{"id":"b53e9475-2806-4bb2-bcb2-01fca4dc4dd2","arxiv_id":"2607.25054","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":2,"one_line_summary":"One-loop AdS supergravity correction to maximal giant graviton–supergraviton correlators is reconstructed in closed Mellin and position form via defect unitarity, including leading-twist anomalous dimensions.","lead":"The authors compute the first quantum-gravity (one-loop) correction to a four-point correlator of two maximal giant gravitons and two light supergravitons in strongly coupled N=4 SYM. The result is given in closed form in Mellin and position space and yields a quantum correction to bound-state binding energies in AdS.","discovery_kind":"extension","skeptic_critique":{"model":"moonshotai/kimi-k3","headline":"The no-single-pole ansatz (4.1) is shown to be sufficient, not necessary: the subleading-log matching validates the constant-c_mn solution but never demonstrates uniqueness against an enlarged ansatz containing single-pole terms.","rationale":"The reader correctly located the load-bearing assumption; I sharpen the nature of the gap. The paper's a posteriori justification (subleading-log matching plus the flat-space Feynman-integral check) establishes that the constant-c_mn simultaneous-pole ansatz is consistent with all available data — an unusually strong sufficiency test, reinforced by the overconstrained Mellin/position-space match (7 residues R_i,j, each with several independent log-coefficients, against 6 unknowns) and by the independent flat-space differential equations (4.34)–(4.35). What is missing is only a necessity/uniqueness argument: single-pole terms were excluded by fiat rather than by fitting and finding zero. This is a real but narrow gap, because (i) the log-degree structure from the OPE already rules out the wider class of non-constant numerators, leaving only constant-residue single poles; (ii) any such sector would have to conspire to vanish in the logB·log²D leading data while hiding entirely in subleading logs that are already matched; and (iii) the flat-space asymptotics would generically detect it. The other candidate concerns do not rise to load-bearing status: the pattern-recognition origin of (3.21), (3.36), and (4.4) is standard for this program and is cross-validated by the Mellin/position-space/flat-space triangle; the regularization ambiguity of the double sum is argued to be a constant with a clear counterterm interpretation, consistent with the flat-space constraint; and the c0/H^(1) scheme ambiguities are stated honestly and lie outside the claimed result. The proposed refit test is concrete, uses the paper's own machinery, and would settle the issue either way. Since the reader's CONDITIONAL already prices in exactly this assumption and the residual risk appears low, I recommend no change to the verdict.","tokens_in":32200,"tokens_out":4850,"duration_ms":127017,"concrete_test":"Enlarge the ansatz to fM22^(2) = Σ c_mn/[(δ+n)(γ−3−2m)] + Σ_{n≤N} a_n/(δ+n) + Σ_{m≤M} b_m/(γ−3−2m) with, say, N=M=6, and re-run the residue matching of §4.1–4.2 against the full LLS coefficient functions of §3.2 (logB, logD, log²D, logB·logD — not only logB·log²D), refitting c_mn simultaneously. If the linear system forces a_n = b_m = 0 and returns (4.4), the uniqueness gap closes and the result stands; any nonzero solution indicates a missing single-pole sector that would shift the residues R_i,j (B.3)–(B.9) and hence (5.6) and (5.16).","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 4, step 2 assumes the reduced one-loop Mellin amplitude contains only simultaneous simple poles with constant numerators, fM22^(2) = Σ c_mn/[(δ+n)(γ−3−2m)]. The justification offered (step 4 and Fig. 3) is that once c_mn are fixed from the logB·log²D coefficient, the ansatz reproduces the full subleading LLS (log²D, logB·logD, logB) obtained in §3.2. This proves the ansatz works; it does not prove it is unique. A single pole a_n/(δ+n) with constant residue contributes to logB×log²D and logB×logD at fixed n for all m, via the double poles of Γ(γ−δ+1)Γ((3−γ)/2)²; similarly b_m/(γ−3−2m) contributes to log²D and logD at all n. Nothing in the paper's fitting procedure excludes a second solution with nonzero {a_n}, {b_m} and correspondingly shifted c_mn, because single poles were never included as free parameters in the fit. The gap matters because c_mn (4.4) was obtained by pattern-matching low-lying slices, and six of the seven unfixed position-space coefficients in §5.2 were then fixed by matching to this Mellin result — so an undetected single-pole sector would propagate coherently into H22^(2) (5.6) and the anomalous dimensions (5.16), rather than being caught by an independent cross-check. Mitigations: the OPE log-degree cap (max logU·log²V, from defect anomalous dimensions at O(c^{-1/2}) and bulk at O(1/c)) already excludes non-constant numerators of degree ≥1 in γ or δ, so the residual freedom is only the finite class of constant-residue single poles; and the flat-space limit tests the large-m,n asymptotics of the total amplitude, which a tuned single-pole sector would likely disturb. The concern is therefore narrow and checkable, not structural.","agreement_with_reader":"agree"},"referee_report":{"model":"moonshotai/kimi-k3","summary":"The paper computes the leading O(1/c) one-loop correction to the correlator of two maximal giant gravitons and two stress-tensor multiplet operators in strongly coupled N=4 SYM, treating the giant gravitons as a zero-dimensional defect. Using the defect AdS unitarity method of [18], the authors glue lower-order OPE data into bulk- and defect-channel leading logarithmic singularities, resum them in closed form (3.21), (3.36), and reconstruct the full reduced correlator H22^(2) in two ways: a Mellin-space ansatz with only simultaneous simple poles whose constant numerators c_mn are fixed to a closed 3F2 form (4.4), and a position-space bootstrap over an eight-function transcendental basis (5.1), (5.4), yielding H22^(2) = Δ̂^(4)P^(2) + ½H22^(1) with explicit coefficients (5.9). The flat-space limit reproduces a single one-loop Feynman diagram off the defect, and the one-loop anomalous dimensions of leading-twist defect-channel operators are extracted (5.16). The answer is determined up to a contact counterterm c0 and an additive multiple of the tree-level correlator.","tokens_in":32718,"tokens_out":4521,"duration_ms":36755,"significance":"If the result holds, this is a substantial advance: the first quantum-gravity (one-loop) correction to a correlator of maximal giant gravitons — heavy operators that retain non-planar physics at N=∞ — computed in closed form in both Mellin and position space. Specific strengths: the reduced amplitude admits the closed simultaneous-pole form (4.1) with coefficients given explicitly by 3F2 hypergeometrics (4.4); the position-space answer takes the compact form (5.6) with a bounded weight-≤3 transcendental basis and a newly identified letter alphabet {z,z̄,1−z,1−z̄,V,1−V}, answering the open position-space question left by [18]; the flat-space limit is shown to agree with a single one-loop Feynman diagram (§4.3); and the construction is validated by multiple internal cross-checks (bulk/defect LLS unmixing, overconstrained Mellin–position-space matching, agreement of the logU·log²V coefficient between channels). The extracted leading-twist defect anomalous dimensions (5.16) are concrete, falsifiable CFT data. The result is also a proof of principle that defect unitarity works for a zero-dimensional defect, the case relevant for light-light-heavy-heavy correlators generally.","major_comments":[{"comment":"§4, ansatz (4.1) and step 2: the reduced one-loop Mellin amplitude is assumed to contain only simultaneous simple poles with constant numerators c_mn, with no single-pole terms a_n/(δ+n) or b_m/(γ−3−2m). The manuscript demonstrates sufficiency, not necessity: the c_mn are extracted from the logB·log²D coefficient under this assumption, and the subsequent reproduction of the subleading LLS (log²D, logB·logD, logB) in §4.2 validates the solution found but does not exclude a second solution with nonzero single-pole residues and shifted c_mn — such terms were never included as free parameters in the fit, and constant-residue single poles are compatible with the OPE log-degree cap of Fig. 2. The gap propagates: six of the seven unfixed position-space coefficients in §5.2 are fixed by matching to the Mellin residues (B.3)–(B.9), so the position-space result (5.6) and the anomalous dimensions (","section":"§4, Eq. (4.1), steps 2–4"},{"comment":"§4.2, Eqs. (4.11)–(4.13): the regularization of the divergent double sum introduces a regulator d_mn and a constant shift C, and it is asserted that the residual scheme ambiguity is exactly one constant, interpreted as the one-loop contact counterterm. This needs a justification. A priori, two admissible regularizations of a doubly-divergent sum can differ by finite terms that are regular but non-constant in (δ,γ) (e.g., derivative-contact-like contributions); the manuscript should argue either from the finite set of available one-loop defect counterterms for this correlator, or from the flat-space analysis of §4.3 (step 5), why no such terms can occur. As written, the claim that step 5 leaves 'just a constant' is stated rather than derived, and the position-space matching in §5.2 uses c0 as the sole free parameter, so the completeness of the final answer (5.6)–(5.9) depends on it.","section":"§4.2, Eqs. (4.11)–(4.13)"}],"minor_comments":[{"comment":"§5.2: the statement that the Mellin matching is 'highly over-constrained' would be more informative with a count: seven residues R_ij are used to fix six parameters plus the c0 relation, but each R_ij is itself a multi-coefficient function of logB and logD. Please state how many independent coefficient equations beyond the number of unknowns were actually checked.","section":"§5.2"},{"comment":"§5.4, Eq. (5.16): the s=0 anomalous dimension retains the c0 dependence, so the leading-twist binding-energy correction is determined only for s≥1. This is acknowledged via the spin-0 pole discussion, but the abstract's claim of extracting 'the quantum correction to the anomalous dimension' should be qualified accordingly.","section":"§5.4, Eq. (5.16)"},{"comment":"Typographical: 'give difference slices' in §4.1 should read 'different slices'; 'These result are' in §4.2 should be 'These results are'; 'position space corelator' in the final comment of §5.3 should be 'correlator'. Footnote 13 contains numbered equations (5.12)–(5.13) that interrupt the numbering flow; consider moving to the main text or an appendix.","section":"Various"},{"comment":"§3.2: the notation m_d in (3.14) and m_b in (3.30) for the shifted summation indices is easy to confuse with the defect-twist label m used elsewhere; suggest renaming. Figure 3 is central to the logic of the paper and would benefit from a caption explicitly stating which coefficients fix c_mn and which are predictions.","section":"§3.2, Fig. 3"},{"comment":"Eq. (5.3): the definition of the one-loop box ϕ(1) omits the customary prefactor 1/(z−z̄); since this convention feeds into the parity assignments of Q1 and Q2 in (5.1), a one-line remark on the normalization would prevent confusion with the defect-free literature [21].","section":"§5.1, Eq. (5.3)"}],"recommendation":"major_revision","confidential_remarks":"The manuscript depends heavily on the authors' own recent pipeline [4,5,18] for the tree-level inputs and the defect unitarity method. Those inputs are independent published computations rather than rearrangements of the present result, so this is not circular, but referees and readers should be aware that external verification of the tree-level data (especially the unmixing formulas (3.16)–(3.18) and (3.31)–(3.34)) would strengthen the foundation. Fit to journal scope is excellent; this is a natural and nontrivial extension of the AdS unitarity program to a genuinely non-planar heavy-heavy-light-light observable."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"This is the first explicit O(1/c) correction to a maximal-giant + two-stress-tensor correlator. They treat the giants as a zero-dimensional defect, glue the known tree data of [4,5] with the defect unitarity method of [18], and produce closed forms: Mellin amplitude with simultaneous poles and cmn built from 3F2’s (4.4), and position-space H22^(2)=Δ̂(4)P^(2)+½H^(1) with an explicit weight-≤3 basis that needs the new letters {V,1−V} and Q8 (5.6–5.9). They also extract the leading-twist defect anomalous dimensions (5.16). That package is new and usable.\n\nWhat works: bulk and defect LLS are unmixed and resummed independently; the Mellin residues then reproduce the full set of subleading logs, not just the simultaneous logB log^{2}D used to fix cmn; the position-space bootstrap is over-constrained by seven Ri,j residues and matches; the flat-space limit collapses to the expected one-loop defect Feynman integral. The free parameters (contact c0 and an additive multiple of the tree correlator) are stated honestly and left for localization/protected data. Citation pattern is appropriate: heavy use of the authors’ own tree and unitarity papers, but those are independent inputs, not rearrangements of the one-loop target.\n\nSoft spot, kept in proportion: the Mellin ansatz assumes only simultaneous simple poles with constant numerators. Matching subleading logs shows that solution is sufficient; it does not rigorously exclude a tuned single-pole sector that could shift cmn while preserving the same LLS. The OPE log-degree bound already kills non-constant numerators, and flat-space asymptotics would likely be disturbed by a large single-pole piece, so the residual freedom is narrow and checkable rather than structural. Six of the seven position-space coefficients are fixed by matching this Mellin answer, so the gap propagates coherently if it exists; it does not, however, invalidate the reconstruction as written.\n\nFor anyone working on holographic defects, precision AdS5×S5 correlators, or binding energies of giant+supergraviton states, this is worth reading and citing. It deserves a serious referee.","headline":"Solid first one-loop giant-graviton correlator via defect unitarity; closed Mellin/position forms and leading-twist γ̂^(2) are real, with a narrow residual uniqueness gap on single poles that does not sink the result.","tokens_in":33040,"tokens_out":586,"would_cite":true,"duration_ms":11563,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"The leading quantum gravity correction to two maximal giant gravitons plus two light supergravitons is reconstructed in closed form from tree-level data via defect AdS unitarity.","keywords":["giant gravitons","AdS unitarity","defect CFT","Mellin amplitude","one-loop correlators","N=4 SYM","anomalous dimensions","holographic defects"],"falsifier":"Compute any independent residue or low-order series coefficient of the one-loop correlator (for example R0,0 or the coefficient of log²V in the defect channel) by a different method—direct Witten-diagram evaluation, localization, or a higher-charge unitarity cut—and check whether it matches the closed-form expressions given in the paper.","tokens_in":32698,"feed_emoji":"⚛️","tokens_out":952,"duration_ms":17570,"temperature":0.7,"pith_summary":"Giant gravitons are heavy, highly supersymmetric operators that still feel genuinely non-planar physics even at infinite N. This paper treats a pair of maximal giants as a zero-dimensional defect and computes the first quantum-gravity (one-loop) correction to their correlator with two light stress-tensor multiplet operators. The calculation does not evaluate new Witten diagrams; instead it glues known tree-level OPE data with a defect version of the AdS unitarity method, producing the full one-loop reduced correlator in closed form both as a Mellin amplitude with simultaneous poles and as an explicit position-space function of weight at most three. From the same object the authors extract the one-loop shift to the binding energy of the lightest giant-plus-supergraviton bound states. The result supplies a concrete, checkable window into quantum corrections that sit between the planar limit and finite-N physics.","feed_headline":"One-loop giant-graviton correlator fixed by defect unitarity","feed_subtitle":"Closed Mellin and position-space forms give the first quantum correction to giant-plus-light binding energies","key_machinery":"Defect AdS unitarity: the one-loop defect two-point function is reconstructed by gluing tree-level bulk and defect OPE data into the leading logarithmic singularities, which are then completed to the full correlator by a simultaneous-pole Mellin ansatz (or a finite transcendental basis in position space).","core_discovery":"The leading O(1/c) one-loop reduced correlator of two maximal giant gravitons and two stress-tensor multiplet operators is completely determined (up to a constant contact counterterm and an additive multiple of the tree-level answer) by defect AdS unitarity from lower-order OPE data. It admits a closed Mellin representation whose numerator coefficients are hypergeometric 3F2 functions, and an equivalent position-space expression written as a fourth-order differential operator acting on a simple weight-≤3 transcendental pre-correlator; the same correlator yields the one-loop anomalous dimensions of the leading-twist defect-channel bound states.","pith_inferences":["The appearance of the same 3F2 building blocks found earlier for surface defects hints that simultaneous-pole Mellin amplitudes with hypergeometric numerators may be universal for a broad class of holographic defects.","Once the contact-term ambiguity is fixed by an integrated correlator or free-theory input, the result becomes a precision benchmark for any future direct string or supergravity computation of D3-brane loop effects.","Extending the construction beyond maximal giants would test how much of the defect unitarity method survives for more general heavy 1/2-BPS backgrounds."],"forward_implications":["The same unitarity glueing determines one-loop giant-graviton correlators with higher Kaluza-Klein light operators once the corresponding tree-level data are known.","The fourth-order differential operator that organizes the position-space answer suggests a possible iterative construction of two-loop and higher corrections.","The extracted one-loop anomalous dimensions give the first quantum correction to the binding energies of giant-plus-supergraviton bound states in AdS.","The flat-space limit of the Mellin amplitude reduces to a single one-loop Feynman diagram of massless particles scattering off an extended defect, providing a direct bulk check."],"fun_headline_variants":["Defect unitarity determines one-loop giant-graviton correlator","Closed Mellin form for quantum-corrected giant graviton correlators","One-loop giant-plus-light binding energies from defect AdS unitarity","Position-space one-loop correlator of two giants and two supergravitons","Leading 1/c correction to maximal giant graviton four-point function"],"cache_read_input_tokens":16512,"weakest_assumption_plain":"The one-loop Mellin amplitude is assumed to contain only simultaneous simple poles with constant numerators and no extra single-pole terms; that assumption is checked after the fact but is not derived from first principles.","fun_headline_variants_meta":{"raw":{"variants":["Defect unitarity determines one-loop giant-graviton correlator","Closed Mellin form for quantum-corrected giant graviton correlators","One-loop giant-plus-light binding energies from defect AdS unitarity","Position-space one-loop correlator of two giants and two supergravitons","Leading 1/c correction to maximal giant graviton four-point function"]},"model":"grok-4.5","effort":"low","cost_usd":0.004632,"raw_usage":{"total_tokens":1293,"prompt_tokens":727,"num_sources_used":0,"completion_tokens":82,"cost_in_usd_ticks":46324000,"prompt_tokens_details":{"text_tokens":727,"audio_tokens":0,"image_tokens":0,"cached_tokens":128},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":484,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":727,"tokens_out":82,"duration_ms":9335,"temperature":1.0,"reasoning_tokens":484,"cache_read_input_tokens":128,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-31T02:27:01.981255+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Compute any independent residue or low-order series coefficient of the one-loop correlator (for example R0,0 or the coefficient of log²V in the defect channel) by a different method—direct Witten-diagram evaluation, localization, or a higher-charge unitarity cut—and check whether it matches the closed-form expressions given in the paper.","supporting_citations":[],"review_version":1}