{"id":"2b61e58f-5256-4ce2-8ba3-472c886a35a8","arxiv_id":"2608.02743","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Smooth filter projection of erratic spectral data reproduces random-matrix ramp and plateau in a single chaotic system, and predicts double-exponential hyper-nonperturbative effects in holographic gravity.","lead":"This paper argues that the random-matrix-like 'ramp' and 'plateau' patterns of quantum chaotic spectra, and their gravitational wormhole counterparts, can come from a single system without ensemble or explicit energy averaging, if a smooth filter separates smooth from erratic parts. The same framework predicts double-exponential 'hyper-structures' in holographic theories and gravity.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The ramp/plateau derivation depends on underived rules for F on products of erratic resolvents; Sec. II B remark 3 concedes the gap, and the plateau adds the equally unproven additive determinant (2.43) and re-exponentiation (2.55).","rationale":"The reader's CONDITIONAL verdict is well placed; this stress test sharpens why. The few-body ramp derivation is the strongest part, but even it is a diagonal approximation without an independent definition of F on products. The plateau is more fragile because it depends on (2.43) and (2.55), neither of which is derived. The Selberg trace formula provides a rare exact setting where (2.43) can be tested directly. The concern is not that the paper is internally inconsistent, but that the central claim is conditional on an unproven filter calculus for products; this does not change the reader's verdict.","tokens_in":55571,"tokens_out":14718,"duration_ms":142136,"concrete_test":"Use a compact hyperbolic surface as the exact Selberg trace formula laboratory: numerically construct Δ±(E) from the Selberg zeta function with the phase convention of (2.40), and compare D(E)=det(E−H) with Δ+(E)+Δ−(E) over a range of E in the semiclassical limit. If the additive reconstruction (2.43) fails at any relevant order, the plateau and all trans-dipole results lose their foundation; if it holds, the remaining gap is the absence of an independent construction of F on products, which should then be tested by computing F{ρ^errρ^err} from exact periodic-orbit data without imposing diagonal phase cancellation.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—that (1.21) and the plateau follow from a single-system projection without averaging—turns on the action of F on products of the erratic Gutzwiller components. Yet F is axiomatized only on linear combinations and on single resolvents: (1.2) and (2.10) fix F{R±} and F{ρ^err}, but not F{ρ^err(E+ϵ/2)ρ^err(E−ϵ/2)}. In (2.22)–(2.23) this action is chosen by keeping n1=−n2 and a1=a2; that is exactly the diagonal approximation whose standard justification is energy averaging. The paper explicitly concedes in Sec. II B remark 3 that 'we currently lack a first-principles method for decomposing products or arbitrary functions of the density.' The plateau then rests on two further unproven inputs: (2.43), D(E)=Δ+(E)+Δ−(E), which is asserted with only a WKB plausibility check, and the re-exponentiation rule (2.55), which assumes all higher connected F-correlations of Y are negligible. Every holographic and hyper-instanton conclusion imports these structures. Until F on products is constructed from independent axioms, the universal RMT forms (1.21) and (2.82) are not derived; they are, in effect, the definition of the projection.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes that the ramp, the plateau, and their Euclidean-wormhole duals need not be interpreted as consequences of ensemble or explicit spectral averaging. Instead, it postulates a smooth filter projection F that separates each observable into a smooth macroscopic part and an erratic microscopic part, and claims that the universal random-matrix form of the spectral form factor follows from applying F to products of erratic densities of a single chaotic system. Section II carries this out for few-body systems using the Gutzwiller trace formula: the ramp is obtained from diagonal orbit pairings plus the Hannay-Ozorio de Almeida sum rule, and the plateau is obtained from correlations of dressed resolvents built from spectral determinants that are reconstructed additively from upper- and lower-half-plane data. Section III extends the formalism to the spectral curve, the forbidden region, and thermal partition functions. Section IV postulates a Gutzwiller-like structure for many-body holographic systems and constructs baby-universe condensates in a multiverse Hilbert space to derive double-exponential effects. Section V studies multiverse instantons in AdS3 gravity, including an explicit c-dependent instanton action. The paper is unusually explicit about its postulates and limitations, including the concession in Sec. II B, Remark 3, that no first-principles method for decomposing products of densities is currently available.","tokens_in":56074,"tokens_out":9100,"duration_ms":91816,"significance":"The few-body ramp derivation in Sec. II B is a genuine and clean result: it adapts Berry's diagonal approximation and, crucially, uses the externally established Hannay-Ozorio sum rule to convert a discrete orbit sum into the universal 1/epsilon^2 two-point singularity without any explicit energy window. This part deserves credit as a concrete demonstration of the proposed filter mechanism. The construction of dressed resolvents, the macroscopic oscillatory term, and the plateau is formally elegant, and the cancellation of the non-universal constants c(E) and g-hat(E) in Eq. (2.80) is a nice structural feature. The paper is also candid about the extent to which the many-body and gravitational sections are postulates rather than derivations. If the filter-on-products rules were derived from independent axioms, the paper would provide a significant reinterpretation of wormhole amplitudes and of the origin of RMT universality in individual chaotic systems. As it stands, the significance is conditional: the plateau and all subsequent hyper-non-perturbative claims rest on unproven prescriptions for how F acts on products of erratic quantities.","major_comments":[{"comment":"The central no-averaging claim rests on the action of F on products of erratic densities, but the axioms (1.2)-(1.3) and the single-resolvent rules (2.10) do not determine F on products. Equation (2.22) fixes F{rho_err(E+epsilon/2) rho_err(E-epsilon/2)} by keeping only n1=-n2 and a1=a2, which is exactly the diagonal approximation whose standard justification is energy averaging. The paper concedes in Remark 3 that 'we currently lack a first-principles method for decomposing products or arbitrary functions of the density.' As a result, the universal ramp (2.23) is not derived from the filter axioms; it is, in effect, an imposed rule for the action of F on this particular product. This does not invalidate the calculation as a formal statement, but it does mean that the claim that the ramp is obtained 'without averaging' is not yet established.","section":"Sec. II B, Eq. (2.22), Remark 3"},{"comment":"Equation (2.43) reconstructs the real spectral determinant as D(E) = Delta_+(E) + Delta_-(E). This additive prescription is load-bearing: it opens the trans-dipole channel that generates both the oscillatory term (2.66) and the plateau (2.82). The paper supports (2.43) only by a 'WKB plausibility check' and an analogy to the Berry-Keating construction. No derivation is given for generic chaotic systems. Since all later claims in Sections II-IV depend on the existence of trans-dipoles, (2.43) should either be proven in a controlled setting (for example, from the Selberg trace formula for hyperbolic surfaces, or in explicit WKB models) or be stated as an additional postulate whose consequences are then conditional. As written, the derivation of the plateau is not complete.","section":"Sec. II C, Eq. (2.43)"},{"comment":"The re-exponentiation rule F{e^A} = e^{F{A} + (1/2) F{A^2}_c} assumes that all higher connected filtered correlations of Y are negligible. This truncation is not justified, and it is not a consequence of the two-point rules (2.48)-(2.54). The plateau calculation relies on the exact cancellation of c(E) and g-hat(E) in Eq. (2.80), and that cancellation would generically be modified if higher cumulants contributed. The paper should state this cumulant-truncation assumption explicitly and provide an estimate of the corrections. Without such control, Eq. (2.82) should be viewed as the leading term of a conjectured projection scheme rather than as a derivation of the plateau.","section":"Sec. II C, Eq. (2.55); Sec. II D, Eqs. (2.75)-(2.80)"},{"comment":"The text states that working with periodic orbits 'inherently implies that a level of coarse-graining over the exact quantum spectrum has already taken place.' If the Gutzwiller representation already coarse-grains the exact spectrum, the distinction between the smooth filter and standard spectral averaging becomes less sharp than the paper claims. The manuscript should specify precisely what type of averaging or coarse-graining is built into the semiclassical representation, and how the filter projection F differs from an energy-window average in a controlled way. This is not merely a terminological point: the claim to derive the ramp and plateau 'without averaging' depends on where this coarse-graining is allowed to enter.","section":"Sec. I B, discussion after Eq. (1.21)"},{"comment":"The many-body and holographic extension is explicitly postulated, and the paper correctly notes that a Gutzwiller trace formula for the genuine many-body regime does not yet exist. Consequently, all double-exponential predictions, including the rapid macroscopic oscillations (4.29), the plateau (4.30), and the hyper-instantons (4.34)-(4.35), are conditional on the assumed existence of the structure in (4.1)-(4.24). This is acceptable if the paper is read as a programmatic proposal, but the abstract and conclusions should carry this qualification into their predictive claims. In particular, the phrase 'predicts universal rapid macroscopic oscillations' should be revised to 'predicts, under the postulated Gutzwiller-like structure, ...' so that the conditional status is not hidden.","section":"Sec. IV A, postulates (4.1)-(4.24)"}],"minor_comments":[{"comment":"The phrase 'linear-tramp' appears in the paragraph after Eq. (1.19); it should be 'linear ramp'.","section":"Sec. I B"},{"comment":"The phrase 'we have used the dentity' should read 'we have used the identity'.","section":"Sec. II C, after Eq. (2.61)"},{"comment":"The divergent delta-function term in (2.83) is regulated in Remark 2 by setting eta = 1/t_H, but the Fourier transforms in (2.86)-(2.88) use an infinitesimal eta and then take eta to zero. The relationship between these two regularization procedures should be spelled out, since the claim that the 1/eta divergences cancel depends on the order of limits.","section":"Sec. II D, Eqs. (2.82)-(2.88)"},{"comment":"The notation F{A^2}_c is not defined precisely. If it denotes the connected part of the filtered two-point function, the definition should be given explicitly so that the re-exponentiation formula can be checked.","section":"Sec. II C, Eq. (2.55)"},{"comment":"The parity of the logarithmic branch in Eq. (4.67) is physically important for the pole structure of the plateau. A brief derivation of the branch choices used to obtain g_+- and g_++ from the double-trump amplitude would help the reader verify the i0 prescriptions in (2.50) and (4.13).","section":"Sec. IV C 2, Eqs. (4.64)-(4.67)"},{"comment":"The sign oscillation of the would-be instanton action with c is an interesting and honest result, but the paper should make clearer in the main text that this provides a sharp falsifiable condition: if the proposed multiverse-instanton interpretation is correct, only values of c for which I_MI > 0 should exhibit the associated non-perturbative corrections.","section":"Sec. V C, Eq. (5.48)"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is honest and well written, and the few-body ramp section is a real contribution. The main risk is that the filter-on-products prescriptions, especially (2.22) and (2.43)-(2.55), are doing essentially all the work in the plateau and hyper-non-perturbative claims. I recommend asking for either a derivation of these rules from an independent principle in a controlled setting, or a clear restatement of the paper's main results as conditional on the postulated filter. I do not think rejection is warranted, because the paper contains a sound derivation of the ramp and is transparent about its limitations."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know about Hong Liu's new paper: it's not a closed proof, but a sprawling, self-aware research program. The one part that is genuinely solid is the few-body ramp. The diagonal approximation in (2.22)–(2.23) is a clean adaptation of Berry's argument, anchored by the Hannay–Ozorio de Almeida sum rule. That gives real support for the single-system filter picture, at least for the ramp.\n\nWhat's new: the explicit construction of the plateau from correlations of trans-bound resolvents in Secs. II C–D, and the translation of those structures into the multiverse Hilbert space as baby-universe condensates and multiverse instantons. The JT gravity checks (4.76), (4.83) and the AdS3 analysis in Sec. V are concrete and honest—the paper plainly says the instantons vanish once descendants are included, which is a good-faith caveat.\n\nThe soft spot is exactly where the stress-test note points. F is axiomatized only on single resolvents. The action of F on products is chosen by the diagonal approximation in (2.22), which in the standard literature is justified by energy averaging; the paper explicitly concedes in Sec. II B remark 3 that it lacks a first-principles construction of F on products. The plateau then rests on two further unproven inputs: the additive determinant prescription D(E)=Δ+(E)+Δ−(E) in (2.43) and the re-exponentiation rule (2.55) that drops all higher connected F-correlations. Until those are derived from independent axioms, the universal forms (1.21) and (2.82) are not so much derived as built into the definition of the projection. That is an honest gap, not a fatal one, but it makes the central claim conditional.\n\nThe holographic half is even more explicitly a postulate: Sec. IV A assumes a minimal Gutzwiller-like structure in the large-N limit, and the double-exponential predictions import that postulate. So the right reading is: the few-body ramp is a result; the plateau is a plausible mechanism; the hyper-structures are a program.\n\nBottom line: this deserves a serious referee. The questions about averaging and factorization are important, and the ramp part is a real contribution. A referee should push hard for a derivation of F on products, or at least a clear statement of which results are axioms and which are consequences. I would not desk-reject. I would send it out, with the expectation of major revision and possibly a recommendation to split the few-body result from the holographic speculation. I would cite it, yes, for the ramp and for the clean statement of the program.","headline":"A serious program that derives the few-body ramp cleanly, but the plateau and hyper-structure claims rest on underived filter axioms; worth refereeing, with a referee who will hold the line on the product-filter gap.","tokens_in":56473,"tokens_out":3057,"would_cite":true,"duration_ms":27980,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The ramp and plateau of the spectral form factor are single-system phenomena produced by a smooth filter projection, not by averaging.","keywords":["quantum chaos","spectral form factor","ramp and plateau","Gutzwiller trace formula","smooth filter projection","wormholes","holography","hyper-non-perturbative effects"],"falsifier":"Take a numerically computed chaotic spectrum, construct the periodic-orbit sums for the product of two densities at $E\\pm\\epsilon/2$, apply the filter by retaining only terms whose actions cancel exactly, and check whether the $\\epsilon\\to 0$ limit reproduces $-1/(2\\pi^2\\epsilon^2)$ together with the known subleading corrections; failure to find the ramp in a single system's filtered product would falsify the central claim. Independently, one can test the additive determinant identity $D(E)=\\Delta_+(E)+\\Delta_-(E)$ order by order in a WKB solvable model, since the plateau derivation collapses if trans-dipole contributions vanish.","tokens_in":55260,"feed_emoji":"🌀","tokens_out":13902,"duration_ms":115006,"temperature":0.7,"pith_summary":"This paper argues that the universal random-matrix signatures of quantum chaos—the linear ramp and the plateau of the spectral form factor—need not come from an ensemble of Hamiltonians or from averaging over energy windows. They are macroscopic smooth structures hidden inside the erratic, rapidly oscillating part of a single system's density of states, and they can be extracted by a smooth filter projection $F$ that discards micro-oscillations. Using the Gutzwiller trace formula, the paper derives the ramp from the diagonal pairing of periodic orbits together with the classical sum rule for periodic-orbit stability amplitudes, and the plateau from correlations of dressed resolvents built from spectral determinants. The same structure is then postulated for holographic large-$N$ systems, predicting density-of-states oscillations and possible hyper-instantons that are double-exponential in $1/N^2$; on the gravity side, wormhole amplitudes play the role of the filter's connected contractions, and multiverse condensates reproduce the same effects. If correct, semiclassical gravity computes smooth filtered quantities of a single boundary theory, not an average over theories.","feed_headline":"No averaging: one chaotic system yields the universal ramp and plateau","feed_subtitle":"A smooth filter extracts random-matrix universality from one chaotic spectrum, with wormholes as its gravity dual.","key_machinery":"The load-bearing object is the smooth filter projection $F$, defined by $F\\{A_{\\rm err}\\}=0$, $F^2=F$, linearity when the coefficients are smooth transseries, and positivity $F\\{A^*A\\}\\ge 0$. It acts on the Gutzwiller decomposition $R_+(E)=\\bar R(E)-(i/\\hbar)\\sum_a T_a F_a e^{iS_a(E)/\\hbar}$, keeping only the Weyl part and killing the periodic-orbit oscillations. Four further pieces carry the argument: the diagonal approximation that pairs orbits with exactly cancelling phases; the periodic-orbit sum rule $\\sum_a |F_a|^2 f(T_a)=\\int dT\\, f(T)/T$, which turns diagonal sums into smooth integrals; the additive determinant reconstruction $D(E)=\\Delta_+(E)+\\Delta_-(E)$, which opens the trans-dipole channel; and the logarithmic singularity $g_{+-}(E_1,E_2)=-\\log(-i(E_1-E_2+i0))+c(E)$, which lets a resolvent contract with a trans-dipole to form the trans-bound resolvent $\\hat R^-_{++}(E)=i e^{\\hat g(E)-c(E)}e^{2\\pi i\\bar N(E)-2i\\phi}:e^{Y^*-Y}:$. The plateau is the overlap of two opposite-orientation trans-bound resolvents; gravity enters because the wormhole amplitude is identified with $F\\{Z_1Z_2\\}_c$, and the multiverse Hilbert space exponentiates single-boundary states into baby-universe condensates.","core_discovery":"The central claim, stated on the paper's own terms, is equation (1.21): for a single chaotic quantum system, the smooth filter projection of the product of two densities of states gives the universal random-matrix two-point function, $F\\{\\rho(E+\\epsilon/2)\\rho(E-\\epsilon/2)\\}_c = -1/(2\\pi^2\\epsilon^2) + \\cos(2\\pi\\bar\\rho(E)\\epsilon)/(2\\pi^2\\epsilon^2)$, with no ensemble and no explicit spectral average. The ramp follows because only diagonal periodic-orbit pairs with $n_1=-n_2$ and $a_1=a_2$ survive the projection, and the periodic-orbit sum rule converts the remaining sum into $-1/(2\\pi^2\\epsilon^2)$. The plateau follows from a second, analogous two-point function in which each resolvent is replaced by a trans-bound dressed resolvent $\\hat R^-_{++}(E)$, formed when a resolvent binds to a spectral dipole through a logarithmic cross-sheet contraction; opposite orientations of two such objects produce the cosine term with the Heisenberg-time phase $e^{2\\pi i\\bar\\rho(E)\\epsilon}$. For holographic systems the paper postulates the same minimal Gutzwiller-like structure, which predicts rapid macroscopic oscillations (1.10) and possible hyper-instantons, both involving double exponentials in $1/N^2$; in the bulk these are reproduced by baby-universe condensates and multiverse instantons built from wormhole amplitudes.","pith_inferences":["Beyond the paper's own claims, the filter should be directly testable numerically: in a chaotic billiard or compact hyperbolic surface one can build the periodic-orbit sums, apply the diagonal projection, and check whether the single-system filtered product of densities converges to the ramp without any energy average.","The additive determinant reconstruction $D(E)=\\Delta_+(E)+\\Delta_-(E)$ is checked only in a WKB limit; a sharper inference is that this identity is the real target to test, since the plateau derivation collapses if trans-dipole contributions fail in an exactly solvable model.","A further extension suggested by the framework is to construct filters for heavy-operator OPE coefficients rather than densities alone; if wormholes filter erratic OPE data the same way, higher-point heavy correlators would inherit the same universal structure.","In the AdS3 example the would-be multiverse-instanton action oscillates with the central charge, so an implication is that the arithmetic of $c$ may decide whether hyper-instantons contribute; modular invariance could settle that in pure AdS3."],"forward_implications":["The ramp is a single-system effect: filtering the product of two erratic Gutzwiller densities gives $-1/(2\\pi^2\\epsilon^2)$ through diagonal orbit pairing and the sum rule, so no ensemble or energy-window average is required.","The plateau is the same type of two-point correlation as the ramp, with each resolvent replaced by a trans-bound resolvent; its Fourier transform cuts off the linear ramp exactly at the Heisenberg time $t_H=2\\pi\\bar\\rho(E)$.","The spectral curve of a chaotic system is the Riemann surface of the smooth Weyl resolvent, and all erratic components and their filtered products live on that curve, making the smoothed two-point functions regulator-independent.","For holographic systems the postulated Gutzwiller-like structure predicts universal rapid oscillations in the density of states, $\\bar\\rho(E)-\\frac{1}{\\pi}e^{\\hat g(E)-c(E)}\\cos(2\\pi\\bar N(E)-2\\phi)$, which are level-2 transseries terms, i.e. double exponentials in $1/N^2$.","Wormhole amplitudes act as connected contractions (baby-universe propagators) in the multiverse Hilbert space; their exponentiation gives baby-universe condensates whose overlaps reproduce the density oscillations and the plateau, with possible multiverse instantons adding double-exponential corrections to partition functions."],"supporting_citations":[{"why":"Supplies the original proposal that large-N observables split into smooth and erratic parts and that the gravitational path integral acts as the smooth filter.","marker":"[1]"},{"why":"Gives the Gutzwiller trace formula for chaotic few-body systems, the Tier-II representation from which the ramp and plateau are derived.","marker":"[25–27]"},{"why":"Provides the periodic-orbit sum rule that converts the diagonal periodic-orbit sum into the smooth integral giving the ramp.","marker":"[58]"},{"why":"Provides the semiclassical spectral-rigidity calculation in the diagonal approximation that this paper reinterprets as a smooth filter projection.","marker":"[59]"},{"why":"Develops the periodic-orbit theory of universal level correlations, including correlated orbit pairs, that the plateau derivation adapts without energy averaging.","marker":"[62,63,72]"},{"why":"Supplies the exact-versus-semiclassical dressed-resolvent and oscillatory-density machinery used for the plateau and hyper-structures.","marker":"[67]"},{"why":"The double-cone wormhole is the gravitational ramp amplitude that the framework reproduces as a filtered product in a single theory.","marker":"[2]"},{"why":"JT gravity as a matrix integral provides explicit double-exponential effects, the double trumpet as a baby-universe propagator, and checks of the plateau.","marker":"[3]"},{"why":"Supplies off-shell wormhole amplitudes that the paper identifies with filtered products and uses as baby-universe propagators in higher dimensions.","marker":"[4]"}],"fun_headline_variants":["Ramp and plateau from one chaotic system, no averaging","Wormholes and spectral universality without any averaging","Smooth filter reveals random-matrix universality from one spectrum","No averages needed: one chaotic spectrum yields the ramp and plateau","Single chaotic system, no ensemble: universal ramp and plateau"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument assumes that one can split each observable into a smooth part and an erratic part in a well-defined way, and that the same split can be applied consistently to products of erratic parts with the cross-sheet contractions taking the specific logarithmic form (2.50); the paper concedes it has no first-principles method for doing this to products.","fun_headline_variants_meta":{"raw":{"variants":["Ramp and plateau from one chaotic system, no averaging","Wormholes and spectral universality without any averaging","Smooth filter reveals random-matrix universality from one spectrum","No averages needed: one chaotic spectrum yields the ramp and plateau","Single chaotic system, no ensemble: universal ramp and plateau"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000204,"raw_usage":{"total_tokens":1487,"prompt_tokens":1139,"completion_tokens":348,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":755,"completion_tokens_details":{"reasoning_tokens":266}},"tokens_in":755,"tokens_out":348,"duration_ms":3179,"temperature":1.0,"reasoning_tokens":266,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T15:01:27.965256+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a numerically computed chaotic spectrum, construct the periodic-orbit sums for the product of two densities at $E\\pm\\epsilon/2$, apply the filter by retaining only terms whose actions cancel exactly, and check whether the $\\epsilon\\to 0$ limit reproduces $-1/(2\\pi^2\\epsilon^2)$ together with the known subleading corrections; failure to find the ramp in a single system's filtered product would falsify the central claim. Independently, one can test the additive determinant identity $D(E)=\\Delta_+(E)+\\Delta_-(E)$ order by order in a WKB solvable model, since the plateau derivation collapses if trans-dipole contributions vanish.","supporting_citations":[],"review_version":2}