Pith. sign in

REVIEW 2 major objections 4 minor 21 references

Large AdS black hole saddles drop out of the spectral form factor at t of order β, curing the d ≡ 5 (mod 4) divergence.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-01 06:56 UTC pith:TSDZVTMF

load-bearing objection A coherent Picard-Lefschetz rescue of the AdS black-hole saddle in the SFF slope, but the resolution rests entirely on a minisuperspace ansatz the paper itself leaves underived. the 2 major comments →

arxiv 2607.21704 v1 pith:TSDZVTMF submitted 2026-07-23 hep-th

AdS Black Holes Are Short-Lived inside the Spectral Form Factor

Jos\'e L. F. Barb\'on , Eduardo Velasco-Aja This is my paper
classification hep-th
keywords spectral form factorAdS/CFT correspondenceblack hole thermodynamicsPicard-Lefschetz theoryStokes phenomenonanalytic continuationminisuperspaceEuclidean path integral
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper tackles a paradox in the semiclassical evaluation of the spectral form factor (SFF) of holographic CFTs on spheres. Naively continuing the large AdS black hole saddle to complex temperature makes the SFF slope grow exponentially for CFT dimensions d = 5, 9, 13, ... (d ≡ 5 mod 4), contradicting the bound |Z(β+it)| ≤ Z(β). Using a minisuperspace integral over off-shell black hole geometries, the authors show that this saddle loses dominance at t = O(β) and then detaches from the integration contour at t = O(β) for d > 3 (or t = O(ℓ) for d = 3), after which the low-energy endpoint controls the slope and gives power-law decay. The result resolves the apparent blow-up and shows how Stokes phenomena in the path integral enforce consistency with unitarity.

Core claim

The paper's central claim is that the exponential divergence of exp(−I_B) obtained by naively continuing the large AdS black hole action to complex temperature is not part of the correct integration cycle for the spectral form factor. Working with the one-dimensional minisuperspace integral over the horizon radius of conical off-shell black holes, the authors find that the large black hole saddle x_B loses dominance at an angle ϕ_D = π/[2(d−1)] and disconnects from the relevant thimble sum at ϕ_S = π/(d−1), i.e., at t_S = β tan(π/(d−1)) for d > 3 and t_S ≈ 2πℓ for d = 3. After the Stokes transition, the contour consists only of the endpoint thimble, and the SFF slope is dominated by the low-

What carries the argument

The central object is the minisuperspace integral Z(β+it) ≈ ∫_0^∞ dx exp[−(β+it)N_* x^{d−2}(1+x^2) + (4π/(d−1)) N_* x^{d−1}], where x is the horizon radius measured in AdS radius units, N_* is the effective central charge of the CFT, and the integral runs over off-shell black hole geometries with conical singularities at the horizon. Picard–Lefschetz theory decomposes the defining contour [0,∞) into steepest-descent thimbles attached to the endpoint x=0 and to the two saddle points x_b and x_B (small and large black holes). The load-bearing mechanism is a Stokes phenomenon: as t increases, the large black hole thimble first exchanges dominance with the endpoint thimble at t_D = β tan(π/[2(d−

Load-bearing premise

The whole argument depends on the one-dimensional model of the path integral being a faithful stand-in for the full gravitational path integral's saddle structure — a point the paper itself says needs further work.

What would settle it

Compute the minisuperspace integral (23) numerically for d = 5 and β ≪ 1, tracking the thimble decomposition as t increases: if the large black hole saddle does not detach from the contour at t ≈ β, or if |Z(β+it)| grows exponentially for t ≫ ℓ, the Stokes mechanism described here is incorrect. Alternatively, a direct spectral-form-factor calculation in a holographic CFT showing the black hole saddle persisting beyond t ~ β would disprove the claim.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • For d > 3, the large AdS black hole saddle contributes to the spectral form factor slope only for t ≲ β; for larger t the slope is set by the low-energy endpoint and decays as a power law.
  • For d = 3, the black hole saddle stays on the integration contour until t ≈ 2πℓ, but it already loses dominance at t = O(β).
  • The naive exponential divergence of the slope in CFT dimensions d ≡ 5 (mod 4) is removed: the divergent saddle is not on the correct contour, and the semiclassical SFF satisfies |Z(β+it)| ≤ Z(β).
  • The inverse Laplace transform defining the microcanonical density of states is well-defined in this framework; exchanging the order of integration gives Ω(E) = e^{S(E)} dS/dE, making earlier local saddle-point analyses safe.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the same Stokes mechanism persists beyond the minisuperspace truncation, any saddle whose analytically continued action grows exponentially with t should be read as a contour artifact rather than a physical instability — a criterion applicable to other gravitational saddles.
  • The parametrically separated timescales in d = 3 (dominance loss at O(β), disconnection at O(ℓ)) offer a concrete signature that could be searched for in numerical studies of holographic models.
  • Including string-scale effects and the Hagedorn transition would likely shift the quantitative Stokes times, but the qualitative conclusion that low-energy states dominate the late-time slope seems robust.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 4 minor

Summary. The paper considers the analytically continued partition function Z(β+it) in holographic CFTs on S^{d-1}. It first shows that the standard semiclassical saddle evaluation of the large AdS black hole action gives an exponentially growing |Z| for d≡5 (mod 4) at large t, in tension with the unitarity bound (5). To resolve this, the authors introduce a one-dimensional minisuperspace integral over off-shell black-hole geometries (Eqs. (23)–(24) and (50)) and perform a Picard–Lefschetz analysis. For d>3 they find that the large black hole saddle loses dominance at t=O(β) and then topologically disconnects from the original integration contour at t=O(β); for d=3, dominance is lost at t=O(β) while disconnection occurs at t=O(ℓ). After disconnection, the spectral form factor slope is controlled by the low-energy endpoint and decays as a power law. The paper concludes that the apparent exponential divergence is removed by Stokes phenomena within this model.

Significance. The result is potentially significant for the semiclassical understanding of the spectral form factor slope: it identifies a concrete mechanism—Stokes disconnection of the large black hole saddle—that enforces the bound (5) and removes the pathology in d≡5 (mod 4). The analysis is careful and largely self-contained: the saddle locations, thimble decompositions, next-to-leading Stokes shifts (Eqs. (31)–(32)), and the d=3 timescales (Eqs. (35)–(36)) are derived from the explicit action (24) with no fitted parameters. The consistency check against (5) anchors the physical need for a resolution. However, all quantitative conclusions are conditional on the unproven one-dimensional minisuperspace ansatz (50); the paper itself identifies the foundation of this ansatz as future work. Because the central claim concerns AdS black holes in the full gravitational path integral, not just the toy integral, this conditionality is significant.

major comments (2)
  1. [§3, Eq. (50); Conclusions] The central result—that the large AdS black hole saddle disconnects from the integration cycle at t=O(β) and that the d≡5 (mod 4) divergence is thereby excised—is established only for the one-dimensional minisuperspace integral (23)–(24)/(50). The full gravitational path integral contains infinitely many field directions and other saddles/complex metrics; the intersection number of the physical contour with the large-BH thimble need not coincide with this toy model's. The paper states in the Conclusions that the foundation of the starting ansatz (50) is future work. I would ask for either (i) a derivation or at least a plausible argument that (50) captures the relevant saddle topology (e.g., using the Lorentzian prescriptions of [13]), or (ii) an explicit weakening of the claims so that they are only about the model integral. Without this, the title and abstract overstate the conclusions
  2. [§3.1, Figs. 2–3 and Eq. (28)] The Stokes disconnection occurs for all d>3, including even d, where the naive saddle action is bounded and oscillatory and no conflict with (5) exists. This suggests that the disconnection may be a generic property of the single-variable truncation rather than a mechanism specifically tied to the d≡5 (mod 4) pathology. The paper does not comment on this. A discussion of why the relevant physics in even d also calls for the loss of the large BH saddle, or a test that distinguishes the truncation from the full path integral, would materially strengthen the argument.
minor comments (4)
  1. [§3.2, Eqs. (40)–(49)] The endpoint-dominated regime is derived using the graviton-gas entropy with an O(1) parameter s_g. The conclusion that the gas saddle never dominates relies on the parametric separation N_*≫1 and on the approximate form of S_g. While plausible, s_g is not determined, and the transition energies (37)–(38) are order-of-magnitude estimates. This does not affect the black-hole Stokes times, but it does affect the quantitative slope in the endpoint regime. Please state clearly that Eq. (49) is illustrative rather than a precise CFT prediction.
  2. [Figures (i)–(iv) and captions] The figure placement and caption referencing ('Figures (i)–(iv)') interrupt the flow after Section 3.1.1. Please renumber or reposition so that the reader can locate them at the relevant discussion.
  3. [Reference formatting] Reference [11] is formatted inconsistently ('Gross, D., Perry, M. and Yaffe, L.'). Please normalize to journal style.
  4. [Throughout] The notation switches between ℓ=1 and restored factors (e.g., Eqs. (35)–(36)). While not incorrect, a short note at first use would improve readability.

Circularity Check

0 steps flagged

No significant circularity: the Stokes/disconnection times follow from the explicit minisuperspace action; the main gap is the unproven ansatz (50), an assumption rather than a circular reduction.

full rationale

The paper's derivation is internally self-contained and does not fit parameters to the quantities it predicts. The large-t divergence (3) is computed directly from the analytically continued black-hole action (16)-(18). The resolution is studied in the explicit minisuperspace integral (23)-(24), whose action is the same off-shell black-hole mass/entropy function. The dominance and Stokes times are then solved from the saddle-point actions themselves, e.g. eqs. (31)-(32) and (35)-(36), not imposed by hand. The external inequality (5) is used only as motivation that the divergence must be an artifact; it does not by itself determine the thimble topology or the timescales. The only substantive limitation is that the minisuperspace ansatz (50) is not derived from the full gravitational path integral; the paper explicitly flags this in Conclusions: 'the foundation of the starting ansatz (50) within the general theory of gravitational path integrals' is left for future work. That is an assumption, not circular reasoning, since the Stokes conclusions follow within the model rather than being presupposed. Self-citations [10] and [16] are used for ancillary points (local saddle analysis and black-hole/string correspondence) and are not load-bearing for the central disconnection claim. Thus no circular step can be exhibited; the score reflects only minor, non-load-bearing self-citations.

Axiom & Free-Parameter Ledger

2 free parameters · 6 axioms · 0 invented entities

The central derivation rests on the assumed minisuperspace ansatz (Eq. 23/50) rather than a first-principles path integral; the paper is transparent about this. No new forces, particles, or dimensions are introduced. The only hand-chosen constants are the O(1) radiation constant s_g and the IR cutoff x0, neither of which affects the black-hole Stokes times. The quantitative results are therefore predictions of the ansatz, not of full AdS/CFT.

free parameters (2)
  • s_g (radiation constant) = unspecified, O(1)
    Introduced in Eq. (40) to model the graviton-gas entropy S_g = s_g N*^{d/(d+1)} x^{d(d-2)/(d+1)}. It affects the endpoint contribution (49) but not the black-hole Stokes times.
  • x0 (infrared cutoff / energy gap) = x0≈0 in the main analysis; y1=1 in boundary layer
    The lower endpoint of the minisuperspace integral (21) is chosen by hand to represent the low-energy cutoff; the paper sets it to zero in the large-N approximation and uses y1=1 in the boundary-layer model.
axioms (6)
  • domain assumption Semiclassical gravitational path integral: Z(β) ≈ exp(-I(X_β^+)) for the large AdS black hole at high temperature (Eq. 1).
    Standard AdS/CFT saddle-point approximation; invoked from the Introduction.
  • domain assumption Off-shell black hole manifolds with conical singularities have action I(β,r_s)=βM(r_s)-S(r_s) (Eq. 20).
    Argued via cutting the manifold and Gauss-Bonnet [12]; basis of the minisuperspace integral.
  • ad hoc to paper The one-dimensional minisuperspace representation Z(β+it) ≈ ∫_0^∞ dx e^{-I(β+it,x)} (Eq. 23/50) with the given action and measure faithfully captures the relevant integration-cycle topology.
    This is the central model assumption; the paper explicitly defers its foundation to future work (Conclusions).
  • domain assumption At large N*, measure Jacobian, loop corrections, and stringy corrections can be neglected in the exponent; boundaries can be extended (x0≈0).
    Invoked in Section 3 before Eq. (23) and in Section 3.2; drops O(log N*) and higher-derivative terms.
  • domain assumption The CFT density of states is dominated by black holes plus a graviton-gas band with entropy ansatz (40).
    Used in the boundary-layer analysis; includes an unspecified O(1) constant s_g.
  • standard math Picard-Lefschetz theory applies to the analytically continued minisuperspace integral.
    Standard steepest-descent/thimble decomposition; referenced to [19].

pith-pipeline@v1.3.0-alltime-deepseek · 13547 in / 19251 out tokens · 168439 ms · 2026-08-01T06:56:56.144106+00:00 · methodology

0 comments
read the original abstract

We analyze the contribution of AdS black hole saddles to the spectral form factor of $d$-dimensional CFTs with a gravity dual. At high temperatures, the analytically continued gravitational action of the large AdS black hole is expected to give a leading large-$N$ approximation to the 'slope' of the spectral form factor. When the CFT is put on a $(d-1)$-dimensional spatial sphere of radius $\ell$, we show that such an evaluation implies unphysical behavior of $Z(\beta +it)$ at large $t\gg \ell$ for $d\equiv5 \,({\rm mod}\; 4)$. A Picard--Lefschetz analysis of a minisuperspace approximation to $Z(\beta+it) $ does reveal the required Stokes phenomena that restore consistency. It is found that for $d>3$, the large AdS black hole saddle loses dominance and subsequently disconnects from the relevant integration cycle at $t=O(\beta)$. For $d=3$, dominance is lost at a time of order $\beta$, while the Stokes disconnection takes place at $t=O(\ell)$. After the large black hole saddle disconnects from the path-integral contour, the slope of the spectral form factor becomes dominated by the low-energy end of the spectrum.

Figures

Figures reproduced from arXiv: 2607.21704 by Eduardo Velasco-Aja, Jos\'e L. F. Barb\'on.

Figure 1
Figure 1. Figure 1: Plot of the Euclidean action as a function of the horizon radius [PITH_FULL_IMAGE:figures/full_fig_p008_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Contour plot for the real part of −I(β + it , x). The two saddle points xb, xB migrate into the complex plane, going to the imaginary axis as t increases. Steepest descent paths are depicted in blue, while steepest ascents are represented in red. For values of t < t(B) S , of order β in d > 3, and order ℓ in the special d = 3 case, the contour defining the integral for the partition C0, can be decomposed i… view at source ↗
Figure 3
Figure 3. Figure 3: Contour plot for the real part of −I(β + it , x). The two saddle points xb, xB migrate into the complex plane, going to the imaginary axis as t increases. Steepest descent paths are depicted in blue, while steepest ascents are represented in red. For values of t > t(B) S , of the order of β in d > 3 and order ℓ in the special d = 3 case, a Stokes transition takes place. Across the Stokes line, the endpoint… view at source ↗
Figure 4
Figure 4. Figure 4: Imaginary part of the action I(β + it, x), as a function of t/β evaluated at the critical points xb, xB and at the endpoint x0 for d = 5 and β = 0.1. For visualization purposes, the action at xb had to be rescaled. The value t/β = tan(π/(d−1)), corresponding to the degenerate triple point in the β → 0 limit, is also represented. In the box, we present a zoom of how the triple point is resolved by O(β 2/ℓ2 … view at source ↗
Figure 5
Figure 5. Figure 5: Real part of the action I(β +it, x), as a function of t/β evaluated at the critical points xb, xB and at the endpoint x0 for d = 5 and β = 0.1. For visualization purposes, the action at xb had to be rescaled. The value t/β = tan(π/2(d − 1)), corresponding to the degenerate triple point in the β → 0 limit, is also represented. In the box, we present a zoom of how the triple point is resolved by O(β 2/ℓ2 ) c… view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

21 extracted references · 15 linked inside Pith

  1. [1]

    Euclidean Quantum Gravity,

    S. W. Hawking, “Euclidean Quantum Gravity,” inRecent Developments in Gravita- tion.Cargese Lectures, Eds. M. Levy and S. Deser (Plenum 1978), NATO Sci. Ser. B44(1979), 145 PRINT-78-0745 (CAMBRIDGE). 1 “The Path Integral Approach to Quantum Gravity”, inGeneral Relativity: An Ein- stein Centenary Survey. Eds. S.W. Hawking and W. Israel, (Cambridge Universit...

    1978

  2. [2]

    Path Integrals and the In- definiteness of the Gravitational Action,

    G. W. Gibbons, S. W. Hawking and M. J. Perry, “Path Integrals and the In- definiteness of the Gravitational Action,” Nucl. Phys. B138(1978), 141-150 doi:10.1016/0550-3213(78)90161-X 1

    doi:10.1016/0550-3213(78)90161-x 1978

  3. [3]

    Action Integrals and Partition Functions in Quantum Gravity

    G. W. Gibbons and S. W. Hawking, “Action Integrals and Partition Functions in Quantum Gravity. ” Phys. Rev.D15, 2752-2756 (1977). 1

    1977

  4. [4]

    Thermodynamics of Black Holes in anti-De Sitter Space,

    S. W. Hawking and D. N. Page, “Thermodynamics of Black Holes in anti-De Sitter Space,” Commun. Math. Phys.87(1983), 577 doi:10.1007/BF01208266 1

    doi:10.1007/bf01208266 1983

  5. [5]

    Anti de Sitter space and holography,

    E. Witten, “Anti de Sitter space and holography,” Adv. Theor. Math. Phys.2(1998), 253-291 doi:10.4310/ATMP.1998.v2.n2.a2 [arXiv:hep-th/9802150 [hep-th]]. 1

    Pith/arXiv arXiv 1998

  6. [6]

    Anti-de Sitter space, thermal phase transition, and confine- ment in gauge theories,

    E. Witten, “Anti-de Sitter space, thermal phase transition, and confine- ment in gauge theories,” Adv. Theor. Math. Phys.2(1998), 505-532 doi:10.4310/ATMP.1998.v2.n3.a3 [arXiv:hep-th/9803131 [hep-th]]. 1

    Pith/arXiv arXiv 1998

  7. [7]

    Black Holes and Random Matrices,

    J. S. Cotler, G. Gur-Ari, M. Hanada, J. Polchinski, P. Saad, S. H. Shenker, D. Stan- ford, A. Streicher and M. Tezuka, “Black Holes and Random Matrices,” JHEP 05(2017), 118 [erratum: JHEP09(2018), 002] doi:10.1007/JHEP05(2017)118 [arXiv:1611.04650 [hep-th]]. 1, 2

    Pith/arXiv arXiv 2017

  8. [8]

    Very long time scales and black hole thermal equilibrium,

    J. L. F. Barbon and E. Rabinovici, “Very long time scales and black hole thermal equilibrium,” JHEP11(2003), 047 doi:10.1088/1126-6708/2003/11/047 [arXiv:hep- th/0308063 [hep-th]]. 1

    arXiv 2003

  9. [9]

    A semiclassical ramp in SYK and in gravity,

    P. Saad, S. H. Shenker and D. Stanford, “A semiclassical ramp in SYK and in gravity,” [arXiv:1806.06840 [hep-th]]. 1 18

    Pith/arXiv arXiv

  10. [10]

    A note on black hole entropy and wormhole in- stabilities,

    J. L. F. Barb´ on and E. Velasco-Aja, “A note on black hole entropy and wormhole in- stabilities,” JHEP08(2025), 103 doi:10.1007/JHEP08(2025)103 [arXiv:2502.00769 [hep-th]]. 2, 17

    Pith/arXiv arXiv 2025

  11. [11]

    Instability of flat space at finite temperature,

    Gross, D., Perry, M. and Yaffe, L., “Instability of flat space at finite temperature,”Phys. Rev. D.25, 330-355 (1982,1). https://link.aps.org/doi/10.1103/PhysRevD.25.330 2

    doi:10.1103/physrevd.25.330 1982

  12. [12]

    Black hole entropy and the dimensional continuation of the Gauss-Bonnet theorem,

    M. Banados, C. Teitelboim and J. Zanelli, “Black hole entropy and the dimensional continuation of the Gauss-Bonnet theorem,” Phys. Rev. Lett.72(1994), 957-960 doi:10.1103/PhysRevLett.72.957 [arXiv:gr-qc/9309026 [gr-qc]]. 5

    Pith/arXiv arXiv 1994

  13. [13]

    Gravitational thermodynamics without the conformal factor problem: partition functions and Euclidean saddles from Lorentzian path integrals,

    D. Marolf, “Gravitational thermodynamics without the conformal factor problem: partition functions and Euclidean saddles from Lorentzian path integrals,” JHEP 07(2022), 108 doi:10.1007/JHEP07(2022)108 [arXiv:2203.07421 [hep-th]]. 5, 17

    Pith/arXiv arXiv 2022

  14. [14]

    How to tame your (black hole) saddles: Lessons from the Lorentzian Gravitational Path Integral,

    M. Kolanowski and D. Marolf, “How to tame your (black hole) saddles: Lessons from the Lorentzian Gravitational Path Integral,” [arXiv:2603.24681 [hep-th]]. 5

    Pith/arXiv arXiv

  15. [15]

    A Correspondence principle for black holes and strings,

    G. T. Horowitz and J. Polchinski, “A Correspondence principle for black holes and strings,” Phys. Rev. D55(1997), 6189-6197 doi:10.1103/PhysRevD.55.6189 [arXiv:hep-th/9612146 [hep-th]]. 5, 14

    Pith/arXiv arXiv 1997

  16. [16]

    Closed string tachyons and the Hagedorn transition in AdS space,

    J. L. F. Barbon and E. Rabinovici, “Closed string tachyons and the Hagedorn transition in AdS space,” JHEP03(2002), 057 doi:10.1088/1126-6708/2002/03/057 [arXiv:hep-th/0112173 [hep-th]]. 5

    Pith/arXiv arXiv 2002

  17. [17]

    A brief note on complex AdS-Schwarzschild black holes,

    R. Mahajan and K. Singhi, “A brief note on complex AdS-Schwarzschild black holes,” JHEP11(2025), 164 doi:10.1007/JHEP11(2025)164 [arXiv:2509.08883 [hep-th]]. 6

    Pith/arXiv arXiv 2025

  18. [18]

    Two dimensional Nearly de Sitter gravity,

    J. Maldacena, G. J. Turiaci and Z. Yang, “Two dimensional Nearly de Sitter gravity,” JHEP01(2021), 139 doi:10.1007/JHEP01(2021)139 [arXiv:1904.01911 [hep-th]]. 7

    Pith/arXiv arXiv 2021

  19. [19]

    Analytic Continuation Of Chern-Simons Theory,

    E. Witten, “Analytic Continuation Of Chern-Simons Theory,” AMS/IP Stud. Adv. Math.50(2011), 347-446 [arXiv:1001.2933 [hep-th]]. 7

    Pith/arXiv arXiv 2011

  20. [20]

    Wick Rotation and the Positivity of Energy in Quantum Field Theory,

    M. Kontsevich and G. Segal, “Wick Rotation and the Positivity of Energy in Quantum Field Theory,” Quart. J. Math. Oxford Ser.72(2021) no.1-2, 673-699 doi:10.1093/qmath/haab027 [arXiv:2105.10161 [hep-th]]. 17

    Pith/arXiv arXiv 2021

  21. [21]

    A Note On Complex Spacetime Metrics,

    E. Witten, “A Note On Complex Spacetime Metrics,” [arXiv:2111.06514 [hep-th]]. 17 19

    Pith/arXiv arXiv