Pith. sign in

REVIEW 3 major objections 5 minor 81 references

Many-body spectral transitions through the lens of the variable-range SYK2 model

T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The many-body spectral form factor of the variable-range SYK2 model is essentially unchanged for interaction-range powers $\alpha<1/2$, then sharply reorganises near $\alpha\simeq 1/2$ into a higher dip and a secondary plateau that track…

desk verdict A genuine analytic extension of the SYK2 spectral form factor to variable-range couplings, with a caveat about the unproven uniform-saddle assumption that needs referee scrutiny. read the letter →

arxiv 2412.14280 v2 pith:PTTGG6E6 submitted 2024-12-18 cond-mat.stat-mech cond-mat.dis-nncond-mat.str-elhep-thquant-ph

Andrea Legramandi , Soumik Bandyopadhyay , Philipp Hauke This is my paper · ORCID
classification cond-mat.stat-mechcond-mat.dis-nncond-mat.str-elhep-thquant-ph MSC 82B4481Q5015B52 PACS 05.30.-d05.45.Mt72.15.Rn
keywords SYK2modelvariable-rangeinteractionsspectralformfactorpower-lawrandombandedmatrixergodic-non-ergodictransitionmany-bodylevelstatisticsAndersonpath-integralsaddlepoint
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper argues that the many-body spectral statistics of the quadratic SYK model remain remarkably robust when the random all-to-all hopping is replaced by a power-law decay $r^{-\alpha}$: for $0 \le \alpha < 1/2$, the spectral form factor is essentially the same as for $\alpha=0$, and the paper derives an analytic formula, Eq. (15), that includes the subleading corrections. At $\alpha = 1/2$ the perturbative expansion around the spatially uniform saddle point breaks down, and for larger $\alpha$ the SFF develops a higher dip and a secondary plateau. These features track the known single-particle transitions of the underlying power-law random banded matrix: the ergodic-to-non-ergodic transition near $\alpha=0.5$, Anderson criticality near $\alpha=1$, and the onset of integrability beyond $\alpha=1.5$. A sympathetic reader would care because the result identifies which spectral markers survive in realistic, finite-range implementations of SYK-type quantum matter and gives a many-body diagnostic of single-particle localization transitions.

What carries the argument

The central object is the path-integral representation of the spectral form factor in terms of site-resolved collective fields $\Sigma^{ab}_i(t,t')$, introduced through local two-point functions $G^{ab}_i=\psi^a_i\psi^b_i$, rather than the spatially averaged field used in the standard SYK treatment. Because the coupling matrix $A_{ij}=a(i-j)^2/N^2$ is circulant, the calculation diagonalises in Fourier space, and its eigenvalues $\lambda_k$, given by Eqs. (12)-(14), control the fluctuation structure. Zero modes of the quadratic fluctuation kernel occur exactly when $\lambda_k=1$; the uniform mode always has $\lambda_0=1$, and the exponential ramp arises from the resulting SU(2)/U(1) coset volume. The remaining eigenvalues vanish in the thermodynamic limit for $\alpha<1/2$ and become order one for $\alpha>1/2$, which is the mechanism that keeps the uniform saddle stable below the threshold and breaks perturbation theory above it. The analytic SFF Eq. (15) combines the classical action, the time-symmetric zero-mode contribution, and the soft-mode corrections.

What would settle it

A direct check would be to fix a system size around $N\approx 200\text{--}400$, compute the first perturbative correction $C$ in Eq. (S57) at $\alpha=0.45$ and $\alpha=0.55$, and see whether $C$ jumps from $O(1/N)$ to $O(N)$ exactly at $1/2$; a jump already at $\alpha=0.45$ would falsify the claim. Alternatively, measure the dip shift $D=(1/N)\log[g(T_{\rm dip},\alpha)/g(T_{\rm dip},0)]$ at $\alpha=0.45$: Eq. (15) predicts it stays near zero for all $\alpha<1/2$, so a measurable rise below $\alpha=0.5$ would break the uniform-saddle assumption.

Watch

Extended reading notes

Core claim

The paper's central claim is that the variable-range SYK2 Hamiltonian with hopping amplitude $a(i-j)=\min(|i-j|,N-|i-j|)^{-\alpha}$ has a many-body spectral form factor that, for $\alpha<1/2$, is given by Eq. (15): the classical slope is unchanged, the exponential ramp keeps its $\alpha=0$ coefficient, and the spatial profile enters only through a subleading sum over eigenvalues $\lambda_k$ of the circulant matrix $A$. The same eigenvalue spectrum changes character at $\alpha=1/2$: for $\alpha>1/2$ infinitely many $\lambda_k$ become order one, producing an infinite set of would-be zero modes of the fluctuation kernel, and the first perturbative correction to the saddle action then scales as $O(N)$ instead of $O(1/N)$, so perturbation theory around the uniform saddle fails. The numerical SFF shows the predicted robustness below $\alpha=1/2$ and, above it, a rising dip and a secondary plateau whose height grows with $\alpha$, merges with the late-time plateau near $\alpha\simeq 3/2$, and shows a concavity change at $\alpha\simeq 1.02$. The paper reads these as the many-body imprints of the single-particle ergodic-to-non-ergodic transition at $\alpha\simeq 1/2$, Anderson criticality at $\alpha\simeq 1$, and integrability for $\alpha\gtrsim 3/2$.

Load-bearing premise

The derivation assumes that the spatially uniform saddle point of Eq. (7) is the dominant saddle for every $\alpha<1/2$; if a non-uniform, site-dependent saddle already wins inside that window, the robust-formula conclusion would fail.

Editorial extensions

If this is right

  • For every $0\le \alpha<1/2$, the spectral form factor follows the single analytic formula Eq. (15), so chaos diagnostics such as the slope and the exponential ramp keep their $\alpha=0$ values with only subleading corrections.
  • The dip height $D(\alpha)=(1/N)\log[g(T_{\rm dip},\alpha)/g(T_{\rm dip},0)]$ rises sharply near $\alpha\simeq 0.5$, with a finite-size extrapolation giving $\alpha=0.49$, providing a practical marker of the ergodic-to-non-ergodic transition.
  • A secondary plateau develops for $1/2\lesssim\alpha\lesssim 3/2$; its height grows with $\alpha$ and merges with the late-time plateau near $\alpha\simeq 3/2$, signalling a prethermalization regime in which the Hilbert space is not fully explored.
  • The plateau-height curve changes concavity at $\alpha\simeq 1.02$, matching the Anderson critical point of the single-particle power-law random banded matrix.
  • For $\alpha\gtrsim 3/2$ the exponential ramp disappears, consistent with the onset of integrability, although the numerics cannot resolve a sharp transition there because the SFF fluctuations grow with $N T$ and limit the attainable statistics.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the uniform saddle actually gives way to a non-uniform one for $\alpha>1/2$, the same $\lambda_k=1$ zero-mode criterion could help locate that new saddle explicitly; finding its form would produce an analytic SFF for the non-ergodic regime, which the paper leaves open.
  • The structural similarity between the matrix $A$ here and the adjacency matrix used for sparse SYK suggests that the eigenvalue criterion $\lambda_k=1$ may serve as a common diagnostics for connectivity-driven transitions; testing whether sparsification triggers the same dip-height and secondary-plateau markers would extend this logic to random-graph ensembles.
  • The interpretation of the secondary plateau as prethermalization tied to nearly conserved charges could be tested dynamically by constructing approximate local integrals of motion from the hopping matrix in the localized regime; if they exist, their support sizes should control the plateau height.
  • The dip height and plateau-height probes could be transferred to out-of-time-order correlators or two-time correlation functions to ask whether the $\alpha\simeq 1/2$ and $\alpha\simeq 1$ thresholds appear in the dynamics itself, not only in spectral statistics.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper studies the spectral form factor (SFF) of a quadratic SYK model with power-law distance-dependent couplings, the variable-range SYK2 model of Eq. (1). The authors derive a path-integral representation for the SFF, identify the saddle-point solutions, and compute the one-loop determinant including zero modes associated with the eigenvalues of the coupling-envelope matrix A. Their central analytic result, Eq. (15), states that for 0 ≤ α < 1/2 the SFF is essentially unchanged from the α = 0 case, with subleading eigenvalue-dependent corrections. For α > 1/2 they argue that perturbation theory around the uniform saddle breaks down, and they numerically observe a higher dip and a secondary plateau, which they connect to the single-particle PRBM ergodic-to-non-ergodic transition at α ≈ 1/2 and to Anderson criticality near α ≈ 1. The supplemental material contains the detailed derivations of the saddle-point structure, the eigenvalue spectrum of A, the one-loop and soft-mode contributions, and a perturbative breakdown argument in Section G.

Significance. If the central claim holds, the paper provides a useful bridge between single-particle PRBM physics and many-body spectral statistics: it gives an analytic SFF formula with explicit finite-N eigenvalue corrections, and it identifies observable SFF markers—dip height and secondary plateau—that track known PRBM transitions. The derivation is largely self-contained and not fitted to data: Eq. (15) follows from the path integral with eigenvalues of A computed from the model, and the PRBM phase diagram is used as an external benchmark rather than as an input. The numerical comparison at α = 0.3, N = 200 is excellent, and the paper is notably honest about statistical limitations, including the large sample-to-sample fluctuations of the quadratic SYK SFF. The significance is, however, conditional on the uniform-saddle assumption, which is not fully justified and underpins the sharp α = 1/2 boundary.

major comments (3)
  1. [Saddle point equation, after Eq. (6); footnote [74]; Supplemental Eq. (S14)] The central robustness claim—that Eq. (15) is valid for all 0 ≤ α < 1/2—rests on the assumption that the dominant saddle is site-independent. This assumption is not established. In the Fourier-transformed saddle equation (S14), the non-uniform component Σ̂_{−k} is proportional to λ_k, and Eq. (14) shows that for every fixed k the eigenvalue λ_k is O(1) for 0 < α < 1/2, not O(1/N); only modes with k ∝ N vanish, per Eq. (13). A finite number of O(1) Fourier modes is sufficient to build a non-uniform saddle with action of order N, so the statement in footnote [74] that 'most eigenvalues vanish' does not force the solution to be uniform. Section G tests only time-translation-invariant perturbations around the uniform saddle and establishes breakdown for α > 1/2; it does not rule out a competing non-uniform saddle for α < 1/2. The numerical agreement at α = 0.3 (Fig. 1b) is encouraging but is a single point. The sharp boundary at α = 1/2 and the 'robustness' conclusion therefore require either a stability analysis of the uniform saddle against non-uniform perturbations or a substantially softened claim.
  2. [SFF and localization, Fig. 3] The numerical markers of the transitions are based on fitting procedures whose robustness is not quantified. The estimate α ≈ 0.49 is obtained from a linear fit to D(α) for α ≥ 0.6 and its intersection with D = 0, and the estimate α ≈ 1.02 comes from a fourth-order polynomial fit above α = 0.8 with a concavity change. No error bars, fit-range dependence, or finite-size scaling analysis is provided. Since these values are used to associate SFF features with PRBM transitions, the authors should either provide such robustness checks or explicitly label these estimates as heuristic diagnostics rather than precise transition points.
  3. [Quadratic fluctuations and Eq. (11)] The statement after Eq. (11) that for α > 1/2 'an infinite number of zero modes should appear in the thermodynamic limit, each of them bringing a factor proportional to the coset volume' is not derived in detail. The t.t.i. one-loop expression in Eq. (S34) contains factors (1 − λ_k)^{-1} that become singular as λ_k → 1, but the paper does not analyze whether the resulting contribution is really a product of coset volumes, or whether a resummation is required. The separate perturbative argument in Section G gives a cleaner breakdown signal, but the intermediate 'infinite zero modes' claim should be clarified or removed.
minor comments (5)
  1. [Eq. (15) and Fig. 1b caption] There are small typos: 'J1 is a is a Bessel function' should read 'J1 is a Bessel function', and 'forα = 0.3' in the Fig. 1b caption needs a space.
  2. [Footnote [74]] The sentence 'since most of the vanishes in the thermodynamic limit' is grammatically incomplete; it should read 'since most of the eigenvalues vanish in the thermodynamic limit'. More importantly, the logical argument should be revised in light of Major Comment 1.
  3. [Fig. 3b and Eq. (16)] The definition of h in Eq. (16) averages over the time interval JT ∈ [10, 15], and the collapse in Fig. 3b is obtained by applying a constant shift to h. The choice of interval and the shift procedure should be stated explicitly in the main text, together with an indication of the sensitivity of the collapsed curves to these choices.
  4. [Supplemental Section D] In Eq. (S32), the zero-mode volume is quoted as 4πN(1 − x_n^2/4). The derivation of the radius of the vacuum manifold would benefit from a few more intermediate steps, especially the factors of sqrt(N), so that the reader can verify the normalization against the α = 0 limit.
  5. [Supplemental Section H and Fig. S1] The discussion of insufficient sampling producing a systematic downward shift of the SFF is valuable and should be mentioned briefly in the main text, since it affects the interpretation of the plateau-height data in Fig. 3b.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: Eq. (15) is derived from the path integral with the eigenvalues of A; PRBM is used as an external benchmark, not as an input; the uniform-saddle assumption is a validity caveat, not an input-output loop.

full rationale

The derivation chain is self-contained: the SFF path integral is reduced to a saddle-point expansion, the eigenvalues of the model's circulant matrix A are computed from Eq. (12)-(14), and the analytic SFF Eq. (15) follows from the classical action, zero-mode volume, and one-loop fluctuations. No parameter is fitted to the SFF data before the formula is produced; the numerical comparison in Fig. 1(b) is a genuine test. The PRBM phase diagram is invoked as an external benchmark for interpreting the numerically observed dip and secondary-plateau markers (e.g., α≈0.49 and α≈1.02), but it is not inserted into Eq. (15) or used to derive the α=1/2 boundary, which instead arises from the kink in the eigenvalues of A and the perturbative estimate in Section G. The uniform-saddle ansatz stated after Eq. (6) is a genuine assumption, and the eigenvalue-based justification in footnote [74] is questionable because a finite number of O(1) Fourier modes is not ruled out by the vanishing of most eigenvalues; however, that is a correctness or robustness concern, not circularity, since the assumption is not defined in terms of, nor fitted to, the quantity it is used to predict. The self-references to the supplemental material are normal derivational citations rather than appeals to an external uniqueness theorem, and they do not make the central claim reduce to its own inputs.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central derivation introduces no free parameters: the analytic SFF Eq. (15) is parameter-free given the model. The fitted numbers appear only in the numerical identification of transition markers (alpha approximately 0.49, alpha approximately 1.02) and in the descriptive constant shifts. The axioms are the standard large-N path-integral toolkit, the uniform-saddle ansatz, the perturbative control assumption, and the external PRBM benchmark. No new entities are invented.

free parameters (3)
  • Linear-fit intercept for dip height D(alpha) = approximately 0.49
    Used to locate the ergodic-to-non-ergodic transition from the numerical dip height data in Fig. 3a; it is a descriptive fit, not a model parameter.
  • Fourth-order polynomial fit coefficients for collapsed plateau height h(alpha) = concavity change at alpha approximately 1.02
    Used to identify the Anderson critical point in Fig. 3b; the polynomial coefficients are fitted to the shifted numerical h(alpha) data.
  • Per-N constant shifts for plateau height collapse
    Applied to h(alpha) in Fig. 3b to collapse curves onto a universal shape; the procedure is described qualitatively and the shift values are not listed.
assumptions (5)
  • standard math The saddle-point path-integral formalism for the SFF (including soft-mode regulator and zero-mode volume) is valid as in the alpha = 0 SYK2 case (Ref. [63]).
    Used to derive Eq. (5), the ramp, and Eq. (15); the regulator in Eq. (S41)-(S42) follows Ref. [63].
  • domain assumption The spatially uniform saddle point (Sigma_i^ab = Sigma^ab) dominates the path integral for 0 <= alpha < 1/2.
    Assumed in the paragraph after Eq. (6) and in Section B of the Supplemental Material; supported only by a qualitative eigenvalue argument and by matching numerics, not by a global stability proof.
  • domain assumption The perturbative series around the uniform saddle is controlled by the first non-trivial term; the O(N) estimate for alpha > 1/2 in Section G is not canceled by higher-order terms.
    The paper computes only the leading correction C and infers breakdown for alpha > 1/2; no resummation or Borel analysis is attempted.
  • standard math The PRBM phase diagram (ergodic for alpha < 1/2, Anderson critical at alpha = 1, integrable for alpha > 3/2) from Mirlin et al. [40] is correct.
    Used as the external benchmark to interpret the SFF markers.
  • domain assumption The fluctuations of the SYK2 SFF scale as (N/T)^T, referenced as [76] 'To Appear' by one of the authors.
    Used to justify the large sample counts and the shift-correction procedure; the result is not yet published elsewhere.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Many-body spectral transitions through the lens of the variable-range SYK2 model." pith.science (2026). https://pith.science/paper/PTTGG6E6

@misc{pith2026241214280,
  author       = {Pith},
  title        = {Pith review of: Many-body spectral transitions through the lens of the variable-range SYK2 model},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PTTGG6E6}},
  note         = {Machine review of arXiv:2412.14280}
}
read the original abstract

The Sachdev-Ye-Kitaev (SYK) model is a cornerstone in the study of quantum chaos and holographic quantum matter. Real-world implementations, however, deviate from the idealized all-to-all connectivity, raising questions about the robustness of its chaotic properties. In this work, we investigate a quadratic SYK model with distance-dependent interactions governed by a power-law decay. By analytically and numerically studying the spectral form factor (SFF), we uncover how transitions present in the single-particle limit carry over to the many-body system. Non-trivial cancellations in the one-loop contributions lead to a robustness of the SFF under a considerable reduction of the interaction range. Further suppression leads to a breakdown of perturbation theory around the infinite-range path-integral saddle and the appearance of new spectral regimes, marked by a higher dip and the emergence of a secondary plateau. Our results highlight the interplay between single-particle criticality and many-body dynamics, offering new insights into the quantum chaos-to-localization transition and its reflection in spectral statistics.

Figures

Figures reproduced from arXiv: 2412.14280 by the authors.

Figure 1
Figure 1. FIG. 1: (a) Illustration of the model in Eq. ( [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2: Eigenvalues [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3: (a) Dip position as a function of [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

81 extracted references · 55 canonical work pages

  1. [74]

    See supplemental material

  2. [1]

    A simple model of quantum holography,

    A. Kitaev, “A simple model of quantum holography,” Talks given at “Entanglement in Strongly-Correlated Quantum Matter,” (Part 1, Part 2), KITP (2015)

    2015

  3. [2]

    Sachdev and J

    S. Sachdev and J. Ye, Phys. Rev. Lett. 70, 3339 (1993)

    work page 1993

  4. [3]

    Sachdev, Phys

    S. Sachdev, Phys. Rev. X 5, 041025 (2015). 6

    work page 2015

  5. [4]

    Jensen, Phys

    K. Jensen, Phys. Rev. Lett. 117, 111601 (2016)

    work page 2016

  6. [5]

    S. A. Hartnoll, A. Lucas, and S. Sachdev, Holographic Quantum Matter (The MIT Press, 2018)

    work page 2018

  7. [6]

    Chowdhury, A

    D. Chowdhury, A. Georges, O. Parcollet, and S. Sachdev, Rev. Mod. Phys. 94, 035004 (2022)

    2022

  8. [7]

    Maldacena and D

    J. Maldacena and D. Stanford, Phys. Rev. D 94, 106002 (2016)

    work page 2016

Show all 81 references
  1. [8]

    Danshita, M

    I. Danshita, M. Hanada, and M. Tezuka, Prog. Theor. Exp. Phys. 2017, 083I01 (2017)

    2017

  2. [9]

    D. I. Pikulin and M. Franz, Phys. Rev. X 7, 031006 (2017)

    2017

  3. [10]

    Garc ´ ıa-´Alvarez, I

    L. Garc ´ ıa-´Alvarez, I. L. Egusquiza, L. Lamata, A. del Campo, J. Sonner, and E. Solano, Phys. Rev. Lett. 119, 040501 (2017)

    2017

  4. [11]

    A. Chew, A. Essin, and J. Alicea, Phys. Rev. B 96, 121119 (2017)

    2017

  5. [12]

    Babbush, D

    R. Babbush, D. W. Berry, and H. Neven, Phys. Rev. A 99, 040301 (2019)

    2019

  6. [13]

    A. Chen, R. Ilan, F. de Juan, D. I. Pikulin, and M. Franz, Phys. Rev. Lett. 121, 036403 (2018)

    2018

  7. [14]

    Luo, Y.-Z

    Z. Luo, Y.-Z. You, J. Li, C.-M. Jian, D. Lu, C. Xu, B. Zeng, and R. Laflamme, Npj Quantum Inf. 5, 53 (2019)

    2019

  8. [15]

    Wei and T

    C. Wei and T. A. Sedrakyan, Phys. Rev. A 103, 013323 (2021)

    2021

  9. [16]

    Uhrich, S

    P. Uhrich, S. Bandyopadhyay, N. Sauerwein, J. Sonner, J.-P. Brantut, and P. Hauke, arXiv:2303.11343 [quant- ph] (2023), 10.48550/arXiv.2303.11343

    2023 doi arXiv

  10. [17]

    Baumgartner, P

    R. Baumgartner, P. Pelliconi, S. Bandyopadhyay, F. Orsi, N. Sauerwein, P. Hauke, J.-P. Brantut, and J. Sonner, (2024), arXiv:2411.17802 [quant-ph]

    2024

  11. [18]

    J. W. Britton, B. C. Sawyer, A. C. Keith, C.-C. J. Wang, J. K. Freericks, H. Uys, M. J. Biercuk, and J. J. Bollinger, Nature 484, 489 (2012)

    2012

  12. [19]

    Jurcevic, B

    P. Jurcevic, B. P. Lanyon, P. Hauke, C. Hempel, P. Zoller, R. Blatt, and C. F. Roos, Nature 511, 202 (2014)

    2014

  13. [20]

    Smith, A

    J. Smith, A. Lee, P. Richerme, B. Neyenhuis, P. W. Hess, P. Hauke, M. Heyl, D. A. Huse, and C. Monroe, Nat. Phys. 12, 907 (2016)

    2016

  14. [21]

    Trautmann and P

    N. Trautmann and P. Hauke, Phys. Rev. A 97, 023606 (2018)

    2018

  15. [22]

    Chomaz, I

    L. Chomaz, I. Ferrier-Barbut, F. Ferlaino, B. Laburthe- Tolra, B. L. Lev, and T. Pfau, Rep. Prog. Phys. 86, 026401 (2022)

    2022

  16. [23]

    Browaeys and T

    A. Browaeys and T. Lahaye, Nat. Phys. 16, 132 (2020)

    2020

  17. [24]

    S. L. Cornish, M. R. Tarbutt, and K. R. A. Hazzard, Nature Physics 20, 730 (2024)

    2024

  18. [25]

    D. D. Awschalom, R. Hanson, J. Wrachtrup, and B. B. Zhou, Nature Photonics 12, 516 (2018)

    2018

  19. [26]

    Chen and T

    X. Chen and T. Zhou, Phys. Rev. B 100, 064305 (2019)

    2019

  20. [27]

    A. L. Burin, Phys. Rev. B 91, 094202 (2015)

    2015

  21. [28]

    A. L. Burin, Phys. Rev. B 92, 104428 (2015)

    2015

  22. [29]

    Hauke and M

    P. Hauke and M. Heyl, Phys. Rev. B 92, 134204 (2015)

    2015

  23. [30]

    K. S. Tikhonov and A. D. Mirlin, Phys. Rev. B 97, 214205 (2018)

    2018

  24. [31]

    X. Deng, V. E. Kravtsov, G. V. Shlyapnikov, and L. San- tos, Phys. Rev. Lett. 120, 110602 (2018)

    2018

  25. [32]

    S. J. Thomson and M. Schir´ o, Phys. Rev. Res.2, 043368 (2020)

    2020

  26. [33]

    X. Deng, S. Ray, S. Sinha, G. V. Shlyapnikov, and L. Santos, Phys. Rev. Lett. 123, 025301 (2019)

    2019

  27. [34]

    Nag and A

    S. Nag and A. Garg, Phys. Rev. B 99, 224203 (2019)

    2019

  28. [35]

    Roy and D

    S. Roy and D. E. Logan, SciPost Phys. 7, 042 (2019)

    2019

  29. [36]

    X. Deng, G. Masella, G. Pupillo, and L. Santos, Phys. Rev. Lett. 125, 010401 (2020)

    2020

  30. [37]

    Modak and T

    R. Modak and T. Nag, Phys. Rev. Res. 2, 012074 (2020)

    2020

  31. [38]

    Modak and T

    R. Modak and T. Nag, Phys. Rev. E 101, 052108 (2020)

    2020

  32. [39]

    Cheng, Phys

    C. Cheng, Phys. Rev. B 108, 155113 (2023)

    2023

  33. [40]

    A. D. Mirlin, Y. V. Fyodorov, F.-M. Dittes, J. Quezada, and T. H. Seligman, Phys. Rev. E 54, 3221 (1996)

    1996

  34. [41]

    HTOkw3a7YZNF9AB+33rWQxsQ/t0=

    and multifractality [42, 43], as it is amenable to an- alytical studies using the supersymmetric sigma-model approach [44]. Here, we take a different route: we study the many-body spectral statistics of the SYK model us- ing the spectral form factor (SFF) [45]. In the thermo- ...

    2024 arXiv

  35. [42]

    P. W. Anderson, Phys. Rev. 109, 1492 (1958)

    1958

  36. [43]

    JANSSEN, Int

    M. JANSSEN, Int. J. Mod. Phys. B 08, 943–984 (1994)

    1994

  37. [44]

    Evers and A

    F. Evers and A. D. Mirlin, Rev. Mod. Phys. 80, 1355 (2008)

    2008

  38. [45]

    Y. V. Fyodorov and A. D. Mirlin, Phys. Rev. Lett. 67, 2405 (1991)

    1991

  39. [46]

    Br´ ezin and S

    E. Br´ ezin and S. Hikami, Phys. Rev. E55, 4067 (1997)

    1997

  40. [47]

    P. Saad, S. H. Shenker, and D. Stanford, (2018), arXiv:1806.06840 [hep-th]

    2018 arXiv

  41. [48]

    P. Kos, M. Ljubotina, and T. Prosen, Phys. Rev. X 8, 021062 (2018), arXiv:1712.02665 [nlin.CD]

    2018 arXiv

  42. [49]

    Gharibyan, M

    H. Gharibyan, M. Hanada, S. H. Shenker, and M. Tezuka, J. High Energ. Phys. 2018, 124 (2018)

    2018

  43. [50]

    A. Chan, A. De Luca, and J. T. Chalker, Phys. Rev. Lett. 122, 220601 (2019)

    2019

  44. [51]

    ˇSuntajs, J

    J. ˇSuntajs, J. Bonˇ ca, T. Prosen, and L. Vidmar, Phys. Rev. E 102, 062144 (2020), arXiv:1905.06345 [cond- mat.str-el]

    2020 arXiv

  45. [52]

    Shackleton, and S

    Nivedita, H. Shackleton, and S. Sachdev, Phys. Rev. E 101, 042136 (2020)

    2020

  46. [53]

    Prakash and A

    R. Prakash and A. Lakshminarayan, Phys. Rev. B 101, 121108 (2020)

    2020

  47. [54]

    Prakash, J

    A. Prakash, J. H. Pixley, and M. Kulkarni, Phys. Rev. Res. 3, L012019 (2021)

    2021

  48. [55]

    ˇSuntajs, T

    J. ˇSuntajs, T. Prosen, and L. Vidmar, Annals of Physics 435, 168469 (2021), special issue on Philip W. Anderson

    2021

  49. [56]

    D. K. Nandy, T. ˇCadeˇ z, B. Dietz, A. Andreanov, and D. Rosa, Phys. Rev. B 106, 245147 (2022)

    2022

  50. [57]

    Anegawa, N

    T. Anegawa, N. Iizuka, A. Mukherjee, S. K. Sake, and S. P. Trivedi, JHEP 11, 234 (2023), arXiv:2305.07505 [hep-th]

    2023 arXiv

  51. [58]

    Barney, M

    R. Barney, M. Winer, C. L. Baldwin, B. Swingle, and V. Galitski, SciPost Phys. 15, 084 (2023)

    2023

  52. [59]

    Hopjan and L

    M. Hopjan and L. Vidmar, Phys. Rev. Res. 5, 043301 (2023)

    2023

  53. [60]

    R. L. Baumgartner, L. V. Delacr´ etaz, P. Nayak, and J. Sonner, (2024), arXiv:2405.19260 [cond-mat.stat- mech]

    2024 arXiv

  54. [61]

    Orman, H

    P. Orman, H. Gharibyan, and J. Preskill, (2024), arXiv:2403.13884 [hep-th]

    2024 arXiv

  55. [62]

    Buijsman, Phys

    W. Buijsman, Phys. Rev. B 109, 024205 (2024)

    2024

  56. [63]

    H. Dong, P. Zhang, C. B. Dag, Y. Gao, N. Wang, J. Deng, X. Zhang, J. Chen, S. Xu, K. Wang, Y. Wu, C. Zhang, F. Jin, X. Zhu, A. Zhang, Y. Zou, Z. Tan, Z. Cui, Z. Zhu, F. Shen, T. Li, J. Zhong, Z. Bao, H. Li, Z. Wang, Q. Guo, C. Song, F. Liu, A. Chan, L. Ying, and H. Wang, (2024...

    2024 arXiv

  57. [64]

    Winer, S.-K

    M. Winer, S.-K. Jian, and B. Swingle, Phys. Rev. Lett. 125, 250602 (2020)

    2020

  58. [65]

    Y. Liao, A. Vikram, and V. Galitski, Phys. Rev. Lett. 125, 250601 (2020)

    2020

  59. [66]

    Xu and W

    W.-F. Xu and W. J. Rao, Sci. Rep. 13, 634 (2023)

    2023

  60. [67]

    V. E. Kravtsov and K. A. Muttalib, Phys. Rev. Lett. 79, 1913 (1997)

    1997

  61. [68]

    Evers and A

    F. Evers and A. D. Mirlin, Phys. Rev. Lett. 84, 3690 (2000)

    2000

  62. [69]

    Varga and D

    I. Varga and D. Braun, Phys. Rev. B 61, R11859 (2000). 7

    2000

  63. [70]

    A. D. Mirlin and F. Evers, Phys. Rev. B 62, 7920 (2000)

    2000

  64. [71]

    Bogomolny and M

    E. Bogomolny and M. Sieber, Phys. Rev. E 98, 042116 (2018)

    2018

  65. [72]

    P. A. Nosov, I. M. Khaymovich, and V. E. Kravtsov, Phys. Rev. B 99, 104203 (2019)

    2019

  66. [73]

    S. Xu, L. Susskind, Y. Su, and B. Swingle, arXiv:2008.02303 [cond-mat.str-el] (2020), 10.48550/arXiv.2008.02303

    2020 doi arXiv

  67. [75]

    In [73] we discuss further motivation for this assumption coming from an analysis of A eigenvalues: since most of the vanishes in the thermodynamic limit, they lead to a vanishing Fourier components of Σ, forcing the solution to be uniform

  68. [76]

    Since we are interested in the N → ∞limit, we preferred using an intuitive but less precise notation that does not distinguish between even and odd N

  69. [77]

    Legramandi and T

    A. Legramandi and T. Neil, To Appear

  70. [78]

    Serbyn, Z

    M. Serbyn, Z. Papi´ c, and D. A. Abanin, Phys. Rev. Lett. 111, 127201 (2013)

    2013

  71. [79]

    D. A. Huse, R. Nandkishore, and V. Oganesyan, Phys. Rev. B 90, 174202 (2014)

    2014

  72. [80]

    This is a consequence of the fact that v1 is an eigenvalue also for A−1 with eigenvector λ−1 1 = N −2

  73. [81]

    Kamenev and M

    A. Kamenev and M. M´ ezard, Phys. Rev. B 60, 3944 (1999). SUPPLEMENT AL MA TERIAL In this supplemental material, we provide detailed explanations and derivations to complement the calculations presented in the main text. In Section A, we show some intermediate steps in the der...

    1999

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.