{"id":"0471cabd-a2a5-4549-a6b8-c045f9889595","arxiv_id":"1301.1597","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"Strong-coupling localization in N=2* theory reproduces the Pilch-Warner supergravity Coulomb-branch vacuum and the perimeter law for large Wilson loops, confirming holography in a non-conformal setting.","lead":"The paper tests holographic duality for N=2* theory, a massive deformation of N=4 super Yang-Mills, by computing the Coulomb-branch eigenvalue distribution, large Wilson loops, and the free energy two independent ways: a localization matrix model at strong coupling and the known Pilch-Warner supergravity background. The two agree on the semicircular vacuum width, the perimeter law, and the leading free energy, giving a precision check of holography away from conformal Sym.","discovery_kind":"unclear","skeptic_critique":{"model":"selfhosted/deepseek-v4.1-flash","headline":"The kernel approximation (2.9) is the load-bearing step; it is self-consistently justified by mu/M = sqrt(lambda)/(2 pi) at large lambda, but the paper does not quantify the neglected O(1/lambda) terms, and the numerical tests asserted in footnote 1 are not shown.","rationale":"The paper's central claim is a leading-order match between localization and supergravity for the PW vacuum width and the Wilson loop perimeter law. The match is supported by two independent computations, and the supergravity side (3.9)-(3.10) is a straightforward minimal-surface calculation that I can verify. The localization side relies on the kernel approximation (2.9), which is self-consistently justified by mu/M = sqrt(lambda)/(2 pi) >> 1 at large lambda. The approximation is controlled in the dominant region |x-y| ~ mu, and the near-diagonal region contributes negligibly because the correction is odd. The main caveat is that the paper does not quantify the neglected O(1/lambda) terms, and the numerical tests in footnote 1 are not shown. This is a real but minor gap: the parametric argument is strong, and the leading result is robust. The free-energy constant bookkeeping between (2.17) and (2.18) is also not fully transparent, but the free energy is a separate prediction, not part of the holographic test. The abstract's 'exact match' is a bit strong for a leading-order match, but the body is clear. Overall, the reader's CONDITIONAL verdict is reasonable, and I would not change it.","tokens_in":10894,"tokens_out":18262,"duration_ms":364980,"concrete_test":"Solve the full saddle-point integral equation (2.8) numerically for N=2* on S^4 at large but finite lambda (e.g., lambda = 100, 400, 1600) with fixed M R, and extract the eigenvalue density width mu and the Wilson loop ln W. Compare to the analytic predictions (2.11) and (2.14). If the relative deviation scales as 1/lambda, the kernel approximation (2.9) is confirmed; if it scales differently or is larger than O(1/lambda), the approximation is missing a contribution and the leading match is less secure.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The derivation of the PW width (2.11) and the Wilson loop (2.14) rests on replacing the exact kernel combination in (2.8) by (1+M^2)/(x-y) via (2.9). This is an expansion in M/mu, and the paper argues self-consistently that mu/M = sqrt(lambda)/(2 pi) >> 1. The argument is sound in the dominant region |x-y| ~ mu, and the near-diagonal region contributes negligibly because the correction is odd and the region is small. However, the paper does not quantify the size of the neglected terms: the next correction is O(M^2/mu^2) = O(1/lambda) relative to the leading term, but this is asserted rather than demonstrated. Footnote 1 claims 'extensive numerical tests' but no numerics, code, or parameter files are shown, so the reader cannot independently verify the size of the neglected terms. This is a real gap, though not a fatal one: the parametric argument is strong, and the leading result is robust. A concrete check would be to solve the full integral equation (2.8) numerically at large but finite lambda and compare the resulting mu and ln W to (2.11) and (2.14), quantifying the deviation as a function of lambda.","agreement_with_reader":"agree"},"referee_report":{"model":"selfhosted/deepseek-v4.1-flash","summary":"The manuscript studies N=2* SU(N) gauge theory on a four-sphere using Pestun localization. At large N and strong 't Hooft coupling the matrix model (2.1) is solved in a saddle-point approximation, and the eigenvalue density is found to be semicircular with width μ = √(λ(1+M²))/(2π) (Eq. 2.11), which in the decompactification limit R→∞ becomes √λ M/(2π), matching the Pilch-Warner supergravity prediction (Eq. 1.4). The same saddle gives ln W(C_circle) = √(λ(1+M²)) + O(λ⁰) (Eq. 2.14), i.e. a perimeter law ln W = √λ M R, which the authors match to the minimal-surface computation in the Pilch-Warner geometry (Eq. 3.10). The paper also computes the free energy at large λ (Eqs. 2.16–2.19) and, in Appendix A, verifies the perimeter law for a large circular loop in the Pilch-Warner background. The central claim is that localization dynamically selects the Pilch-Warner Coulomb-branch vacuum and reproduces the string-theory Wilson-loop result.","tokens_in":11192,"tokens_out":7714,"duration_ms":150045,"significance":"If correct, this is a clean and genuinely non-trivial test of holography in a non-conformal, half-supersymmetric setting. The two sides of the comparison are independent: the field-theory side uses the Pestun matrix model (2.1) and a strong-coupling saddle, while the gravity side uses the Pilch-Warner metric (3.1) and a minimal-surface area. The agreement of the width (1.4) and of the perimeter-law coefficient (3.10) is a parameter-free, falsifiable prediction, and the free-energy result (2.18) is a useful bonus. The appendix's treatment of the circular loop, including the scaling function f(x) and the 1/a correction (A.16), is a careful piece of work. The paper deserves credit for making the comparison explicit rather than qualitative.","major_comments":[{"comment":"The kernel approximation (2.9), replacing ½K(x+M)+½K(x−M)−K(x) by M²/x, is the load-bearing step of the localization derivation, and footnote 1 asserts 'an extensive set of numerical tests' without showing any numerics, code, or parameter values. The approximation is self-consistently justified by μ/M = √λ/(2π) ≫ 1, but the size of the neglected O(M²/μ²) = O(1/λ) correction is asserted rather than demonstrated. A concrete check would be to solve the full integral equation (2.8) numerically at large but finite λ and compare the resulting μ and ln W to (2.11) and (2.14), quantifying the deviation as a function of λ; a figure or table would make the claim verifiable. It would also help to state explicitly that K''(x) ≈ 2/x is the large-x asymptotic, valid in the region |x−y| ~ μ ≫ 1 that dominates the integral, and to argue that the near-diagonal region where it fails is subleading.","section":"§2, Eq. (2.9) and footnote 1","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The abstract (abstract:s2) and §2 describe the result as a 'dynamical selection' of the Pilch-Warner Coulomb-branch vacuum, but the paper exhibits a single one-cut semicircular saddle and does not compare its free energy against competing saddles (multi-cut, or the unbroken phase). 'Selection' implies a comparison that is not performed. Either soften the wording to 'the strong-coupling saddle reproduces the Pilch-Warner distribution' or supply the free-energy comparison that would justify the stronger claim.","section":"§2 and abstract:s2, 'dynamical selection'","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The free-energy chain (2.16)→(2.17)→(2.18) uses '≈' at each step and drops O(M³) and non-logarithmic terms; the M=0 normalization is fixed by matching to the N=4 result, but the constant in (2.18) should be displayed and checked against the known N=4 S⁴ free energy. Eq. (2.19) is explicitly only leading-log, so the abstract's unqualified 'compute free energy at large 't Hooft coupling' (abstract:s4) should be qualified to match. Showing the intermediate integration that produces (2.18) from (2.17) would remove the ambiguity.","section":"§2, Eqs. (2.16)–(2.19)","authors_acknowledge":null,"authors_acknowledge_p":null}],"minor_comments":[{"comment":"The phrase 'exact match' should be qualified: the agreement holds at leading order in 1/λ and in the decompactification limit R→∞, with O(λ⁰) corrections acknowledged in (2.14). 'Leading-order match' would be more accurate.","section":"Abstract, abstract:s3","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The coefficient 9/32 in S_ren = −√λ a − 9√λ ln a/(32a) is quoted while the text claims only 'logarithmic accuracy' and states that the finite parts are not computed. Clarify whether 9/32 is derived from the matching or asserted, and note explicitly which finite parts remain undetermined.","section":"Appendix A, Eq. (A.16)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The reinstatement of R-dependence from (2.11) to (2.12) should be shown in one line (M→MR and the 1/R² term) so the reader can verify that the decompactification limit reproduces (1.4).","section":"§2, Eq. (2.12)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The text says the strong-coupling solution is 'a slight modification of the small-M limit considered in [7]'; it would help to state precisely what is new here relative to [7,8] and which results are being imported.","section":"§2, relation to [7,8]","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"There are typographical issues to fix: 'Yang-Mill' in the abstract, 'enhan¸ con' in §3 and Appendix A (presumably 'enhanced/conformal'), and inconsistent spacing in the displayed equations. These do not affect the physics but should be cleaned up.","section":"Presentation","authors_acknowledge":null,"authors_acknowledge_p":null}],"recommendation":"minor_revision","confidential_remarks":"This is a short, correct, and useful letter-style contribution; the central comparison is sound and the two sides are independent. The main thing an editor may wish to see verified is the footnote-1 numerics claim, which is currently unshown, and a slightly more careful presentation of the free-energy chain. Neither is likely to change the conclusions."},"author_rebuttal":null,"desk_editor":{"model":"selfhosted/deepseek-v4.1-flash","letter":"Worth your time, and short enough to read in one sitting. The real content is that the Pestun matrix model for N=2* on S4, taken at large N and strong 't Hooft coupling, dynamically lands on the Pilch-Warner Coulomb branch: semicircular density with width mu = sqrt(lambda(1+M^2))/(2 pi), which at R -> infinity is exactly the supergravity value sqrt(lambda) M/(2 pi). The same saddle gives log W = sqrt(lambda) M R for the circular loop, matching the minimal-surface area in the PW geometry, and a leading free energy F = -N^2 M^2 R^2 ln(MR). The two sides really are independent — one uses the localization matrix model, the other the ten-dimensional PW metric — so this is a legitimate test, not a consistency check dressed up as one. The kernel expansion in (2.9) is the load-bearing step, and it is self-consistently justified: mu/M = sqrt(lambda)/(2 pi), so the neglected terms are O(1/lambda). That is a parametric argument, not a numerical one, and it holds. The appendix on the circular loop is careful work; the 9 ln a/(32 a) correction is a nice touch and the matching-scale cancellation between (A.9) and (A.15) checks out on inspection. On the stress-test note: the missing numerics behind footnote 1 are a presentational gap, not a hole in the argument. The authors flag it themselves, and the parametric control is strong enough that I would not hold the paper for it. Two genuinely minor things. First, the constant bookkeeping between (2.17) and (2.18) is not quite the naive integral — there is an extra O(N^2 M^2) piece absorbed into the M=0 normalization — harmless for (2.19) but the text glides past it. Second, 'exact match' in the abstract is doing some work: the agreement is leading order in 1/lambda and leading order in 1/(MR), with subleading terms not compared, which the authors know perfectly well since they say so in the body. Neither is worth a referee round on its own. This is for anyone doing precision holography beyond N=4, or using localization on curved backgrounds with mass deformations. Send it out; the referees will have an easy job and the paper will be cited.","headline":"A clean, short, and genuinely two-sided check of holography in N=2*: localization reproduces the Pilch-Warner vacuum width, the perimeter law, and the leading free energy, with only cosmetic soft spots.","tokens_in":11829,"tokens_out":1520,"would_cite":true,"duration_ms":31197,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"pith_extraction":{"msc":[],"pacs":["11.25.Tq","11.15.Pg","11.30.Pb"],"model":"selfhosted/deepseek-v4.1-flash","headline":"A localization computation at large $N$ and strong 't Hooft coupling reproduces, in $N=2^*$ gauge theory, both the Coulomb-branch eigenvalue distribution and the Wilson-loop expectation value predicted by the Pilch-Warner supergravity dual.","keywords":["N=2* gauge theory","localization","holographic duality","Wilson loops","perimeter law","Coulomb branch","Pilch-Warner supergravity","matrix models"],"falsifier":"Solve the full saddle-point equation $2$ numerically at finite $2$ and $M$, without the kernel expansion, and compare the resulting eigenvalue density and $2$ against the semicircle of width $2$; a deviation that persists as $2$ grows would break the claimed exact match with supergravity.","tokens_in":12200,"feed_emoji":"🧲","tokens_out":3346,"duration_ms":67624,"temperature":0.7,"pith_summary":"The paper asks whether holography survives when conformal symmetry and half the supersymmetry are broken. $N=2^*$ theory is the right testing ground: it is $N=4$ super-Yang-Mills with a common mass $M$ for four scalars and half the fermions, so it has an explicitly known ten-dimensional supergravity dual (the Pilch-Warner background) and is also solvable at any coupling through Pestun's localization. The authors evaluate the resulting matrix model at large $N$ and strong coupling and find that the saddle point dynamically lands on the Coulomb-branch vacuum that supergravity singles out: a semicircular eigenvalue density whose width is $2$ times the strong-coupling value $2$ times $2$ times, in the decompactified limit, exactly $2$ times the supergravity width. The same saddle point gives a Wilson loop that grows with the perimeter rather than the area, with a coefficient that matches the minimal-surface computation in the deformed geometry. If the authors are right, holography passes an exact, non-conformal test with no free parameters on either side.","feed_headline":"Field theory picks the supergravity vacuum, exactly","feed_subtitle":"A strong-coupling matrix model reproduces the Pilch-Warner eigenvalue width and the perimeter law for large Wilson loops.","key_machinery":"The carrying object is the localization matrix integral on $S^4$, whose integrand is a product of pairwise interactions built from a single function $H(x)$ (a regularized infinite product) plus a Gaussian weight $2$. Its large-$N$ saddle-point equation is a singular integral equation for the eigenvalue density, with kernel $2$. The load-bearing step is the Taylor expansion $2$, which converts the equation into the standard Wigner form solved by a semicircle; the Wilson loop is then the exponential average $2$, dominated by the largest eigenvalue. On the gravity side the corresponding machinery is the Nambu-Goto action for a string ending on the contour in the warped metric $2$, renormalized by subtracting the perimeter divergence.","core_discovery":"The paper's central claim is that the localization matrix model of $N=2^*$ theory, taken at large $N$ and large 't Hooft coupling, reproduces two independent supergravity predictions for the Pilch-Warner background. First, the saddle-point eigenvalue density is Wigner's semicircle with width $2$ in units of the four-sphere radius, $2$, which becomes $2$ when the sphere is decompactified — the same width read off from the supergravity solution. Second, the same saddle point gives $2$, so that at large radius $2$: a perimeter law, not an area law, whose coefficient matches the renormalized minimal-surface area in the Pilch-Warner metric, where the complicated factor $2$ cancels between the induced metric and the dilaton. The authors also compute the free energy at strong coupling, $2$, which at large radius behaves as $2$.","pith_inferences":["Because the reduction to a Gaussian matrix model only uses the semicircular form of the kernel at strong coupling, the same route should apply to other $2$ quiver theories whose gravity duals exhibit semicircular eigenvalue densities; checking whether the free-energy formula survives there is a concrete next test.","The finite-$R$ results are predictions for a half-BPS supergravity solution whose boundary is $S^4$ rather than $2$; constructing that solution would independently test the $2$ shift in the eigenvalue width.","The subleading correction to the circular loop quoted in the appendix, $2$, is a sharp target: a numerical string computation at large but finite loop radius, or a localization computation at finite $2$, could confirm or contradict it."],"forward_implications":["At strong coupling the eigenvalue density is semicircular with $2$, so on $2$ the symmetry-breaking scale is $2$, parametrically larger than the deformation mass $M$ by a factor of order $2$.","Large Wilson loops obey $2$ rather than an area law, in both the localization computation and the minimal-surface computation, with coinciding coefficients.","The strong-coupling free energy is $2$, reducing to $2$ at large radius with logarithmic accuracy.","The agreement holds at finite four-sphere radius $R$ as well, so the results $2$, $2$ and $2$ are predictions that go beyond the decompactified supergravity limit."],"supporting_citations":[{"why":"Supplies the localization formula that reduces the $S^4$ partition function and Wilson loop to the matrix integral the whole argument is built on.","marker":"[2]"},{"why":"Provides the Pilch-Warner supergravity background whose metric, dilaton and Coulomb-branch eigenvalue density are the comparison target.","marker":"[1]"},{"why":"Identifies the semicircular eigenvalue distribution of width $2$ as the Coulomb branch of the Pilch-Warner solution, fixing the supergravity side of the match.","marker":"[4]"},{"why":"Establishes the $N=4$ benchmark, where the Gaussian matrix model gives $2$ and matches the classical string area, the limit the present calculation deforms.","marker":"[6]"},{"why":"Supplies the prescription equating the Wilson loop expectation value to the renormalized on-shell string action used in the supergravity computation.","marker":"[10]"},{"why":"Provides the small-$M$ strong-coupling treatment of the same matrix model that the present kernel expansion modifies.","marker":"[7]"},{"why":"Develops the large-$N$ limit of $N=2$ localization matrix models, the technical setting for the saddle-point equation used here.","marker":"[8]"},{"why":"Justifies the perimeter-divergence subtraction used to renormalize the minimal surface in the deformed geometry.","marker":"[11]"}],"fun_headline_variants":["Localization reproduces supergravity for N=2* Wilson loops","Large N=2* Wilson loops obey a perimeter law","Matrix model selects the Pilch-Warner vacuum at large N","N=2* free energy computed at large 't Hooft coupling","Semicircle eigenvalue width matches the Pilch-Warner geometry"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The derivation replaces the exact interaction kernel by its second-order Taylor term, $2$, which is legitimate only when the symmetry-breaking scale $2$ is much larger than the deformation mass $M$; the paper assumes this before solving, confirms it only afterwards from its own answer, and supports it with numerical tests that are asserted in a footnote but not displayed.","fun_headline_variants_meta":{"raw":{"variants":["Localization reproduces supergravity for N=2* Wilson loops","Large N=2* Wilson loops obey a perimeter law","Matrix model selects the Pilch-Warner vacuum at large N","N=2* free energy computed at large 't Hooft coupling","Semicircle eigenvalue width matches the Pilch-Warner geometry"]},"model":"selfhosted/deepseek-v4.1-flash","effort":"low","attempts":1,"cost_usd":0.0,"raw_usage":{"total_tokens":1172,"prompt_tokens":827,"completion_tokens":345,"prompt_tokens_details":null,"completion_tokens_details":{"reasoning_tokens":263}},"tokens_in":827,"tokens_out":345,"duration_ms":7615,"image_pages":[],"temperature":1.0,"reasoning_budget":50,"reasoning_tokens":263,"transport_provider":"selfhosted","cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-11-10T13:03:30.404622+00:00","model_set":{"reader":"selfhosted/deepseek-v4.1-flash"},"falsifier":"Solve the full saddle-point equation $2$ numerically at finite $2$ and $M$, without the kernel expansion, and compare the resulting eigenvalue density and $2$ against the semicircle of width $2$; a deviation that persists as $2$ grows would break the claimed exact match with supergravity.","supporting_citations":[{"cited_title":"Localization of gauge theory on a four-sphere and supersymmetric Wilson loops","cited_arxiv_id":"0712.2824","evidence_quote":"Supplies the localization formula that reduces the $S^4$ partition function and Wilson loop to the matrix integral the whole argument is built on."},{"cited_title":"N=2 Supersymmetric RG Flows and the IIB Dilaton","cited_arxiv_id":"hep-th/0004063","evidence_quote":"Provides the Pilch-Warner supergravity background whose metric, dilaton and Coulomb-branch eigenvalue density are the comparison target."},{"cited_title":"Gauge Dual and Noncommutative Extension of an N=2 Supergravity Solution","cited_arxiv_id":"hep-th/0008076","evidence_quote":"Identifies the semicircular eigenvalue distribution of width $2$ as the Coulomb branch of the Pilch-Warner solution, fixing the supergravity side of the match."},{"cited_title":"Wilson Loops in N=4 Supersymmetric Yang--Mills Theory","cited_arxiv_id":"hep-th/0003055","evidence_quote":"Establishes the $N=4$ benchmark, where the Gaussian matrix model gives $2$ and matches the classical string area, the limit the present calculation deforms."},{"cited_title":"Wilson loops in large N field theories","cited_arxiv_id":"hep-th/9803002","evidence_quote":"Supplies the prescription equating the Wilson loop expectation value to the renormalized on-shell string action used in the supergravity computation."},{"cited_title":"A note on perturbation series in supersymmetric gauge theories","cited_arxiv_id":"1203.5061","evidence_quote":"Provides the small-$M$ strong-coupling treatment of the same matrix model that the present kernel expansion modifies."},{"cited_title":"Large N Limit of N=2 SU(N) Gauge Theories from Localization","cited_arxiv_id":"1207.3806","evidence_quote":"Develops the large-$N$ limit of $N=2$ localization matrix models, the technical setting for the saddle-point equation used here."},{"cited_title":"Wilson Loops and Minimal Surfaces","cited_arxiv_id":"hep-th/9904191","evidence_quote":"Justifies the perimeter-divergence subtraction used to renormalize the minimal surface in the deformed geometry."}],"review_version":1}