N_f=8$ data to dilaton-pion effective field theory', arXiv:1910.10331 (2019), https://arxiv.org/abs/1910.10331
10. Yasumichi Aoki, Tatsumi Aoyama, Ed Bennett, Toshihide Maskawa, Kohtaroh Miura, Hiroshi Ohki, Enrico Rinaldi, Akihiro Shibata, Koichi Yamawaki, Takeshi Yamazaki (LatKMI Collaboration), 'A novel view of the flavor-singlet spectrum from multi-flavor QCD on the lattice', Phys. Rev. D 112, 114503 (2025), DOI: 10.1103/vnml-g6nx, arXiv:2505.08658, https://arxiv.org/abs/2505.08658
11. Florian Goertz, Álvaro Pastor-Gutiérrez, Jan M. Pawlowski, 'Gauge-Fermion Cartography: from confinement and chiral symmetry breaking to conformality', Phys. Rev. D 112, 034029 (2025), DOI: 10.1103/7dzj-k6k8, https://doi.org/10.1103/7dzj-k6k8, arXiv:2412.12254, https://arxiv.org/abs/2412.12254