Quenched Eguchi-Kawai volume independence: control of $1/N$ corrections and exact planar results for the reduced model

Parisi's 1982 paper "A simple expression for planar field theories" (Phys. Lett. B 112, 463), developed simultaneously with the work of Bhanot, Heller and Neuberger, introduced quenching to rescue large-N reduction: in the Quenched Eguchi-Kawai (QEK) model, the eigenvalue phases of the one-site link matrices are treated as quenched random variables, and quenched expectation values of gauge-invariant observables are argued to be volume independent in the $N\to\infty$ limit. The claim was established only at leading order in $1/N$, by planar, perturbative and free-field arguments, and two issues remain open beyond it: the control of $1/N$ corrections, including nonperturbative contributions that could spoil the equivalence non-uniformly in $N$, and the construction of exact planar solutions of the reduced model by analytic matrix-model methods in the spirit of Brezin, Itzykson, Parisi and Zuber's exact solution of the planar one-matrix model. The literature shows: twisted and adjoint-fermion reductions now provide practical large-N volume independence on the lattice, and an all-order planar perturbative proof of equivalence among quenched models on finite tori was given in 2022, but the in-principle validity of quenching at weak coupling and beyond leading order in $1/N$ remains contested and unresolved (Bringoltz and Sharpe argued breakdown in 2008; Neuberger rebutted their argument in 2020 and stated that the question requires future extensive numerical work), the weak-coupling boundary conditions of the lattice loop equations remain unknown, and no exact analytic planar solution of the four-dimensional reduced model exists.

Background

In the 't Hooft large-$N$ (planar) limit, a gauge theory with gauge group $SU(N)$ or $U(N)$ is studied as $N\to\infty$ with the 't Hooft coupling $\lambda=g^2N$ held fixed. Feynman diagrams organize by topology and only planar diagrams (genus zero) survive at $N=\infty$, giving the expectation values of single-trace operators - in particular Wilson loops, the traces of parallel transporters of the gauge field around closed contours - a systematic expansion in $1/N$. Eguchi and Kawai discovered in 1982 that in this limit the theory on a large (or infinite) lattice is equivalent to a reduced model defined on a single lattice site, whose only variables are the $N\times N$ link matrices (the parallel transporters along the links of the lattice) in each direction. The proof goes through the loop equations, the exact Schwinger-Dyson hierarchy for Wilson loop expectation values: on an infinite lattice and on the one-site reduced lattice the hierarchy coincides, provided that translation invariance and the center symmetry of the reduced model are not spontaneously broken. Center symmetry here is the invariance under gauge transformations that act independently in each lattice direction, multiplying the link matrices in direction $\mu$ by elements of the center $Z_N$ of the gauge group; in its unbroken phase the $N$ eigenvalue phases of each link matrix effectively behave as continuous lattice momenta on a torus, which is how a one-site model can imitate an infinite volume. The equivalence fails at finite $N$: Bhanot, Heller and Neuberger showed by weak-coupling analysis and Monte Carlo simulation that the original Eguchi-Kawai model spontaneously breaks the required symmetry, invalidating the reduction precisely in the weak-coupling regime relevant for the continuum limit. Parisi's paper, the source of this problem, proposed the fix called quenching: the eigenvalue phases of the link matrices - the emergent momenta - are removed from the dynamical (annealed) variables and treated as fixed external random variables, observables are computed at fixed phases, and only afterwards are they averaged over the phases with the measure induced by the change of variables (a Jacobian of Vandermonde type in the eigenvalue phases); this mirrors the quenched averaging over frozen disorder in spin-glass physics. The paper argues that such quenched expectation values of $U(N)$-invariant objects are volume independent in the $N\to\infty$ limit. Bhanot, Heller and Neuberger simultaneously developed this Quenched Eguchi-Kawai (QEK) model, supplied the planar-diagram argument for its equivalence at leading order in $1/N$, and studied it numerically. Like the original Eguchi-Kawai proof, this is a leading-order-in-$1/N$ statement of perturbative and free-field character: it does not control the $1/N$ corrections, and it is not a nonperturbative construction. Both limitations define open directions beyond the source's leading-order claim: the control of $1/N$ corrections, including possible nonperturbative contributions that could invalidate the equivalence non-uniformly in $N$, and the exact solution of the reduced model by analytic matrix-model methods of the type used by Brezin, Itzykson, Parisi and Zuber (BIPZ), who had solved the planar one-matrix model exactly by saddle-point eigenvalue-density techniques. Two neighboring lines frame the modern status. First, the twisted Eguchi-Kawai (TEK) model of Gonzalez-Arroyo and Okawa implements the reduction with twisted boundary conditions, i.e. twist matrices from the center that modify the periodicity of the fields in the different directions and force the Polyakov loops (traces of holonomies wrapping a compact direction) to vanish at weak coupling; with an appropriate $N$-dependent choice of twist, numerical simulations show volume independence works up to very large $N$. Related prescriptions - adding adjoint (color-neutral) fermions, trace deformation, and continuum reduction on moderately-sized lattices - also achieve practical volume independence, and the general conditions under which large-$N$ volume independence holds in QCD-like theories have been formulated. Second, on the theory side, the all-order planar perturbative equivalence of quenched models on finite tori has been proven, but the finite-$N$ behavior of the original quenching prescription remains unsettled: Bringoltz and Sharpe argued in 2008 that the large-$N$ equivalence in the QEK model likely breaks down, while Neuberger in 2020 rebutted their specific argument and concluded that the in-principle validity of quenching remains an open question requiring extensive numerical work.

Problem Statement

Make Parisi's quenched volume-independence conjecture an exact and controlled statement, for $U(N)$ (or $SU(N)$) lattice gauge theory and its one-site quenched reduction - the QEK model, in which each link matrix is decomposed as $U_\mu = V_\mu e^{i\theta_\mu}V_\mu^\dagger$, the eigenvalue phases $\theta_\mu$ are held fixed as quenched random variables distributed with the induced Jacobian measure, and the eigenvector matrices $V_\mu$ are integrated at fixed $\theta_\mu$. Two linked questions serve one objective, namely to make the one-site quenched model an exact, controlled representation of the planar theory rather than a leading-order heuristic:

(1) Validity and $1/N$ corrections. Determine, with analytic control beyond leading order in $1/N$, whether quenched expectation values of gauge-invariant observables (Wilson loops, free energy) in the QEK model agree with those of the infinite-volume planar theory, and characterize the size and structure of the $1/N$ corrections across the coupling range - in particular at weak coupling, where the continuum limit is taken and where the original Eguchi-Kawai model failed. This includes the nonperturbative questions of whether large-$N$ transitions or spontaneous breaking of center symmetry or of the symmetry under permuting the eigenvalue phases (the emergent momenta) invalidates the equivalence non-uniformly in $N$ (the point on which Bringoltz-Sharpe (2008) and Neuberger (2020) reach opposite conclusions).

(2) Exact planar results. Obtain exact results at $N=\infty$ for the reduced model (an exact planar solution, or a proof of obstruction), going beyond the perturbative and free-field evidence.

An obstruction is a demonstration that no exact planar solution of the reduced model can be obtained by analytic methods, whether because those methods are obstructed for this model or because no such exact planar solution exists. The two components are linked: a validity verdict fixes the phase window in which an exact solution is meaningful, and an exact solution is the strongest possible form of a positive validity verdict. Both components belong to the single scientific objective.

References

1. G. Parisi, 'A simple expression for planar field theories', Physics Letters B 112 (1982) 463-464. DOI: 10.1016/0370-2693(82)90849-8, https://doi.org/10.1016/0370-2693(82)90849-8
2. T. Eguchi, H. Kawai, 'Reduction of Dynamical Degrees of Freedom in the Large N Gauge Theory', Physical Review Letters 48 (1982) 1063. DOI: 10.1103/PhysRevLett.48.1063, https://doi.org/10.1103/PhysRevLett.48.1063
3. G. Bhanot, U. M. Heller, H. Neuberger, 'The Quenched Eguchi-Kawai Model', Physics Letters B 113 (1982) 47. DOI: 10.1016/0370-2693(82)90106-X, https://doi.org/10.1016/0370-2693(82)90106-X
4. G. Bhanot, U. M. Heller, H. Neuberger, 'A Phase Transition in the Quenched Eguchi-Kawai Model', Physics Letters B 115 (1982) 237-238. DOI: 10.1016/0370-2693(82)90652-9, https://doi.org/10.1016/0370-2693(82)90652-9
5. M. Okawa, 'Monte Carlo Study of the Quenched Eguchi-Kawai Model', Physical Review Letters 49 (1982) 705. DOI: 10.1103/PhysRevLett.49.705, https://doi.org/10.1103/PhysRevLett.49.705
6. E. Brézin, C. Itzykson, G. Parisi, J. B. Zuber, 'Planar diagrams', Communications in Mathematical Physics 59 (1978) 35-51. DOI: 10.1007/BF01614153, https://doi.org/10.1007/BF01614153
7. A. Gonzalez-Arroyo, M. Okawa, 'A Twisted Model for Large N Lattice Gauge Theory', Physics Letters B 120 (1983) 174. DOI: 10.1016/0370-2693(83)90647-0, https://doi.org/10.1016/0370-2693(83)90647-0
8. A. Gonzalez-Arroyo, M. Okawa, 'Twisted-Eguchi-Kawai model: A reduced model for large-N lattice gauge theory', Physical Review D 27 (1983) 2397. DOI: 10.1103/PhysRevD.27.2397, https://doi.org/10.1103/PhysRevD.27.2397
9. M. Teper, H. Vairinhos, 'Symmetry breaking in twisted Eguchi-Kawai models', Physics Letters B 652 (2007) 359. DOI: 10.1016/j.physletb.2007.06.037, arXiv:hep-th/0612097, https://arxiv.org/abs/hep-th/0612097
10. T. Azeyanagi, T. Hirata, M. Hanada, T. Ishikawa, 'Phase structure of twisted Eguchi-Kawai model', Journal of High Energy Physics 01 (2008) 025. DOI: 10.1088/1126-6708/2008/01/025, arXiv:0711.1925, https://arxiv.org/abs/0711.1925
11. A. González-Arroyo, M. Okawa, 'Large N reduction with the Twisted Eguchi-Kawai model', Journal of High Energy Physics 07 (2010) 043. DOI: 10.1007/JHEP07(2010)043, arXiv:1005.1981, https://arxiv.org/abs/1005.1981
12. M. Garcia Perez, 'Prospects for large N gauge theories on the lattice', Proceedings of Science LATTICE2019 (2020) 276. DOI: 10.22323/1.363.0276, arXiv:2001.10859, https://arxiv.org/abs/2001.10859
13. B. Bringoltz, S. R. Sharpe, 'Breakdown of large-N quenched reduction in SU(N) lattice gauge theories', Physical Review D 78 (2008) 034507. DOI: 10.1103/PhysRevD.78.034507, arXiv:0805.2146, https://arxiv.org/abs/0805.2146
14. H. Neuberger, 'Quenched Eguchi-Kawai model revisited', Physical Review D 102 (2020) 094503. DOI: 10.1103/PhysRevD.102.094503, arXiv:2009.09539, https://arxiv.org/abs/2009.09539
15. H. Neuberger, 'Planar diagrams for lattice gauge theory on finite tori', arXiv:2206.13988 (2022). arXiv:2206.13988, https://arxiv.org/abs/2206.13988
16. A. González-Arroyo, I. Kanamori, K.-I. Ishikawa, K. Miyahana, M. Okawa, R. Ueno, 'Numerical stochastic perturbation theory applied to the twisted Eguchi-Kawai model', Journal of High Energy Physics 06 (2019) 127. DOI: 10.1007/JHEP06(2019)127, arXiv:1902.09847, https://arxiv.org/abs/1902.09847
17. K.-I. Ishikawa, M. Okawa, H. Takei, 'Perturbative gradient flow coupling of the twisted Eguchi-Kawai model with the numerical stochastic perturbation theory', International Journal of Modern Physics A 40 (2025) 2550013. DOI: 10.1142/S0217751X25500137, arXiv:2410.15668, https://arxiv.org/abs/2410.15668
18. Y. Hamada, T. Misumi, 'Z_N stability and continuity in the twisted Eguchi-Kawai model with two-flavor adjoint fermions', Physical Review D 112 (2025) 094503. DOI: 10.1103/dbg6-9d71, arXiv:2507.20123, https://arxiv.org/abs/2507.20123
19. C. Bonanno, J. L. Dasilva Golán, M. García Pérez, M. D'Elia, A. Giorgieri, 'Scale setting of SU(N) Yang-Mills theory, topology and large-N volume independence', Physical Review D 113 (2026) 074501. DOI: 10.1103/qtqt-hdpz, arXiv:2511.07355, https://arxiv.org/abs/2511.07355
20. M. Hirasawa, A. Matsumoto, J. Nishimura, A. Yosprakob, 'Tensor renormalization group and the volume independence in 2D U(N) and SU(N) gauge theories', Journal of High Energy Physics 12 (2021) 011. DOI: 10.1007/JHEP12(2021)011, arXiv:2110.05800, https://arxiv.org/abs/2110.05800
21. R. Maeta, 'Regularized master-field approximation for large-N reduced matrix models', arXiv:2605.10720 (2026). arXiv:2605.10720, https://arxiv.org/abs/2605.10720
22. P. Kovtun, M. Ünsal, L. G. Yaffe, 'Volume independence in large $N_c$ QCD-like gauge theories', Journal of High Energy Physics 06 (2007) 019. DOI: 10.1088/1126-6708/2007/06/019, arXiv:hep-th/0702021, https://arxiv.org/abs/hep-th/0702021
23. T. Azeyanagi, M. Hanada, M. Ünsal, R. Yacoby, 'Large-N reduction in QCD-like theories with massive adjoint fermions', Physical Review D 82 (2010) 125013. DOI: 10.1103/PhysRevD.82.125013, arXiv:1006.0717, https://arxiv.org/abs/1006.0717
24. T. Sulejmanpasic, 'Global Symmetries, Volume Independence, and Continuity in Quantum Field Theories', Physical Review Letters 118 (2017) 011601. DOI: 10.1103/PhysRevLett.118.011601, arXiv:1610.04009, https://arxiv.org/abs/1610.04009
25. J. Kiskis, R. Narayanan, H. Neuberger, 'Does the crossover from perturbative to nonperturbative physics in QCD become a phase transition at infinite N?', Physics Letters B 574 (2003) 65. DOI: 10.1016/j.physletb.2003.08.070, arXiv:hep-lat/0308033, https://arxiv.org/abs/hep-lat/0308033