{"id":"515ef8e0-bbc0-496c-98f9-3df61ef1ef8f","arxiv_id":"2504.17840","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The next-to-leading Wilson coefficient in D=7 maximal supergravity is minimized at the same U-duality symmetric point as the leading coefficient, supporting string universality.","lead":"This paper computes the lowest value of the second Wilson coefficient in seven-dimensional string theory and finds it occurs at the same special lattice point as the first coefficient. The result is numerical and conditional on an unproved conjecture, but it supports the idea that string theory is the unique consistent completion of maximal supergravity.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Global minimum claim is conditional on an unproved symmetric-point conjecture and a single-line check; a full-moduli numerical search is needed.","rationale":"The paper contains a careful two-loop string amplitude analysis and a well-defined renormalized Eisenstein series evaluation, and the numerical value at H_D5 is stable within the chosen truncation. However, the central claim that this value is the global minimum on the full moduli space is not established by the evidence presented. The authors explicitly restrict to a two-dimensional surface and then to a one-dimensional line, and the extension to the full space relies on a conjecture from the earlier paper [34] that is not proved here. The proposed concrete test directly targets that gap: a global numerical search using their own Fourier expansions would either find a lower point or substantially increase confidence. Since the reader's verdict already marks the paper CONDITIONAL on essentially this assumption, I would not move the verdict; I agree with the reader's assessment.","tokens_in":17955,"tokens_out":4813,"duration_ms":50026,"concrete_test":"Run a global numerical minimization of E^(1,0) over the full 14-dimensional fundamental domain of SO(5)\\SL(5,R)/SL(5,Z), using the paper's truncated Fourier expansions (eqs. 2.4 and 2.13) with nmax=5 and nmax=6 for selected candidates. Initialize from thousands of random points and from all irreducible symmetric points in [41] (D5, D*5, A5, A*5, A+2_5, A+3_5), recording every converged local minimum. If any point yields E^(1,0) < 26.2315, or a local minimum off the x = -2/5 line with value below the D5 value, the headline claim is refuted; if all minima found are at D5 and D*5 with D5 lowest, the conditional conclusion is strengthened.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The decisive gap is not merely the inherited conjecture about individual Eisenstein series; it is the step from those conjectures to the minimum of the positive linear combination E^(1,0) = ζ(5)\\hat E_{5/2Λ1} + (π/(15ζ(5)))\\hat E_{5/2Λ3}. The two summands have different global minima at the relevant s (D5 for Λ1, D*5 for Λ3), so a positive linear combination can in principle have its minimum anywhere; positivity does not force the minimum onto the S5-invariant line x = -2/5. Section 1 replaces proof by \"with these hypothesis and with the help of numerical analysis\", and Section 2 evaluates only the surface H(r,x) and the line x = -2/5. The paper itself concedes (after eq. 2.21) that proving only D5 and D*5 are local minima would require systematically studying the next-to-minimal Eisenstein series on the entire moduli space. Unless that gap is closed, the abstract's \"we find the minimum at the same point\" is an unsupported numerical extrapolation from a one-dimensional slice of a 14-dimensional moduli space.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper computes (numerically) the next-to-leading Wilson coefficient E^(1,0) in the D=7 low-energy expansion of the type II four-graviton amplitude. This coefficient is expressed as a positive linear combination of renormalised SL(5,Z) Eisenstein series attached to the weights Λ1 and Λ3. The authors develop Fourier expansions, including for the next-to-minimal Eisenstein series, evaluate the function on a two-parameter surface H(r,x), and claim that its global minimum is attained at the D5 U-duality symmetric point, with E^(1,0)(H_D5) ≈ 26.2315. Section 3 derives the two-loop string-theory contribution and fixes the renormalisation that removes the supergravity divergence, leading to the coefficient in (1.7).","tokens_in":18252,"tokens_out":7369,"duration_ms":70609,"significance":"The claimed result is potentially significant for the S-matrix bootstrap programme: it would identify the next-to-leading Wilson coefficient of maximal supergravity with the minimum of the bootstrap bound at the same self-dual point as the leading coefficient. The renormalisation computation in Section 3 is detailed, and the Fourier-truncation stability check (nmax=4 vs 5) is a genuine strength. The numerical value (1.10) is reproducible in principle from the described expansion. However, the central global-minimum statement is conditional: it relies on an unproved symmetric-point conjecture and on evaluation on a one-dimensional slice of a 14-dimensional moduli space. The value 26.2315 should therefore be read as a candidate minimum until the global claim is established.","major_comments":[{"comment":"The abstract's statement that the next-to-leading Wilson coefficient is minimum at the same point as the leading Wilson coefficient is not supported by the arguments in Sections 1–2. The derivation assumes the conjecture from [34] that local minima of the relevant Eisenstein series occur only at symmetric points, and the numerical work is restricted to the surface H(r,x) and, ultimately, to the S5-invariant line x = −2/5. These restrictions are insufficient for locating a global minimum on the 14-dimensional moduli space. In particular, the two summands in (1.7) have their respective minima at D5 (Λ1 term) and D*5 (Λ3 term at s = 5/2), so positivity alone does not force the minimum of the positive linear combination to lie on that line. The text itself concedes, after (2.21), that proving that D5 and D*5 are the only local minima would require systematic study of the entire moduli space. A full numerical search over a fundamental domain, or a proof that the minimum of the combination occurs at a symmetric point, is required before (1.10) can be reported as a global minimum.","section":"§1 and §2 (global minimum claim)"},{"comment":"The claim that the renormalised next-to-minimal Eisenstein series has only D5 and D*5 as minima in SO(5)\\SL(5) goes beyond what is computed. The numerical results in that section are evaluations on the S4-invariant surface (2.18) and on the line x = −2/5; they do not exclude other local minima, including at the other symmetric points A5 and A*5 or on lower-strata loci. The following sentence acknowledges this gap. Since the Wilson coefficient is a positive combination of this function with the Λ1 Eisenstein series, the Hessian analysis at the symmetric points and the possible existence of off-surface minima are load-bearing for the main conclusion and need to be addressed.","section":"§2, after Eq. (2.20)"},{"comment":"The determination that the minimum on the line is at H_D5 is graphical: the authors state that the plots of the two summands are nearly identical and that their derivatives nearly always have the same sign. No quantitative bound or error control is provided for this comparison, and the stitching of two dual Fourier representations in Fig. 2 could in principle introduce an artificial feature at the transition point r = 1/5^{1/4}. The quoted value 26.2315 is stable against increasing nmax, but this cutoff stability does not test the restriction of the domain to one line. I would ask for explicit values at D5, D*5, A5, A*5 and at sample points away from the line, together with Hessian eigenvalues or gradient information in off-line directions.","section":"§2, Fig. 2 and Eq. (2.22)"}],"minor_comments":[{"comment":"There are several typos: 'in this detail' should be 'in detail' (p.6); 'Madelstam' should be 'Mandelstam' (p.12); and 'hypothesis' should be 'hypotheses' in Section 1.","section":"Throughout"},{"comment":"The notation Γ1-loop is used in (3.20) and (3.22) without an explicit definition; it should be defined in the same way as Γ2-loop in (3.4).","section":"§3, Eq. (3.20)"},{"comment":"In Eq. (2.18), x denotes a scalar coordinate while x_N in (2.3) is a vector; a different symbol for the scalar would avoid ambiguity.","section":"§2, Eq. (2.18)"},{"comment":"The phrase 'almost degenerate' for the values at D5 and D*5 should be quantified by giving the numerical difference to the stated precision.","section":"§2, Eq. (2.21)"}],"recommendation":"major_revision","confidential_remarks":"The Section 3 two-loop renormalisation computation appears sound and valuable, but the headline claim in the title and abstract is stronger than the evidence presented. The global-minimum statement depends on an unproved conjecture from the authors' earlier paper and on a restricted numerical scan over a one-dimensional line. If the authors can provide an exhaustive numerical minimisation over the moduli space, or a proof that the positive linear combination has its minimum at a symmetric point, the paper would be suitable for publication; otherwise the abstract and conclusions should be reworded as a candidate minimum under explicit hypotheses."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Read it. The genuinely new pieces are the two-loop renormalization scheme in Section 3 and the numerical analysis of the next-to-minimal Eisenstein series on the S4-symmetric surface. The determination that E^(1,0)(H_D5) ≈ 26.2315, with stable truncation checks, is a concrete numerical result that I trust. That part is careful and reproducible in principle.\n\nBut the headline claim—that this is the global minimum of the Wilson coefficient—is not proven. Section 1 says “with these hypothesis and with the help of numerical analysis,” and Section 2 evaluates only the H(r,x) surface and the line x = -2/5. The paper itself concedes after eq. (2.21) that proving D5 and D*5 are the only local minima would require systematically studying the function on the entire moduli space. The stress-test point is sharper: the two summands in (1.7) have their minima at different points (D5 for the L1 term, D*5 for the L3 term at this value of s), and positivity does not force the minimum of the positive linear combination onto the S5-invariant line. The observed derivative-similarity on that line is suggestive, but it is not a substitute for a search over the 14-dimensional moduli space.\n\nSo the abstract overstates. “We find the minimum at the same point” should read “assuming the symmetric-point conjecture inherited from [34], the minimum is at the same point.” That is a real presentational flaw, but it is not a sign of sloppy work: the renormalization computation is careful, the truncation checks are explicit, and the paper flags its own missing proof. I would not call this a takedown. It moves the bootstrap-saturation program one step further at D=7, but only provisionally.\n\nRecommendation: send this to a serious referee. The referee should demand either a proof of the symmetric-point conjecture or an explicit caveat in the abstract and conclusion. I would cite it for the value at D5 and the renormalization scheme, not for the global-minimum claim.","headline":"The two-loop renormalization and the numerical value at D5 are solid, but the global-minimum claim is conditional on an unproved symmetric-point conjecture and a one-line check; the abstract overstates the result.","tokens_in":18692,"tokens_out":1835,"would_cite":true,"duration_ms":18123,"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 next-to-leading Wilson coefficient in seven-dimensional maximal supergravity is minimized at the same U-duality symmetric point as the leading coefficient, the D5 point.","keywords":["maximal supergravity","Wilson coefficient","Eisenstein series","next-to-minimal automorphic representation","U-duality","type II string theory","S-matrix bootstrap","two-loop superstring amplitude"],"falsifier":"Compute $E^{(1,0)}$ on a fine grid over the full 14-dimensional moduli space, or at least on other symmetric strata such as the three-dimensional family invariant under S4, with a more complete truncation of the lattice sums; any value below 26.2315, or a non-symmetric stationary point with a downward direction, would refute the global-minimum claim.","tokens_in":17779,"feed_emoji":"📉","tokens_out":16772,"duration_ms":138792,"temperature":0.7,"pith_summary":"For seven-dimensional type II string theory on a three-torus, the paper identifies where the exact non-perturbative next-to-leading Wilson coefficient $E^{(1,0)}(\\varphi)$ takes its smallest value. The coefficient is a specific positive combination of two renormalised SL(5) Eisenstein series, one attached to a minimal automorphic representation and one to a next-to-minimal one. The paper argues, using numerical evaluation on a specially chosen surface and along the unique S5-invariant line joining the two candidate points, that the global minimum lies at the D5 U-duality symmetric point — the same point that minimises the leading Wilson coefficient — and evaluates the minimum as $E^{(1,0)}(H_{D5})\\approx 26.2315$. This matters because it shows the next order of the low-energy effective action also selects a single U-duality symmetric point, in line with the expectation that string theory saturates S-matrix bootstrap bounds.","feed_headline":"Next Wilson coefficient in 7D is smallest at the D5 point","feed_subtitle":"Both the leading and next-to-leading coefficients are smallest at D5, supporting saturation at the next bootstrap bound.","key_machinery":"The load-bearing object is the pair of renormalised Eisenstein series $\\hat{E}^{\\mathrm{SL}(5)}_{\\frac{5}{2}\\Lambda_1}$ and $\\hat{E}^{\\mathrm{SL}(5)}_{\\frac{5}{2}\\Lambda_3}$ on the symmetric space $SO(5)\\setminus SL(5,\\mathbb{R})$, the first attached to the minimal automorphic representation and the second to the next-to-minimal one. Their Fourier expansions, expressed with Bessel functions and sums over lattice charges, give the numerical approximations used in the paper. The other central piece is the stratification of the moduli space by stabiliser subgroups: symmetric points — isolated fixed points of finite subgroups of $SL(5,\\mathbb{Z})$ — are forced extrema of all automorphic functions, so the paper restricts its scan to the surface $H(r,x)$ containing the irreducible symmetric lattices $D_5$, $D^*_5$, $A_5$, $A^*_5$, $A^{+2}_5$ and $A^{+3}_5$, and to the unique S5-invariant line $x=-2/5$ joining $D_5$ to $D^*_5$. Along that line the derivatives of the two Eisenstein series almost always have the same sign, so the positive combination inherits their minima.","core_discovery":"On the paper's terms, the central discovery is that the next-to-leading Wilson coefficient $$$E^{{(1,0)}}$(\\varphi)=\\zeta(5)\\hat{E}^{\\mathrm{SL}(5)}_{\\frac{5}{2}\\Lambda_1}(H)+\\frac{\\pi}{15\\zeta(5)}\\hat{E}^{\\mathrm{SL}(5)}_{\\frac{5}{2}\\Lambda_3}(H)$$ has its global minimum at $H=H_{D5}$, with numerical value $\\approx 26.2315$. Here $H$ parametrises the coset $SO(5)\\setminus SL(5,\\mathbb{R})/SL(5,\\mathbb{Z})$, and the hatted Eisenstein series are the finite parts at the singular value $s=5/2$, defined by subtracting the pole dictated by the functional equations. The two ingredients vary almost identically along the S5-invariant line joining the D5 and $D^*_5$ points, so their positive combination has the same local minima; among the symmetric points, D5 is the smallest. The paper states this conclusion under the hypothesis, inherited from its earlier analysis, that all local minima of these automorphic functions occur at symmetric points, and with the help of numerical analysis on the surface $H(r,x)$.","pith_inferences":["One could test the saturation picture directly by computing the S-matrix bootstrap lower bound for the next-to-leading Wilson coefficient in seven dimensions; if that bound approaches $26.2315$, the D5 point would be confirmed as the locus where string theory saturates unitarity at the next order.","Because the two candidate points are nearly degenerate for the $\\Lambda_3$ Eisenstein series, one can in principle vary the relative coefficient of the two terms in (1.7) and locate the ratio at which the minimum flips from D5 to $D^*_5$; the physical ratio $\\pi/(15\\zeta(5))$ sits on the D5 side of that flip.","A proof of the symmetric-point-minima conjecture would upgrade the numerical scan to a theorem, and the paper's evidence suggests the same qualitative behaviour holds for other next-to-minimal Eisenstein series, so the method may transfer to higher derivative coefficients."],"forward_implications":["If the claim is correct, the exact value $E^{(1,0)}(H_{D5})\\approx 26.2315$ is the smallest possible value of the next-to-leading Wilson coefficient in seven-dimensional maximal supergravity with type II string-theory boundary conditions.","The D5 point then simultaneously minimises the leading and next-to-leading Wilson coefficients, making it the natural locus at which string theory would saturate the next order of the S-matrix bootstrap unitary bound.","Because the two Eisenstein series vary almost identically along the S5-invariant line, the positive combination in (1.7) inherits its local minima from them; under the stated hypothesis no other symmetric point competes with D5.","The renormalisation prescription fixed by matching the two-loop superstring integral removes the logarithmic divergence of two-loop supergravity, leaving a finite Wilson coefficient consistent with the Eisenstein-series formula."],"supporting_citations":[{"why":"Supplies the conjecture that local minima occur at symmetric points, the H(r,x) surface, and the previous result that the leading coefficient is minimised at D5.","marker":"[34]"},{"why":"Gives the expression of the next-to-leading Wilson coefficient as the combination of SL(5) Eisenstein series used throughout.","marker":"[38]"},{"why":"Provides the low-energy expansion of the four-graviton amplitude in which the next-to-leading Wilson coefficient is defined at order $\\ell_P^{10}$.","marker":"[7,16]"},{"why":"Supplies the two-loop superstring four-graviton amplitude whose low-energy limit fixes the renormalised coefficient.","marker":"[42]"},{"why":"Provides the three-region decomposition of the genus-two moduli space used to isolate the two-loop low-energy contribution.","marker":"[44]"},{"why":"Gives the Fourier-expansion method, including the appendix computation, used to evaluate the next-to-minimal Eisenstein series numerically.","marker":"[40]"},{"why":"Classifies the maximal finite irreducible subgroups of GL(5,Z), identifying the D5 and A5 lattices and their duals as the candidate symmetric points.","marker":"[41]"},{"why":"Provides the Fourier coefficients of minimal and next-to-minimal automorphic representations that underlie the numerical evaluation.","marker":"[26]"}],"fun_headline_variants":["7D next-to-leading Wilson coefficient bottoms at D5","Maximal supergravity: next Wilson coefficient minimal at D5","Next Wilson coefficient in 7D hits minimum at D5","D5 point minimizes next-to-leading Wilson coefficient in 7D","Bootstrap next-order bound: D5 point is minimal in 7D"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole argument rests on the unproved conjecture that every local minimum of these Eisenstein series occurs at a symmetric point; under that assumption the paper only scans a special surface and the single S5-invariant line joining D5 to $D^*_5$, so if a nonsymmetric minimum existed the true value could be below 26.2315.","fun_headline_variants_meta":{"raw":{"variants":["7D next-to-leading Wilson coefficient bottoms at D5","Maximal supergravity: next Wilson coefficient minimal at D5","Next Wilson coefficient in 7D hits minimum at D5","D5 point minimizes next-to-leading Wilson coefficient in 7D","Bootstrap next-order bound: D5 point is minimal in 7D"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000642,"raw_usage":{"total_tokens":2912,"prompt_tokens":859,"completion_tokens":2053,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":475,"completion_tokens_details":{"reasoning_tokens":1964}},"tokens_in":475,"tokens_out":2053,"duration_ms":12823,"temperature":1.0,"reasoning_tokens":1964,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T10:30:55.688929+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute $E^{(1,0)}$ on a fine grid over the full 14-dimensional moduli space, or at least on other symmetric strata such as the three-dimensional family invariant under S4, with a more complete truncation of the lattice sums; any value below 26.2315, or a non-symmetric stationary point with a downward direction, would refute the global-minimum claim.","supporting_citations":[{"cited_title":"Saturating unitarity bounds at U-duality symmetric points","cited_arxiv_id":"2308.02847","evidence_quote":"Supplies the conjecture that local minima occur at symmetric points, the H(r,x) surface, and the previous result that the leading coefficient is minimised at D5."},{"cited_title":"Infrared divergences and harmonic anomalies in the two-loop superstring effective action","cited_arxiv_id":"1510.02409","evidence_quote":"Provides the three-region decomposition of the genus-two moduli space used to isolate the two-loop low-energy contribution."},{"cited_title":"On Maximal Finite Irreducible Subgroups of GL(n, Z): I. The five and seven dimensional cases.,","cited_arxiv_id":null,"evidence_quote":"Classifies the maximal finite irreducible subgroups of GL(5,Z), identifying the D5 and A5 lattices and their duals as the candidate symmetric points."}],"review_version":1}