{"id":"c229b762-cf28-43e8-994e-66f7a47decdd","arxiv_id":"2608.05963","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":8.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"Gautschi's conjecture about endpoint ratios of subrange Jacobi polynomials is proved for all degrees when β≥0 and for additional negative-parameter regions, leaving only a bounded parameter wedge open.","lead":"The paper proves Gautschi's 2018 conjecture on subrange Jacobi polynomials for a large new set of parameters, showing certain quadrature nodes move monotonically as the interval grows. The proof introduces a first-crossing argument plus a statistical-physics-style Ward identity for orthogonal polynomial ensembles, which could be reused on similar monotonicity questions.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified","rationale":"The reader's weakest_assumption correctly identifies Lemma 5.1 as the most externally dependent step, since it imports a nontrivial association theorem and a covariance sign that is not derived from first principles in the paper. However, my detailed check of that lemma shows the MTP2 hypotheses are satisfied: the ordered chamber is a sublattice, the pairwise lattice inequality for ∆^2 is correct in both same-orientation and crossed cases, one-particle factors cancel exactly, and the functions S and e^{2λH} are bounded, coordinatewise increasing, and satisfy the symmetry needed for E_0 S = 0. The Ward identity is algebraically verified, including the pairing of the double sum giving −(N−1)Σx_i, and the boundary terms vanish. The subsequent induction in Theorem 1.4 uses the bound (5.3) with the inequality direction exactly as required. The positive-sector theorem (1.2) is independent of Lemma 5.1 and its proof via the Pearson identity (3.3), crossing lemma (3.4), and root-sum propagation is internally consistent. The degree-one proof, eventual-validity proof via the weak arcsine limit, and the Möbius reformulation also check out. The only caveats are the explicitly stated limitations in Remark 5.2 and the non-effectiveness of Theorem 1.6(2), neither of which affects correctness. Hence I do not identify a load-bearing concern, and the ACCEPT verdict stands. The proposed numerical test would still be a worthwhile independent check of the single most imported step.","tokens_in":14321,"tokens_out":43034,"duration_ms":350482,"concrete_test":"Run a high-precision numerical check of Lemma 5.1: for a grid of N = 1..10, r in (0,1), λ in (0,0.5), a in (0.1,0.99), compute the OPE expectations in (5.7) via Gaussian quadrature with the weight w_λ on [-a,a], and verify (N−ra²)Σ_N < Nλa². Also directly estimate Cov_{0}(S, e^{2λH}) by sampling or quadrature; any negative value would contradict the MTP2 input and would invalidate Theorem 1.4.","verdict_should_be":"UNCHANGED","load_bearing_attack":"After checking the main proof chains, I find no load-bearing flaw. The most imported step is Lemma 5.1's application of the MTP2 association theorem [7, Theorem 4.2] to conclude Cov_{0}(S, e^{2λH}) ≥ 0. The necessary hypotheses hold: the ordered chamber is a sublattice; the Vandermonde-square lattice inequality is verified pairwise in §5; the one-particle factors cancel under coordinatewise min/max; S and e^{2λH} are bounded and coordinatewise increasing on the support; and the base law is invariant under the reversal involution, giving E_{0}S = 0. The Ward identity leading to (5.7) is algebraically consistent, the boundary fluxes vanish, and the dominated convergence bound for the truncated chamber is sound. The induction in Theorem 1.4 uses the bound exactly as derived, and the crossing lemma is internally correct. The positive-sector proof rests only on the Pearson identity, the crossing lemma, and the induction, all of which I verified step-by-step. The limitations stated in Remark 5.2 and Section 6 are honest boundaries, not hidden gaps. Thus the central claims appear sound.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies Gautschi's 2018 conjecture on subrange Jacobi polynomials: for -1<α<β and 0<c≤1, the endpoint ratio (1.1) is claimed to be <1. The authors introduce an equivalent endpoint difference D_n, derive exact identities, and prove the conjecture uniformly in degree for 0<α<β (Theorem 1.2). They also prove a degree-uniform family in the remaining negative wedge under c^2≤3/(3+r) (Theorem 1.4), yielding the full conjecture for all admissible parameters when c≤√3/2 (Corollary 1.5), plus degree-one and eventual-in-degree results for arbitrary parameters. The proof combines a boundary identity, Pearson and root-motion identities, a first-crossing argument, and an MTP2 ensemble Ward identity.","tokens_in":14559,"tokens_out":35061,"duration_ms":296196,"significance":"If correct, this is a major advance on a problem posed by Gautschi. It closes the β≥0 range completely and reduces the negative-wedge case to a parameter region Oc that is empty for c≤√3/2; for fixed parameters only finitely many degrees remain open. The paper's main tools are exact and parameter-free: no numerical evidence is used, thresholds such as 3/(3+r) are derived rather than fitted, and the positive-sector proof rests on transparent identities. The negative-wedge proof is the most delicate part; its use of the Karlin–Rinott association theorem is carefully checked, including the sublattice property, the Vandermonde-square inequality in the ordered chamber, and the Ward-identity boundary fluxes. I found no load-bearing errors.","major_comments":[],"minor_comments":[{"comment":"The sentence \"Orthogonality annihilates the part containing p_n p'_n\" is potentially misleading, because the integral of (1-x^2)p_n p'_n w is not zero; the nonzero contribution comes from -∫ x^2 p_n p'_n w after using ∫ p_n p'_n w = 0. A short clarification of the two integrals would improve readability.","section":"3.1, proof of Lemma 3.1"},{"comment":"The parameter t in the density dP_{N,t} is never explicitly set equal to λ before the exponential tilting by e^{2λH} is used; stating t=λ at the definition would remove a small ambiguity.","section":"5, equation (5.4)"},{"comment":"The panel label \"c=2√5\" appears to omit a slash; the caption states c=2/√5, and the label should be corrected to match.","section":"Figure 1, panel (b)"},{"comment":"For α=0 the equivalence of (7.4) with Conjecture 1.1 degenerates to the trivial inequality 0<β; a one-line parenthetical comment would prevent confusion in that boundary case.","section":"7, equation (7.4)"}],"recommendation":"accept","confidential_remarks":"I have no concerns about provenance or citation practice. The only imported external result with real weight is the MTP2 association theorem; its application in Lemma 5.1 is valid, and the remaining proof is internally consistent. The paper is a good fit for the journal."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Candid take: this is the real thing. It resolves Gautschi's conjecture for the entire β≥0 half of the admissible range, and for the negative wedge when c²≤3/(3+r), leaving only the explicitly characterized O_c region for c>√3/2. Theorem 1.2 and Theorem 1.4 are genuine new results, not repackagings of Milovanović's criterion. The paper also proves degree-one exactly and eventual validity via weak arcsine limits.\n\nThe proof structure is sound. The boundary identity (2.2)–(2.3), the Pearson identity (3.3), the root-motion formula (3.6), and the crossing lemma (3.4) all check out. The induction in Theorem 1.2 is airtight. In the negative wedge, the Ward identity (5.7) is algebraically consistent, the boundary fluxes vanish, and the covariance nonnegativity follows from the MTP2 association theorem. I read the ensemble bound carefully and the hypotheses of [7, Theorem 4.2] are satisfied: the ordered chamber is a sublattice, the Vandermonde-square inequality is verified pairwise, the one-particle factors cancel under min/max, and S and e^{2λH} are bounded and coordinatewise increasing. The limitation in Remark 5.2 is honest and correctly states that the method just doesn't reach beyond c²=3/(3+r).\n\nSoft spots, in proportion: (1) The MTP2 application is the most imported step; if that theorem were misapplied, Theorem 1.4 collapses. But I don't see a misapplication. A referee should still check that step first. (2) The eventual-validity result is non-effective, so the finite middle block is not bounded uniformly; the authors say this openly. (3) The degree-one proof uses a covariance inequality that is correct but a bit intricate. None of these are load-bearing.\n\nCitation pattern is clean: earlier results by Gautschi and Milovanović are re-derived in Section 2 or cited as external criteria, not leaned on as their own. No fitted parameters; the threshold 3/(3+r) is derived. No self-citation.\n\nWho should read this: anyone working on orthogonal polynomials, Gaussian quadrature, or zero monotonicity. It deserves a serious referee. I would send it to peer review; my expectation is acceptance after a routine check of the ensemble step.","headline":"Solid paper: resolves Gautschi's conjecture for β≥0 and for the negative wedge when c²≤3/(3+r), with careful proofs and an honest boundary; the MTP2 step is the one to scrutinize but it holds up.","tokens_in":15087,"tokens_out":7792,"would_cite":true,"duration_ms":62385,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["33C45","42C05","60E15","65D32"],"pacs":[],"model":"deepseek-v4-flash","headline":"Gautschi's conjecture proven for every degree in the positive sector","keywords":["Gautschi's conjecture","subrange Jacobi polynomials","zero monotonicity","orthogonal polynomial ensembles","MTP2 association","Pearson identity","root-motion identities","Ward identity"],"falsifier":"Compute the normalized endpoint difference $D_n/h_n$ at a fixed degree (say $n=2$ or $3$) for parameters in the remaining wedge $c^2>3/(3+r)$ using high-precision quadrature; a non-positive value would refute the conjecture in that region. Alternatively, evaluate the covariance $\\mathrm{Cov}_{N,0}(S,e^{2\\lambda H})$ for a small ensemble (for example $N=5$, $r=0.5$, $\\lambda=0.4$, $a=0.9$); a negative value would disprove the MTP2 association hypothesis used in Lemma 5.1.","tokens_in":14160,"feed_emoji":"📈","tokens_out":14460,"duration_ms":108031,"temperature":0.7,"pith_summary":"Gautschi's conjecture concerns monic orthogonal polynomials for the Jacobi weight $(1-x)^\\alpha(1+x)^\\beta$ restricted to a subinterval $[-c,c]$, and asserts that the endpoint ratio $[\\pi_n(-c)/\\pi_n(c)]^2((1-c)/(1+c))^{\\beta-\\alpha}$ is strictly less than $1$. Because of Gautschi's variation formula, this inequality is sufficient to conclude that every positive zero moves strictly to the right as $c$ increases. The paper proves the conjecture uniformly in the degree for $0<\\alpha<\\beta$, completing the entire admissible range $\\beta\\ge 0$, and proves it also in the negative wedge $\\alpha=-r-\\lambda$, $\\beta=-r+\\lambda$ whenever $c^2\\le 3/(3+r)$. Together with earlier results this settles the conjecture for every admissible pair of parameters when $c\\le\\sqrt{3}/2$. The proof combines an exact boundary identity, a first-crossing lemma, and a Ward identity with positive association for an orthogonal polynomial ensemble.","feed_headline":"Gautschi's conjecture proven for every degree in the positive sector","feed_subtitle":"The endpoint inequality now holds for every degree in the positive sector, so positive zeros move right as the interval grows.","key_machinery":"The argument revolves around the endpoint difference $D_n=w(c)\\pi_n(c)^2-w(-c)\\pi_n(-c)^2$, which the boundary identity rewrites as $\\int_{-c}^{c} w'(x)\\pi_n(x)^2\\,dx$; the conjecture is exactly $D_n>0$. In the positive sector, the Pearson identity $d_n=\\lambda+(n+\\nu-1)\\Sigma_n-(n+\\nu)\\Sigma_{n+1}$ and the root-motion formula $\\partial_c\\Sigma_n=a_n(U_nU_{n-1}-V_nV_{n-1})$ feed a crossing lemma: at a first zero of $d_n$ its derivative must be positive, contradicting the definition of a first crossing. In the negative wedge, a Ward identity for the ordered-chamber orthogonal polynomial ensemble, combined with the MTP$_2$ association theorem, gives the root-sum estimate $(N-ra^2)\\Sigma_N < N\\lambda a^2$, which supplies the root-sum bound required by the crossing lemma.","core_discovery":"The paper's central discovery is that Gautschi's endpoint inequality, verified pointwise at the smallest interval and prevented from crossing zero by a local transversality argument, holds for every degree uniformly in the positive sector $0<\\alpha<\\beta$ (Theorem 1.2). The same first-crossing mechanism, fed by a root-sum estimate derived from an MTP$_2$ orthogonal polynomial ensemble, proves the inequality throughout the negative wedge when $c^2\\le 3/(3+r)$ (Theorem 1.4). Consequently, for $0<c\\le\\sqrt{3}/2$ the conjecture holds for every admissible pair of parameters (Corollary 1.5). The paper also establishes the case $n=1$ in full generality and shows eventual validity for each fixed admissible triple as $n\\to\\infty$, so at any remaining parameter point only finitely many degrees are unresolved.","pith_inferences":["The paper leaves implicit that a quantitative strengthening of the ensemble covariance bound used in Lemma 5.1 would close the remaining wedge and prove the conjecture for all admissible parameters.","Because eventual validity holds at every fixed parameter point, any counterexample would have to occur at a low degree; checking the finite intermediate block at each parameter point would settle the residual region.","The same MTP2-tilted Ward identity likely applies to other subrange weight families, offering a general recipe for the root-sum estimates needed in first-crossing proofs of zero monotonicity."],"forward_implications":["For every $n\\ge 1$, $0<c<1$, and $0<\\alpha<\\beta$, every positive zero of $\\pi_n$ moves strictly to the right as $c$ increases, by Gautschi's variation formula.","The endpoint inequality holds uniformly in the degree throughout the positive sector and in the negative-wedge slice $c^2\\le 3/(3+r)$, making zero monotonicity degree-uniform in those regions.","For $0<c\\le\\sqrt{3}/2$, Gautschi's conjecture holds for every admissible parameter pair and every degree, closing the full admissible range.","At every fixed admissible parameter point outside the degree-uniform regions, the remaining question is reduced to finitely many degrees $n\\ge 2$, since degree one is proved and eventual validity as $n\\to\\infty$ is established.","The Möbius reformulation converts the conjecture into the reciprocal-moment inequality $M_n>\\beta/\\alpha$, giving an exact finite-degree criterion in the remaining wedge."],"supporting_citations":[{"why":"states Gautschi's conjecture and the variation formula that turns the endpoint inequality into zero monotonicity.","marker":"[6]"},{"why":"supplies the published criterion $\\beta-\\alpha\\ge c|\\alpha+\\beta|$ and the boundary-identity computation that the paper extends.","marker":"[9]"},{"why":"the MTP$_2$ association theorem used to prove the non-negative covariance bound in Lemma 5.1.","marker":"[7]"},{"why":"the determinantal representation of orthogonal polynomial ensembles used to identify $\\mathbb{E}_{N,\\lambda}\\sum_i x_i$ with $\\Sigma_N$.","marker":"[8]"},{"why":"ratio asymptotics for orthogonal polynomials used in the weak arcsine limit that gives eventual validity.","marker":"[10]"},{"why":"the weak-convergence framework used to obtain the arcsine limit for the subrange measure.","marker":"[11]"}],"fun_headline_variants":["Gautschi's conjecture holds for all degrees in positive parameter sector","Gautschi's conjecture resolved for all parameters when c ≤ √3/2","Gautschi's conjecture eventually holds for all parameters"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof in the negative wedge rests on the premise that a certain weighted sum of the roots remains non-negative on average after exponential tilting; if that positive correlation failed, the root-sum estimate and Theorem 1.4 would collapse.","fun_headline_variants_meta":{"raw":{"variants":["Gautschi's conjecture holds for all degrees in positive parameter sector","Gautschi's conjecture resolved for all parameters when c ≤ √3/2","Gautschi's conjecture eventually holds for all parameters"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.002338,"raw_usage":{"total_tokens":9062,"prompt_tokens":1050,"completion_tokens":8012,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":666,"completion_tokens_details":{"reasoning_tokens":7950}},"tokens_in":666,"tokens_out":8012,"duration_ms":55799,"temperature":1.0,"reasoning_tokens":7950,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T20:13:39.132774+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the normalized endpoint difference $D_n/h_n$ at a fixed degree (say $n=2$ or $3$) for parameters in the remaining wedge $c^2>3/(3+r)$ using high-precision quadrature; a non-positive value would refute the conjecture in that region. Alternatively, evaluate the covariance $\\mathrm{Cov}_{N,0}(S,e^{2\\lambda H})$ for a small ensemble (for example $N=5$, $r=0.5$, $\\lambda=0.4$, $a=0.9$); a negative value would disprove the MTP2 association hypothesis used in Lemma 5.1.","supporting_citations":[{"cited_title":"Gautschi,On the zeros of subrange Jacobi polynomials, Numer","cited_arxiv_id":null,"evidence_quote":"states Gautschi's conjecture and the variation formula that turns the endpoint inequality into zero monotonicity."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"supplies the published criterion $\\beta-\\alpha\\ge c|\\alpha+\\beta|$ and the boundary-identity computation that the paper extends."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"ratio asymptotics for orthogonal polynomials used in the weak arcsine limit that gives eventual validity."},{"cited_title":"Van Assche,Weak convergence of orthogonal polynomials, Indag","cited_arxiv_id":null,"evidence_quote":"the weak-convergence framework used to obtain the arcsine limit for the subrange measure."}],"review_version":1}