{"id":"288c2461-8d1f-4885-b642-a02841fd76da","arxiv_id":"1908.03502","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"For an eternal black hole, the interior partition function factorizes as the square of the exterior one, fixing the behind-the-horizon cutoff in terms of the UV cutoff.","lead":"Using the Papadodimas Raju construction of interior operators, this paper relates the partition function of modes behind a black hole horizon to those outside, and shows that a UV cutoff fixes a cutoff behind the horizon. It offers a new derivation of a known result in black hole complexity and TTbar-deformed CFTs.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Scalar-to-graviton extension of Z^II ∝ (Z^I)^2 is unproven; Eq. (24) inherits this unsupported step.","rationale":"The reader's weakest assumption is exactly the scalar-to-graviton extension, and I agree that it is the most load-bearing step. The paper derives the factorization for a generalized free scalar using Eq. (10), then switches to the graviton effective action with no justification beyond the assertion that the leading effective action is the classical action. The stress tensor is not a generalized free field: its connected three-point function does not vanish at leading order in 1/N, so the factorization that underlies Z^II ∝ (Z^I)^2 does not follow. Since Eq. (24) is obtained by substituting on-shell gravitational actions into Eq. (14), the central claim that a UV cutoff enforces a behind-horizon cutoff is contingent on this unsupported assumption. I also note that the paper assumes the existence of r0 rather than deriving it, which reinforces the conditional character of the result. The agreement with the holographic-complexity formula in [8] is independent support for the final relation, but it does not validate the proposed derivation. The appropriate disposition is conditional acceptance, matching the reader's verdict; no verdict change is needed.","tokens_in":8470,"tokens_out":10279,"duration_ms":115331,"concrete_test":"Compute the leading-order connected three-point function of the boundary stress tensor dual to the graviton on the black brane background. If it is nonzero, as expected in Einstein gravity, then the correlator factorization (10), and hence Z^II ∝ (Z^I)^2 for the graviton effective action, is false. This directly tests the step after Eq. (13) that leads to Eq. (24).","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 3 derives Z^II ∝ (Z^I)^2 from large-N factorization (Eq. (10)) for a generalized free scalar. Immediately after Eq. (13), the paper extends this to the graviton effective action, saying only that at leading order the effective action is the classical on-shell action. But the boundary dual of the graviton is the stress tensor, which is not a generalized free field: in any holographic CFT the stress tensor has non-vanishing connected three-point functions at leading order in 1/N, so correlators do not factorize as in Eq. (10). Consequently the proportionality Z^II ∝ (Z^I)^2, and hence Eq. (14) and the key relation Eq. (24), are not derived for the gravitational action. The existence of the interior cutoff r0 is also assumed before Eq. (23) (the paper posits a cutoff rather than deriving its necessity), so the 'enforced' behind-horizon cutoff is only as secure as the unproven graviton factorization.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper uses the Papadodimas-Raju construction of black-hole interior operators to propose a relation between the partition function of interior modes and the partition functions of the left and right exterior modes of an eternal black brane. For generalized free fields at large N the relation is Z^(II) ∝ Z^(I) Z^(III), which in the symmetric case becomes Z^(II) ∝ (Z^(I))^2. The paper then extends this relation to the graviton effective action, replaces the restricted partition functions by on-shell Einstein-Hilbert actions with Gibbons-Hawking and counterterm terms, and computes the effect of a finite radial UV cutoff r_c. By postulating a behind-the-horizon cutoff r_0, it derives Eq. (24), which fixes r_0 in terms of r_c and reproduces the cutoff relation obtained earlier in the holographic-complexity context. Generalizations to unequal left/right cutoffs and to a single exterior plus mirror-partner construction are given in Eqs. (26)-(29). The central new step is the application of the scalar factorization relation to gravity, and the central assumption is the existence of the behind-the-horizon cutoff r_0.","tokens_in":8659,"tokens_out":15954,"duration_ms":154069,"significance":"If the derivation were completed, Eq. (24) would be a concrete and checkable relation tying a UV cutoff to a cutoff behind the horizon, and it would explain the different powers of r_0 and r_c seen in the complexity analysis. The on-shell action computations are explicit, the small-cutoff asymptotics are clean, and the paper is honest about some of its own limitations: it assumes rather than derives the existence of r_0 before Eq. (23), and it acknowledges the heuristic nature of the restricted partition functions. The main strength is that the final relation is not a vague expectation but a specific equation that can be compared with independent computations. Its significance is currently conditional, because the derivation relies on an unproven scalar-to-graviton extension and on the existence of solutions to Eq. (24). The agreement with [8] is a useful consistency check, but it is not an independent external validation because [8] shares two of the present authors.","major_comments":[{"comment":"The extension of the factorization relation from the scalar case to the graviton effective action is not justified. Equation (10) relies on generalized free-field factorization at large N, a property that fails for the stress-tensor multiplet: the connected three-point function of the stress tensor is non-vanishing at leading order in 1/N and does not factorize. Since Eq. (14) and hence Eq. (24) are obtained by applying Eq. (13) to the Einstein-Hilbert action, the central result is unsupported at this step. The authors should either prove the graviton factorization at the required order or explicitly restrict the claim to scalar fields.","section":"Section 3, after Eq. (13)"},{"comment":"The existence of the behind-the-horizon cutoff r_0 is assumed, not derived. The text states 'we will assume that there is also a finite radial cutoff behind the horizon located at r_0', but the abstract and conclusions claim that a UV cutoff 'enforces' a cutoff behind the horizon. At most, the computation shows that if r_0 exists, Eq. (24) fixes its value. Moreover, no existence proof is given: for r_c close to r_h the right-hand side of Eq. (24) approaches r_h^{-(d+1)}, while the left-hand side as a function of r_0>r_h is bounded above by approximately 0.207 r_h^{-(d+1)}; hence Eq. (24) has no real solution in that regime. The claimed enforcement should be restricted to a proven range of r_c.","section":"Section 3, before Eq. (23)"},{"comment":"The 'restricted partition functions' Z^(I) and Z^(II) are not defined with sufficient precision. A bulk gravitational path integral is a global object; restricting it to field configurations of the form (3)-(5) is a formal manipulation, and the delta-functional remark in footnote 2 does not specify integration measures, boundary conditions on the cutoff surfaces, or the treatment of the horizon. The subsequent saddle-point identification of these objects with on-shell actions in separated bulk regions is therefore heuristic. The paper should either define these restricted partition functions carefully or present the on-shell-action relation as a conjecture motivated by the Papadodimas-Raju construction.","section":"Section 3, Eqs. (7)-(9)"}],"minor_comments":[{"comment":"There are several typographical errors: 'Penrose digram' in Section 2 should be 'Penrose diagram'; 'redial' in Section 3 should be 'radial'; 'react into' before Eq. (22) should be 'recast into'; 'grater' in Section 4 should be 'greater'; 'gravity daul' in footnote 2 should be 'gravity dual'; and 'liner' in Section 3 should be 'linear'.","section":"Throughout"},{"comment":"Region IV is labeled in the Penrose diagram but is never defined or used in the text; please add a definition or remove the label.","section":"Figure 1"},{"comment":"The domain of r_0 should be stated explicitly: the square root requires r_0>r_h, so the text should specify that the behind-the-horizon cutoff lies in the branch r_0>r_h rather than relying on the phrase 'behind the horizon' alone.","section":"Eq. (24)"},{"comment":"The proportionality constants in Eqs. (11) and (12) are not tracked; footnote 5 states that they are cutoff-independent, but this cutoff-independence is asserted rather than demonstrated. Please either prove it or state it as an additional assumption.","section":"Eqs. (11)-(14)"},{"comment":"The phrase 'in terms those partition functions' is missing a word and should read 'in terms of those partition functions'.","section":"Abstract"}],"recommendation":"major_revision","confidential_remarks":"The external benchmark [8] is by the same group (Akhavan, Alishahiha, Naseh, and Zolfi), so the agreement in Eq. (24) is a self-consistency check rather than independent confirmation. This is not a defect in the derivation itself, but it should be weighed when calibrating how strongly the result is advertised as validated. The paper fits the journal's scope and is clearly written; the main risk is that the abstract overstates the conclusiveness of the derivation relative to the assumptions made in Section 3."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: the paper is a competent re-derivation of a result the authors already published in [8], plus a few new asymmetric cutoff formulas. The on-shell action algebra is standard, and the final relation (24) follows from Eq. (14) without any obvious error. The main soft spot is that Eq. (14) is not actually derived for gravity.\n\nThe paper starts with the Papadodimas-Raju construction. For a generalized free scalar, large-N factorization gives Z^II proportional to Z^I Z^III, hence Z^II proportional to (Z^I)^2. That step is fine. The trouble appears right after Eq. (13): the same proportionality is extended to the graviton effective action, with only the remark that the leading effective action is the classical action. The stress tensor is not a generalized free field. It has a nonzero connected three-point function at leading order in the 1/N expansion, so the factorization in Eq. (10) does not apply. A nonlinear bulk extension of the interior-operator construction would be needed, and none is provided. Since Eq. (14) is the load-bearing relation, the main claim is conditional on an unproven assumption.\n\nSecond, the behind-the-horizon cutoff r0 is assumed to exist, not derived. The sentence \"we will assume that there is also a finite radial cutoff behind the horizon\" appears just before Eq. (23), and the word \"enforce\" in the abstract is too strong. What is actually shown is: if Eq. (14) holds for gravity and if such a cutoff exists, then consistency fixes its value. Those are real conditionals, but they are not enforcement.\n\nCredit where it is due: the asymmetric formulas (26) and (29) are not in [8], and the relation Z^II proportional to Z^I Z^III gives a clean explanation of the power mismatch between r0 and rc. The paper is short, readable, and the action computations are transparent. The agreement with [8] is not suspicious—the overlap is stated—but it does mean the benchmark is self-referential, so it cannot validate the new derivation.\n\nWho should read this: people working on holographic complexity and interior probes. It is a useful cross-check, not a breakthrough. I would send it to a serious referee rather than desk-reject, because the gap is identifiable and might be fixable by either a proper justification of the graviton step or by presenting Eq. (14) as a conjecture. As it stands, I would not cite it as the primary derivation of the behind-the-horizon cutoff.","headline":"Clean but incremental: an alternative derivation of the authors' own behind-the-horizon cutoff result, with a genuinely unjustified step from scalar factorization to the graviton effective action.","tokens_in":9179,"tokens_out":5907,"would_cite":false,"duration_ms":65151,"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 paper derives a fixed relation between a UV cutoff and a cutoff behind the horizon, matching holographic complexity.","keywords":["black hole interior","Papadodimas-Raju operators","partition function","UV cutoff","behind the horizon cutoff","holographic complexity","eternal black hole","generalized free fields"],"falsifier":"Compute the one-loop (i.e., $O(1/N)$) graviton partition function in region II with the interior cutoff in place and check whether the exact relation $\\Gamma^{(II)} = 2\\Gamma^{(I)} + \\text{const}$ survives; any non-vanishing $1/N$ correction would modify Eq. (24), and the deviation would show up as a shift in the effective cutoff $r_0$ inferred from interior correlators.","tokens_in":8252,"feed_emoji":"🕳️","tokens_out":12261,"duration_ms":111430,"temperature":0.7,"pith_summary":"Using the Papadodimas–Raju construction of interior operators, this paper establishes a general relation between partition functions for the inside and outside of an eternal black hole: the interior partition function is proportional to the product of the left- and right-exterior partition functions, and for identical exterior modes it is the square of either one. Treating the graviton effective action at leading order as the on-shell action, the authors convert this into a relation between cutoff versions of the actions and derive a formula that fixes a behind-the-horizon cutoff $r_0$ in terms of the UV cutoff $r_c$ and the horizon radius $r_h$. The formula, Eq. (24), is exactly the one previously obtained from holographic complexity. If correct, it means a physicist who imposes a finite UV cutoff—for example through a $T\\bar{T}$-type deformation—cannot leave the interior of the black hole uncut; the interior cutoff is forced by the exterior one.","feed_headline":"Finite UV cutoff forces a cutoff behind the black hole horizon","feed_subtitle":"Interior operators split into two exterior copies, yielding Eq. (24) and matching holographic complexity.","key_machinery":"The load-bearing object is the Papadodimas–Raju interior operator $\\varphi^{(II)}_{\\mathrm{CFT}}$, a CFT operator built from two copies of exterior generalized free fields, $O_{\\omega,k}$ and $\\tilde{O}_{\\omega,k}$, with the same Klein–Gordon modes continued through the horizon. The key identity is large-$N$ factorization of correlators for generalized free fields,\n$$\\langle O_1\\cdots O_n \\tilde{O}_1\\cdots \\tilde{O}_m\\rangle = \\langle O_1\\cdots O_n\\rangle \\langle \\tilde{O}_1\\cdots \\tilde{O}_m\\rangle + O(1/N),$$\nwhich yields $Z^{(II)} \\propto Z^{(I)} Z^{(III)}$. The rest of the argument is a saddle-point computation: the Einstein–Hilbert action with Gibbons–Hawking and counterterm terms is evaluated on shell in regions I and II, with and without the radial cutoff, and the cutoff/no-cutoff difference is equated between the two regions through the exponentiated relation. This comparison produces Eq. (24), the fixed relation between $r_0$, $r_c$, and $r_h$.","core_discovery":"The central discovery is the factorization of interior physics: for generalized free fields with large-$N$ factorization, the interior partition function satisfies $Z^{(II)} \\propto Z^{(I)} Z^{(III)}$, and for an eternal black hole with commuting left and right operators, $Z^{(II)} \\propto (Z^{(I)})^2$. Promoting this proportionality to the graviton effective action through $e^{i\\Gamma^{(II)}} \\propto e^{2i\\Gamma^{(I)}}$ and comparing on-shell actions with and without a cutoff at $r = r_c$, the paper derives\n$$\\frac{1}{$r_0^{{d+1}}$}\\left(\\sqrt{\\frac{$r_0^{{d+1}}$}{$r_h^{{d+1}}$} - 1} - 1\\right) = \\frac{1}{$r_c^{{d+1}}$}\\left(1 - \\sqrt{1 - \\frac{$r_c^{{d+1}}$}{$r_h^{{d+1}}$}}\\right)^2,$$\nwhich fixes the behind-the-horizon cutoff $r_0$ once the exterior cutoff $r_c$ is chosen. In the small-cutoff limit this reduces to $r_0 r_c^2 \\approx 2^{4/(d+1)} r_h^3$. The authors also show that unequal left/right cutoffs lead to a symmetric generalization, Eq. (26), and that a half-sided geometry (one exterior side capped) still induces an interior cutoff. The result matches the cutoff relation found earlier in holographic complexity, and the paper points out that the different powers of $r_0$ and $r_c$ in the small-cutoff limit are explained by the interior being built from two copies of exterior modes.","pith_inferences":["One can test the relation beyond the saddle point: loop corrections to the on-shell actions should shift Eq. (24) by powers of $G_N$ (i.e., $1/N$), and measuring that shift would reveal how state-dependent the interior cutoff is.","The same factorization logic suggests a universal principle: any bulk or boundary cutoff that restricts exterior modes will restrict interior modes through the doubling of degrees of freedom, so interior reconstructions in any approach (not just the Papadodimas–Raju one) should exhibit an inherited cutoff.","For the typical-microstate geometry discussed in the paper, the predicted interior cutoff is larger by a factor $2^{2/(d+1)}$ when one exterior side is capped; this could serve as a sharp signature distinguishing a typical microstate from the full eternal black hole.","If $T\\bar{T}$-like deformations are the physical origin of the UV cutoff, then the behind-the-horizon cutoff provides a concrete geometric target: finite-cutoff gravity should exhibit a modified interior geometry with a new boundary at $r_0$, and correlation functions of interior operators should reflect this boundary."],"forward_implications":["A finite UV cutoff at $r_c$ forces a behind-the-horizon cutoff $r_0$ given by Eq. (24); the interior cutoff is not an independent choice.","For small cutoffs, $r_0 r_c^2 \\approx 2^{4/(d+1)} r_h^3$, so moving the boundary cutoff inward pushes the interior cutoff toward the horizon with a different power.","The same interior/exterior factorization predicts how unequal left and right cutoffs determine the interior cutoff, Eq. (26), which complexity computations did not directly address.","The result explains the power mismatch between $r_0$ and $r_c$ in the complexity literature as a consequence of the interior operator being formed from two copies of exterior modes.","If the proportionality $e^{i\\Gamma^{(II)}} \\propto e^{2i\\Gamma^{(I)}}$ survives higher orders, the interior cutoff should be visible in other interior-sensitive quantities, such as commutators or entanglement probes."],"supporting_citations":[{"why":"Supplies the construction of interior operators from two copies of exterior modes, which is the basis for the partition-function relation.","marker":"[13]"},{"why":"Establishes the large-N factorization of generalized free field correlators used to derive $Z^{(II)} \\propto Z^{(I)}Z^{(III)}$.","marker":"[18]"},{"why":"Provides the earlier holographic-complexity derivation whose behind-the-horizon cutoff relation, Eq. (24), is reproduced here.","marker":"[8]"},{"why":"Provides the eternal black hole and thermofield double background on which the on-shell actions are evaluated.","marker":"[19]"},{"why":"Motivates treating a finite radial cutoff as the holographic description of a $T\\bar{T}$-type deformation, the setting in which the cutoff result matters.","marker":"[24]"}],"fun_headline_variants":["Black hole interior cutoff fixed by exterior UV cutoff","Interior modes factorize: two exteriors set the cutoff","Behind-horizon cutoff emerges from exterior modes","Gravity's interior: UV cutoff dictates a far-side cutoff","Holographic complexity match: interior cutoff from exterior"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument assumes without proof that the relation 'interior partition function equals the square of the exterior one,' derived for free scalar fields in the large-$N$ limit, also applies to the gravitational action itself; if the gravitational on-shell action does not factorize this way, Eq. (24) does not follow.","fun_headline_variants_meta":{"raw":{"variants":["Black hole interior cutoff fixed by exterior UV cutoff","Interior modes factorize: two exteriors set the cutoff","Behind-horizon cutoff emerges from exterior modes","Gravity's interior: UV cutoff dictates a far-side cutoff","Holographic complexity match: interior cutoff from exterior"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000261,"raw_usage":{"total_tokens":1616,"prompt_tokens":992,"completion_tokens":624,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":608,"completion_tokens_details":{"reasoning_tokens":546}},"tokens_in":608,"tokens_out":624,"duration_ms":6974,"temperature":1.0,"reasoning_tokens":546,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:26:24.823498+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the one-loop (i.e., $O(1/N)$) graviton partition function in region II with the interior cutoff in place and check whether the exact relation $\\Gamma^{(II)} = 2\\Gamma^{(I)} + \\text{const}$ survives; any non-vanishing $1/N$ correction would modify Eq. (24), and the deviation would show up as a shift in the effective cutoff $r_0$ inferred from interior correlators.","supporting_citations":[],"review_version":1}