The mode-locked (ML) 4-phasor model describes a glassy random laser as complex spin amplitudes under a global spherical constraint, with quenched Gaussian four-body couplings restricted by the frequency matching condition to an intermediate-dilution graph with O(N3)O(N^3) interacting quadruplets out of O(N4)O(N^4), whose deterministic, correlated connectivity lies between the fully connected and sparse regimes where spin-glass theory is controlled. Monte Carlo finite-size scaling (Niedda, Gradenigo, Leuzzi and Parisi, 2023) locates the glassy random-lasing transition and yields effective exponents νeff=2β+γ\nu_{\mathrm{eff}}=2\beta+\gamma of 0.91(8)0.91(8) with free boundary conditions and 1.2(2)1.2(2) with periodic boundary conditions on the frequencies, against the mean-field window 1≤2β+γ≤21\le 2\beta+\gamma\le 2 and the Random Energy Model value 22 - suggesting a mean-field class distinct from the fully connected narrow-band model. But no analytic solution incorporating the mode-locking topology exists, and the asymptotic status of the exponents, the robustness to the model's simplifying assumptions (dissipative limit, extended-mode approximation, independent Gaussian couplings), and the physical fidelity of the boundary-condition trick remain unresolved. The problem asks for a controlled analytic mean-field theory of the model on its mode-locked graph - closed equations for global order parameters, the transition temperature, and critical exponents - and for the thermodynamic-limit universality class of the transition. As of September 2026 the problem is open: the only direct attack (Niedda's 2023 thesis) derives but does not solve the self-consistency equations via a mapping of uncontrolled validity, and the 2026 follow-up by Benedetti and Leuzzi studies a different constraint variant numerically.

Background

Random lasers are lasers in which the optical feedback sustaining amplification comes from multiple scattering in a disordered gain medium rather than from cavity mirrors. In some random lasers the shot-to-shot fluctuations of the emission spectra display glassy behavior: they are organized as predicted by the statistical mechanics of disordered systems, with replica symmetry breaking (RSB) - the Parisi framework in which the equilibrium measure splits into many nearly degenerate pure states, diagnosed experimentally through a nontrivial distribution of the overlap between intensity fluctuations of independent emission shots, and theoretically through the Parisi overlap (the similarity between equilibrium configurations of two independent copies of the same disordered sample). A minimal statistical-mechanics description is given by phasor models: each lasing mode kk is a complex amplitude aka_k, and gain saturation is modeled by a global spherical constraint fixing the total intensity ∑k∣ak∣2\sum_k|a_k|^2. In the mode-locked (ML) 4-phasor model the Hamiltonian is a disordered four-body interaction with couplings Jk1k2k3k4J_{k_1k_2k_3k_4} that are quenched (fixed during the dynamics) independent Gaussian variables, restricted to quadruplets obeying the frequency matching condition (FMC) ∣ωk1−ωk2+ωk3−ωk4∣≲Γ|\omega_{k_1}-\omega_{k_2}+\omega_{k_3}-\omega_{k_4}|\lesssim\Gamma, with Γ\Gamma the typical mode linewidth, which encodes resonant four-wave mixing between modes (the third-order nonlinear-optical coupling of four frequency components). Of the O(N4)O(N^4) quadruplets of NN modes only O(N3)O(N^3) interact: a dense but intermediate dilution whose pattern is deterministic and correlated, the local connectivity depending on the mode frequency index.

In the narrow-band approximation all quadruplets satisfy the FMC, the interaction graph is fully connected, and the model is analytically solved: the RSB solution of the complex spherical 2+42+4 spin glass by Antenucci, Crisanti and Leuzzi (2015) yields the glassy (random-lasing) transition, and the equivalence between the intensity-fluctuation overlap measured in experiments and the Parisi overlap was proved in that limit. Both the fully connected model and a uniformly random O(N)O(N) dilution of it lie in the mean-field universality class of the Random Energy Model (REM), the reference exactly solvable model of disordered systems, with hyperscaling combination 2β+γ=22\beta+\gamma=2 (where β\beta and γ\gamma are the standard order-parameter and susceptibility critical exponents, and the combination is the one that scaling theory ties to the correlation-length exponent and the dimension). Controlled spin-glass solutions, however, exist only for fully connected graphs (replica and full-RSB methods) and for sparse graphs with O(1)O(1) connectivity per variable (cavity methods - message-passing equations for the distribution of local fields on a sparse graph - as in the spherical p-spin computation of Gradenigo, Angelini, Leuzzi and Ricci-Tersenghi, 2020); as the source paper emphasizes, no spin-glass model has been solved exactly outside these two regimes.

Niedda, Gradenigo, Leuzzi and Parisi (SciPost Phys. 14, 144 (2023)) simulated the ML 4-phasor model on its FMC graph with exchange Monte Carlo (a parallel-tempering algorithm swapping configurations at different temperatures to accelerate equilibration) on GPUs. Finite-size scaling (FSS - the standard extraction of critical points and exponents from the size-dependence of thermodynamic peaks) of specific-heat peaks yields thermodynamic-limit critical temperatures Tc(∞)=0.86(3)T_c(\infty)=0.86(3) with free boundary conditions (FBC) on the frequency indices and 0.61(3)0.61(3) with periodic boundary conditions (PBC) - a trick wrapping the frequency indices on a ring that eliminates weakly coupled band-edge modes and mimics the bulk of a larger system. The effective exponent νeff≡2β+γ\nu_{\mathrm{eff}}\equiv 2\beta+\gamma comes out as 0.91(8)0.91(8) (FBC) and 1.2(2)1.2(2) (PBC), to be compared with the general mean-field window 1≤2β+γ≤21\le 2\beta+\gamma\le 2 derived in the paper from Landau free energies with leading nonlinearity ϕn\phi^n, with the REM at the value 22. The transition thus appears compatible with a mean-field class yet distinct from the REM class of the fully connected model - a conclusion limited by very strong finite-size effects: equilibrated sizes reach only about one hundred modes, and equilibration times grow roughly as eaNe^{aN}, as expected for a problem that is NP-complete-like in the number of modes (combinatorially intractable in the worst case).

Problem Statement

Determine, with analytic control, the thermodynamic-limit critical behavior of the mode-locked glassy random laser - the ML 4-phasor model (complex spin amplitudes aka_k under a global spherical constraint ∑k∣ak∣2=N\sum_k|a_k|^2=N, quenched Gaussian four-body couplings restricted by the frequency matching condition) on its FMC-diluted interaction graph. Concretely:

(1) Controlled analytic theory. Construct a mean-field solution of the model that explicitly incorporates the deterministic FMC-induced topology (the intermediate dense dilution - O(N3)O(N^3) interacting quadruplets out of O(N4)O(N^4) - and its correlation structure, with local connectivity depending on the mode frequency index): after averaging over the quenched Gaussian couplings, obtain closed self-consistency equations for global order parameters, solve them, and derive predictions for the glass-transition temperature and the critical exponents. Standard replica/full-RSB machinery (applicable to fully connected limits) and cavity/population methods (applicable to sparse graphs) do not straightforwardly apply; no spin-glass model has been solved exactly outside these two regimes.

(2) Universality-class settlement. Establish whether the numerically estimated exponents - νeff=2β+γ=0.91(8)\nu_{\mathrm{eff}}=2\beta+\gamma=0.91(8) with free boundary conditions and 1.2(2)1.2(2) with periodic boundary conditions on the frequencies, against the mean-field window 1≤2β+γ≤21\le 2\beta+\gamma\le 2 and the REM value 22 - are the true thermodynamic-limit universality-class values, or are shifted by residual finite-size and preasymptotic corrections (including those induced by the PBC trick, and the earlier estimate of about 2/3 from smaller simulations). A controlled theory of (1) predicts these thermodynamic-limit values.

References

  1. Niedda, J., Gradenigo, G., Leuzzi, L., & Parisi, G. (2023). Universality class of the mode-locked glassy random laser. SciPost Phys. 14, 144. DOI: 10.21468/scipostphys.14.6.144. https://doi.org/10.21468/scipostphys.14.6.144
  2. Niedda, J. (2023). Realistic Model for Random Lasers from Spin-Glass Theory (PhD thesis, Sapienza University of Rome). arXiv:2306.08972. https://arxiv.org/abs/2306.08972
  3. Benedetti, M., & Leuzzi, L. (2026). Smoothed-cubic spin-glass model of random lasers. SciPost Phys. 20, 153. DOI: 10.21468/scipostphys.20.6.153. https://doi.org/10.21468/scipostphys.20.6.153
  4. Niedda, J., Leuzzi, L., & Gradenigo, G. (2023). Intensity pseudo-localized phase in the glassy random laser. Journal of Statistical Mechanics: Theory and Experiment 2023, 053302. DOI: 10.1088/1742-5468/acd2c4. https://doi.org/10.1088/1742-5468/acd2c4
  5. Gradenigo, G., Antenucci, F., & Leuzzi, L. (2020). Glassiness and lack of equipartition in random lasers: The common roots of ergodicity breaking in disordered and nonlinear systems. Phys. Rev. Research 2, 023399. DOI: 10.1103/physrevresearch.2.023399. https://doi.org/10.1103/physrevresearch.2.023399
  6. Okuyama, M., & Ohzeki, M. (2025). Free energy equivalence between mean-field models and nonsparsely diluted mean-field models. J. Phys. A: Math. Theor. 58, 045004. DOI: 10.1088/1751-8121/adaab4. https://doi.org/10.1088/1751-8121/adaab4
  7. Antenucci, F., Crisanti, A., & Leuzzi, L. (2015). Complex spherical 2+4 spin glass: A model for nonlinear optics in random media. Phys. Rev. A 91, 053816. DOI: 10.1103/physreva.91.053816. https://doi.org/10.1103/physreva.91.053816
  8. Antenucci, F., Crisanti, A., & Leuzzi, L. (2015). The glassy random laser: replica symmetry breaking in the intensity fluctuations of emission spectra. Scientific Reports 5, 16792. DOI: 10.1038/srep16792. https://doi.org/10.1038/srep16792
  9. Gradenigo, G., Angelini, M. C., Leuzzi, L., & Ricci-Tersenghi, F. (2020). Solving the spherical p-spin model with the cavity method: equivalence with the replica results. Journal of Statistical Mechanics: Theory and Experiment 2020, 113302. DOI: 10.1088/1742-5468/abc4e3. https://doi.org/10.1088/1742-5468/abc4e3
  10. Kroeze, R. M., Marsh, B. P., Atri Schuller, D., Hunt, H. S., Bourzutschky, A. N., Winer, M., Gopalakrishnan, S., Keeling, J., & Lev, B. L. (2025). Directly observing replica symmetry breaking in a vector quantum-optical spin glass. Science 389, 1122-1126. DOI: 10.1126/science.adu7710. https://doi.org/10.1126/science.adu7710
  11. Marsh, B. P., Atri Schuller, D., Ji, Y., Hunt, H. S., Socolof, G. Z., Bowman, D. P., Keeling, J., & Lev, B. L. (2025). Multimode cavity QED Ising spin glass. Phys. Rev. Lett. 135, 160403. DOI: 10.1103/x19r-pzyb. https://doi.org/10.1103/x19r-pzyb
  12. Palacios, G., Raposo, E. P., Amaral, A. M., Acioli, L. H., Falcão-Filho, E. L., de Araújo, C. B., & Siqueira, A. C. A. (2025). Replica symmetry breaking in higher-order optical soliton fission. Chaos, Solitons & Fractals 199, 116742. DOI: 10.1016/j.chaos.2025.116742. https://doi.org/10.1016/j.chaos.2025.116742
  13. Niedda, J., Tonolo, T., & Gradenigo, G. (2024). Probing marginal stability in the spherical p = 2 model. Journal of Statistical Mechanics: Theory and Experiment 2024, 113301. DOI: 10.1088/1742-5468/ad890c. https://doi.org/10.1088/1742-5468/ad890c
  14. Baldovin, M., Gradenigo, G., Vulpiani, A., & Zanghì, N. (2025). On the foundations of statistical mechanics. Physics Reports 1132, 1-79. DOI: 10.1016/j.physrep.2025.05.003. https://doi.org/10.1016/j.physrep.2025.05.003