{"id":"001c5e07-1bf8-41b4-9bfd-eec9ad08b7d3","arxiv_id":"1908.01800","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"A linearized graviton field in a sphere has entanglement entropy equal to that of two massless scalars with l=0 and l=1 modes removed, giving a logarithmic coefficient of -61/45.","lead":"The paper computes the entanglement entropy of a free massless spin-2 field (linearized gravitons) in a sphere in flat space, finding a universal logarithmic coefficient of -61/45. It reduces the problem to two free scalar fields with the lowest angular momentum modes subtracted, a result that helps clarify how gravitational degrees of freedom localize.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The -61/45 coefficient depends on the l=1 scalar mode having log coefficient exactly 1/6; this is asserted from dimensional analysis plus an undescribed lattice check, so the central number is not independently verifiable as written.","rationale":"I read the paper in good faith: the derivation is detailed, the spherical-harmonic reduction is explicit, and the overall structure leading to two scalar copies with l=0 and l=1 modes subtracted is coherent. The weakest point is exactly the one flagged by the reader: the l=1 scalar mode log coefficient enters the final -61/45 with a multiplicity of twelve, and the published text supports it with a scale-invariance argument plus an unreported numerical check. I do not see an internal inconsistency in the main reduction, and the dimensional argument is reasonably plausible because the entangling surface at r=R sees a negligible potential 2/R^2 for high-energy modes, so the UV coefficient should indeed be the free-scalar one. Nevertheless, as written the central number is not reproducible without the missing lattice computation. That is a genuine but fixable gap, so the appropriate verdict remains conditional: accept only after the l=1 lattice check is supplied or independently confirmed. My stress-test pass therefore does not change the reader's verdict.","tokens_in":22993,"tokens_out":13939,"duration_ms":158550,"concrete_test":"Implement the lattice version of (5.56): H = 1/2 sum_{i>=1}[p_i^2 + (phi_{i+1} - phi_i)^2 + (2/i^2) phi_i^2] with phi_0 = 0, and compute the ground-state entanglement entropy of the first N sites using the standard free-field correlator method (Srednicki/Casini-Huerta). Fit S(N) = a log N + b for N ranging from 10^2 to 10^4, including subleading corrections, and compare a with 1/6 to five digits. If a departs from 1/6 by more than 10^-3, the coefficient -61/45 in (5.59) must be revised accordingly.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The headline universal coefficient (5.59) is obtained in (5.58) by subtracting, for each of the two scalar copies, the l=0 mode and three l=1 modes of the scalar sphere entropy. The l=1 subtraction uses S = (1/6) log(R/epsilon) + const for the Hamiltonian in (5.56), H = 1/2[P^2 + (partial_r phi)^2 + (2/r^2) phi^2]. This coefficient is load-bearing: if the true coefficient were 1/6 + delta, the final coefficient would shift from -61/45 by 12 delta (two scalar copies times six subtracted l=1 modes), so even a 1% error moves the result by 0.12, far above the precision claimed. The text at (5.57) justifies 1/6 by saying the UV divergent piece is the same as for the scalar and the model has no dimensionful scales, and reports a lattice check to five digits without giving the computation. Scale invariance alone fixes the functional form S = a log(R/epsilon) + const, but it does not tie a to the free-scalar value unless one also establishes the UV coefficient; the identification with the free scalar at the entangling surface is plausible but not derived in the text. The missing lattice check is therefore the weakest link in the otherwise coherent reduction of the graviton entropy to scalar modes.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper computes the entanglement entropy of a massless spin-2 field (linearized gravitons) in a sphere in flat Minkowski space. The authors decompose h_μν in tensor spherical harmonics, fix a gauge adapted to spherical symmetry, and show that for each angular momentum l>=2 the two dynamical modes reduce to two independent scalar spherical modes with the same Hamiltonian as a free massless scalar. They then analyze the low-angular-momentum modes and argue that the entropy is equivalent to that of two free scalars with the l=0 and l=1 modes subtracted. The universal logarithmic coefficient is computed as -61/45, Eq. (5.59), and is argued to agree with the mutual-information regularization.","tokens_in":23325,"tokens_out":6597,"duration_ms":76229,"significance":"If the result holds, this is an important universal coefficient for the entanglement entropy of free gravitons on a sphere, extending the known scalar (-1/90) and Maxwell (-16/45) coefficients to spin 2. The paper's methodological contribution is also significant: it shows how to choose a gauge that preserves the localization of the gauge-invariant algebra inside the sphere, which is essential for interpreting the entropy as a physical quantity. The reduction for l>=2 is explicit and detailed, and the treatment of the l=0 and l=1 modes is a useful step. The paper also gives a concrete prediction that is not fitted to the target value. However, two load-bearing points are not fully supported as written: the logarithmic coefficient of the l=1 scalar mode and the locality relations between the gauge-fixed fields and the curvature tensor.","major_comments":[{"comment":"The logarithmic coefficient 1/6 for the l=1 scalar mode is load-bearing for the central result, since Eq. (5.58) subtracts six such modes and the final coefficient -61/45 shifts by 12 delta if the true coefficient is 1/6 + delta. The text justifies (5.57) by scale invariance plus the statement that the UV divergent piece is the same as for a free scalar, and by an undescribed lattice check. Scale invariance alone only fixes the functional form S = a log(R/epsilon) + const; it does not determine a without an independent evaluation of the UV coefficient. The claimed five-digit lattice verification is not described or reproducible. Please provide an explicit derivation of a=1/6 for the Hamiltonian (5.56), or a complete description of the lattice computation and its numerical output, before the final coefficient can be considered established.","section":"Section 5.11, Eq. (5.57)"},{"comment":"The claim that the gauge-fixed fields h_{lm}^{1m} and h_l^{te} can be locally expressed in terms of the gauge-invariant curvature inside the sphere is central to identifying the computed entropy with the entropy of the gauge-invariant algebra. The text states that this was obtained by 'computer based algebraic manipulation' but displays explicit angular functions only for m=0, Eqs. (5.51) and (5.55). For general m the locality relations are asserted without explicit formulas or a demonstration from rotational covariance. Please provide the general formulas or a supplementary derivation, since the physical interpretation of the result and the gauge-choice justification rest on this step.","section":"Section 5.10, Eqs. (5.50)-(5.55)"}],"minor_comments":[{"comment":"In the sentence 'the model = 0 for the scalar' the intended wording is 'the mode l = 0 for the scalar'; please correct this and similar lapses.","section":"Section 3"},{"comment":"There are numerous typographical and language issues, including 'studding' (Introduction), 'hT y ξV' (Section 5.3), 'Laplancians' (Appendix A), and 'cames' (Section 6). A thorough English and typographical editing pass is needed.","section":"Throughout"},{"comment":"The abbreviation 'cons.' for 'constant' is informal; please write 'constant' or define the notation.","section":"Section 5.11, Eq. (5.57)"},{"comment":"The statement that the UV divergence of the mutual information for the l=0,1 modes 'cannot change due to the potential' is a key step but is only discussed qualitatively; even if a full derivation is deferred, the logic would be clearer if this were expanded into an explicit argument.","section":"Section 6"}],"recommendation":"major_revision","confidential_remarks":"The main result is likely correct, but the current manuscript is not self-contained at exactly the point that the headline coefficient depends on: the l=1 mode entropy coefficient is asserted with an unreported lattice check. The authors should be asked to provide the missing derivation or reproducible numerical data. The locality relations in Section 5.10 are likewise abbreviated; a supplementary file or an appendix would resolve this."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe one thing to know: this paper computes the universal logarithmic entanglement coefficient for free linearized gravitons in a sphere and gets -61/45, a new result. The route is a gauge-fixed Hamiltonian reduction to two free scalars with the l=0 and l=1 modes subtracted. It is the first derivation of this coefficient directly from the field Hamiltonian, and it is consistent with Dowker's later thermodynamic computation in de Sitter space.\n\nWhat is genuinely good: the tensor spherical harmonic decomposition is done carefully, the l>=2 modes are explicitly reduced to two scalar Hamiltonians of the form (1/2)(P^2 + (∂_r φ)^2 + l(l+1)/r^2 φ^2), and the gauge fixing is chosen so that the gauge-fixed fields are reconstructible from the curvature inside the sphere, which is what makes the entropy a well-defined quantity for the algebra of gauge-invariant operators. The l=0 and l=1 modes are handled separately and shown to drop out of the graviton Hilbert space. The Maxwell parallel and sphere computations are reviewed and used as a consistency check.\n\nThe soft spot is the one the stress-test flags: the l=1 scalar mode with the 2/r^2 potential is assigned log coefficient 1/6 from a dimensional/UV argument plus a lattice check that is reported to five digits but not shown. That coefficient enters the final answer multiplied by twelve, so the headline -61/45 depends on it. The paper's Section 6 gives a second argument via mutual information: the UV divergence of the mutual information for these 1D modes is -1/3 log(epsilon), independent of the potential, which would fix the entropy log term. That helps, but it is stated rather than demonstrated in detail. For a referee, the key request should be the lattice computation and a cleaner derivation of the 1/6 for the potential mode. A second, minor gap is that the locality functions R_erem and R_emem are written only for m=0; the m-independence of the Hamiltonians makes this a small omission, not a structural one.\n\nNone of this changes my bottom line. The central reduction is coherent, the l=1 coefficient is not fitted to the target, and the independent Dowker agreement is real evidence. This is a solid paper that would benefit from one revision round to fill in the numerical and locality details. It deserves a serious referee, and I would bring it to a reading group focused on entanglement in gauge theories and gravity.\n\nBest,\n[your name]","headline":"New universal log coefficient -61/45 for free gravitons in a sphere, derived from a careful scalar-mode reduction; the main weak point is the l=1 coefficient resting on an undescribed lattice check, but the mutual-information argument and Dowker's independent result make the number credible.","tokens_in":23792,"tokens_out":4342,"would_cite":true,"duration_ms":46546,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The entanglement entropy of free gravitons in a sphere is that of two free massless scalars with the l=0 and l=1 modes removed, giving a universal logarithmic coefficient of -61/45.","keywords":["entanglement entropy","graviton","spin-2 field","tensor spherical harmonics","gauge fixing","logarithmic coefficient","sphere","mutual information"],"falsifier":"Compute the entanglement entropy of the l=1 scalar Hamiltonian $H=\\frac{1}{2}[P^2+(\\partial_r\\varphi)^2+\\frac{2}{r^2}\\varphi^2]$ on the half-line $r\\in(0,R)$ by exact lattice diagonalization for several lattice spacings; if the coefficient of $\\log(R/\\epsilon)$ is not $1/6$, the final $-61/45$ shifts by six times the deviation.","tokens_in":22815,"feed_emoji":"⚛️","tokens_out":8594,"duration_ms":78458,"temperature":0.7,"pith_summary":"The paper computes the entanglement entropy of a free massless spin-2 field, the linearized graviton, in a spherical region of flat Minkowski space. It claims that, after a gauge choice adapted to the sphere, the entropy is exactly that of two free massless scalar fields with the l=0 and l=1 angular-momentum modes removed. The universal coefficient of the logarithmic term follows as -61/45, and the same value comes out of a mutual-information regularization of the entropy. This matters because it gives a concrete, unambiguous entanglement-entropy prediction for a free graviton theory in flat space, a benchmark that can be compared with holographic or black-hole entropy formulas.","feed_headline":"Graviton sphere entropy equals two scalars, minus l=0,1","feed_subtitle":"A gauge-fixing argument pins the universal logarithmic term -61/45 for massless spin-2 fields.","key_machinery":"The central object is the decomposition of the metric perturbation $h_{\\mu\\nu}$ into tensor spherical harmonics, followed by a radial gauge fixing that reduces each angular momentum sector to two independent Hamiltonians of the scalar-spherical form $H = \\frac{1}{2}[P^2 + (\\partial_r\\varphi)^2 + \\frac{l(l+1)}{r^2}\\varphi^2]$. The gauge parameters are fixed ($\\alpha=0$, $\\gamma=-\\beta/\\sqrt{(l-1)(l+2)}$) so that the gauge-fixed fields can be recovered from the linearized curvature tensor by relations with no radial derivatives, ensuring that the operator algebra inside the sphere matches the algebra of gauge-invariant operators. The l=0 and l=1 sectors are then analyzed separately; they drop out of the entropy, so the final answer is two scalar fields with the l=0 and l=1 modes subtracted.","core_discovery":"On the paper's own terms, the discovery is that vacuum entanglement of linearized gravitons across a sphere does not need any new machinery beyond the scalar spherical modes: the graviton entropy equals the entropy of two free massless scalars in the same sphere with the l=0 and l=1 sectors subtracted. The l=0 and l=1 graviton modes turn out to be non-dynamical or to cancel, while each l≥2 angular momentum contributes two scalar modes with the usual $\\frac{l(l+1)}{r^2}$ potential. Using the known scalar sphere coefficient $-1/90$ and the half-line scalar coefficient $1/6$ for the subtracted modes, the paper obtains $2\\times(-1/90) - 2\\times(1/6) - 6\\times(1/6) = -61/45$ for the coefficient of $\\log(R/\\epsilon)$. The calculation is done in real time with the gauge-fixed metric perturbation $h_{\\mu\\nu}$, in a gauge chosen so that the fields inside the sphere generate the same algebra as the curvature tensor localized there.","pith_inferences":["The same gauge-localization logic could be applied to a graviton in a region of arbitrary shape; a natural test is whether non-spherical boundaries change which low angular modes are subtracted, or whether the subtraction pattern is tied to the sphere's symmetry.","The paper's l=1 scalar mode coefficient $1/6$ could be verified by an independent numerical diagonalization of the one-dimensional Hamiltonian with the $2/r^2$ potential; a deviation would shift every helicity coefficient by six times the error.","Because the result is regulator-independent and tied to mutual information, it offers a clean benchmark for holographic calculations of graviton entanglement, where the same logarithmic coefficient might be extracted from the dual theory.","The pattern 'two scalars minus low-l modes' suggests a general rule for free higher-spin fields on the sphere that could be tested mode by mode with the same harmonic decomposition."],"forward_implications":["If the calculation is right, the universal logarithmic coefficient for a massless spin-2 field in a sphere in flat space is $-61/45$, independent of the short-distance regulator.","The graviton sphere entropy is the same as a Maxwell field's entropy with the $l=1$ mode contribution removed, since Maxwell is already two scalars minus the $l=0$ mode.","The coefficient computed from the entropy coincides with the one obtained from the mutual-information-regulated entropy $S_\\epsilon(A)=\\frac{1}{2}I(A_+,A_-)$, so the result is insensitive to center-term or edge-mode ambiguities.","The paper's conjecture for higher helicity $h>2$ is that the coefficient becomes $-(1+15h^2)/45$, from subtracting the $l=0,\\dots,h-1$ scalar modes.","For parallel planes, the graviton entropy reduces to two scalar fields and shares the same universal coefficient as the Maxwell field."],"supporting_citations":[{"why":"Supplies the Maxwell-field template: gauge-fixed field reduced to two scalar spherical modes, with the l=0 subtraction and its 1/6 log coefficient.","marker":"[10]"},{"why":"Gives the -1/90 universal logarithmic coefficient for a free massless scalar in a sphere.","marker":"[11]"},{"why":"Establishes the localization criterion that gauge-fixed fields inside a region must be recoverable from gauge-invariant operators, used to fix the graviton gauge.","marker":"[17]"},{"why":"Provides the original spherical reduction of a massless scalar to radial modes, the model for the graviton mode Hamiltonians.","marker":"[20]"},{"why":"Supplies the 1/6 log coefficient for a two-dimensional massless scalar on a half-line, used for the l=0 and l=1 subtracted modes.","marker":"[25]"},{"why":"Defines the mutual-information regularization used to identify the entropy unambiguously and to confirm the coefficient.","marker":"[31]"}],"fun_headline_variants":["Graviton sphere entropy: two scalars minus l=0,1","Spin-2 sphere entropy: two scalars minus l=0,1","Universal log term -61/45 for graviton sphere entropy","Graviton entanglement in sphere: -61/45 log coefficient","Massless spin-2 sphere entropy: scalar subtraction gives -61/45"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the l=1 scalar mode, a one-dimensional field with an inverse-square potential $2/r^2$, has the same logarithmic entropy coefficient $1/6$ as the plain massless scalar; the paper asserts this from dimensional analysis and a lattice check that is not shown in the text.","fun_headline_variants_meta":{"raw":{"variants":["Graviton sphere entropy: two scalars minus l=0,1","Spin-2 sphere entropy: two scalars minus l=0,1","Universal log term -61/45 for graviton sphere entropy","Graviton entanglement in sphere: -61/45 log coefficient","Massless spin-2 sphere entropy: scalar subtraction gives -61/45"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000947,"raw_usage":{"total_tokens":4027,"prompt_tokens":916,"completion_tokens":3111,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":532,"completion_tokens_details":{"reasoning_tokens":3015}},"tokens_in":532,"tokens_out":3111,"duration_ms":20305,"temperature":1.0,"reasoning_tokens":3015,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:02:58.613154+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the entanglement entropy of the l=1 scalar Hamiltonian $H=\\frac{1}{2}[P^2+(\\partial_r\\varphi)^2+\\frac{2}{r^2}\\varphi^2]$ on the half-line $r\\in(0,R)$ by exact lattice diagonalization for several lattice spacings; if the coefficient of $\\log(R/\\epsilon)$ is not $1/6$, the final $-61/45$ shifts by six times the deviation.","supporting_citations":[{"cited_title":"Entanglement entropy of a Maxwell field on the sphere","cited_arxiv_id":"1512.06182","evidence_quote":"Supplies the Maxwell-field template: gauge-fixed field reduced to two scalar spherical modes, with the l=0 subtraction and its 1/6 log coefficient."}],"review_version":1}