{"id":"a74acc5b-14c5-4671-b031-409f24b5f2f8","arxiv_id":"2507.03862","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Sieved Jacobi polynomials are eigenfunctions of an explicit Dunkl-type operator with cyclic reflections, confirming Askey's conjecture and establishing their bispectrality.","lead":"The paper constructs explicit Dunkl-type operators for which Askey's sieved Jacobi polynomials are eigenfunctions, settling a question Askey posed in the 1980s. This gives the sieved Jacobi families, both on the unit circle and on the real line, the bispectrality property that had been missing.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Lemma 5.3's summation identities are false for odd j (e.g., N=2, j=1), leaving Theorem 5.4's odd-parity cases unproven.","rationale":"The central claim is that sieved Jacobi polynomials satisfy the eigenvalue equations (6.6)-(6.7). This follows from Theorem 5.4 by straightforward algebra: H=L^2-N(alpha+beta+1)L and the pairing property lambda_{2n}=lambda_{2n-1}=Lambda_n are easy to verify. Thus the whole paper stands or falls on Theorem 5.4. The theorem's proof has two gaps: Lemma 5.3 is delegated to the authors' preprint [9], and the parity cases are asserted by analogy. I checked the lemma and found it internally inconsistent for odd j; the counterexample above is direct. This is a correctness risk, not merely an exposition preference. However, I found no reason to believe the final theorem is false; the residue identity (5.12) is correct and likely supplies the necessary sums once the correct h is chosen for each parity class. So the appropriate verdict remains CONDITIONAL: require the authors to state the summation identities with correct parity conditions and to provide the case-by-case verification (or a Maple/SymPy script). The reader's weakest_assumption pointed at the same lemma; my check sharpens it from 'unproven' to 'false as stated'.","tokens_in":14800,"tokens_out":13396,"duration_ms":134229,"concrete_test":"Compute the two functions F_n^{(1)}(z) and F_n^{(2)}(z) in equation (5.20) for N=2 and N=3, for all parity classes of n and j and k=0,1,2, using the correct residue identity (5.12) with integer h instead of the false (5.8)-(5.9). If both functions vanish identically, Theorem 5.4 is true and the defect is only the statement of Lemma 5.3; if not, the central claim of the paper fails for the sieved Jacobi OPUC.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The proof of Theorem 5.4 (the operator L(N) diagonalizing the sieved CMV Laurent polynomials) is the load-bearing step: Theorem 6.1 inherits the eigenvalue equation from it. The proof is a sketch that verifies only the case n, N, j all even and then says the other parity cases are analogous. These cases rely on Lemma 5.3, whose identities (5.8) and (5.9) are not merely delegated to [9]; as stated they are false. For N=2, q=-1, j=1, z=2, the left side of (5.9) is -4/3 - 4i/5 while the right side is -4/15; the left side of (5.8) is -4/3 + 4i/5 while the right side is -16/15. No convention rescues these equalities, since q^{1/2}=i is well defined. The residue method does yield the correct identity (5.12) for integer h, but (5.8)-(5.9) are a specialization valid only for even j (or with a different exponent structure). Odd-j branches appear in (4.7)-(4.9) for both parities of n, so the cancellation of the reflection terms in Theorem 5.4 is not demonstrated for those cases. The paper's assertion that 'all other possible situations can be treated analogously' therefore hides a missing family of summation identities.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper claims to settle Askey's conjecture that the sieved Jacobi polynomials are eigenfunctions of a differential/difference operator of Dunkl type carrying cyclic reflections. The strategy is to work on the unit circle: the authors define an operator L(N) acting on the CMV Laurent polynomials associated with the sieved Jacobi OPUC, state that these polynomials satisfy L(N)ψ_n = λ_n(N)ψ_n, and then form H(N)=L(N)^2-N(α+β+1)L(N). Using the Szegő correspondence, they derive eigenvalue equations for the sieved Jacobi polynomials of the first and second kind on the real line, with eigenvalues Λ_n(N)=n(n+N(α+β+1)) and Λ_{n+1}(N) respectively. The ultraspherical specialization is also presented. The central technical step is Theorem 5.4, whose proof is sketched and relies on the summation identities of Lemma 5.3.","tokens_in":15061,"tokens_out":7617,"duration_ms":81215,"significance":"If the central result is correct, it settles a question posed by Askey, provides a concrete Dunkl-type operator with cyclic reflections for which the sieved Jacobi polynomials are eigenfunctions, and establishes the bispectrality of these polynomials and of their CMV counterparts. The construction is elegant and fits naturally into the framework previously developed by the authors for Bannai–Ito polynomials. The paper also gives explicit formulas for the second-order operator H(N) and its conjugation for the second-kind polynomials, which is valuable for applications. However, the paper's main theorem is not fully proven in the manuscript, and one of the key summation lemmas is false as stated. These issues must be addressed before the claim can be considered established.","major_comments":[{"comment":"The identities (5.8) and (5.9) are false for odd j. For N=2, q=-1, j=1, z=2, the left-hand side of (5.8) equals -4/3 + 4i/5 while the right-hand side equals -16/15; for (5.9) the left-hand side is -4/3 - 4i/5 while the right-hand side is -4/15. The derivation from (5.12) can only produce these formulas when j is even and the parity of N is compatible; as stated, the lemma overreaches. Since the proof of Theorem 5.4 invokes these identities and then claims that all other parity cases are analogous, the odd-j cases are not established by the given argument. Please provide a correct version of the summation identities that covers all parity cases, or restrict the statement of the lemma and prove the remaining cases separately.","section":"Lemma 5.3, Eqs. (5.8)–(5.9)"},{"comment":"The central eigenvalue equation (5.13) is not proven in the manuscript. The proof explicitly treats only the case where n, N, and j are all even, and then says that all other possible situations can be treated analogously. Because the analogous situations include odd j, for which Lemma 5.3 as stated fails, this is not a harmless omission. Moreover, the proof of the key summation Lemma 5.3 is delegated to the authors' preprint [9], and the verification of the vanishing of F_n^{(1)} and F_n^{(2)} is not displayed. Since Theorem 6.1 and Corollaries 7.1–7.2 all depend on Theorem 5.4, the paper's headline result is not self-contained. The authors should either include a complete proof of Theorem 5.4 for all parity cases or clearly present the paper as an exposition of results proved in [9], with Theorem 5.4 stated as a quoted theorem and the missing proof supplied or referenced in a complete form.","section":"Theorem 5.4, proof"}],"minor_comments":[{"comment":"The title contains a typo: 'ultrasperical' should be 'ultraspherical'.","section":"Section 7 title"},{"comment":"The word 'Hovever' should be 'However'.","section":"Remark 6.3"},{"comment":"The phrase 'with the the relation' should be 'with the relation'.","section":"Proof of Theorem 5.4"},{"comment":"The lemma introduces a variable h but the identities are stated in terms of j; please clarify the relationship between h and j and explicitly state the parity assumptions on j under which (5.8) and (5.9) are claimed to hold.","section":"Lemma 5.3"}],"recommendation":"major_revision","confidential_remarks":"The manuscript depends heavily on the authors' own work [8] and [9] for the main lemmas and theorem. If [9] is already accepted, the current paper could serve as an announcement, but the false statement of Lemma 5.3 is a serious issue that must be corrected in any case. The refereeing process should insist on a complete and correct proof of Theorem 5.4, at least for the parity cases that are needed, before the paper can be accepted."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know two things before reading this paper. First, it gives explicit Dunkl-type eigenvalue equations for sieved Jacobi polynomials, which would confirm Askey's old conjecture. Second, the key summation lemma that powers the proof is false as stated for odd j, so the central theorem is not established the way the paper claims.\n\nWhat is genuinely new and useful: the explicit expression for H(N) = L(N)^2 - N(α+β+1)L(N), the companion equation for the second-kind polynomials Q_n, and the simplified ultraspherical corollaries. The N=1 base case and much of the bispectrality result already appear in the authors' own [8] and [9]; this paper is more of an expanded exposition than a first proof. That said, the explicit formulas are valuable, and the self-adjointness proof for L(N) in Theorem 5.5 is clean and convincing.\n\nThe soft spot is not just a missing detail; it is a false identity. Lemma 5.3 states (5.8) and (5.9) for all integers h < N, but for N=2, j=1, q=-1, z=2, the left side of (5.8) is -4/3 + 4i/5 while the right side is -16/15, and similarly for (5.9). The residue method in the proof only produces (5.12) for integer h; the half-integer exponents in (5.8) and (5.9) are not specializations of it. The proof of Theorem 5.4 uses these sums to cancel the reflection terms, and the remaining parity cases are asserted without details. So as written, the proof of the main theorem does not go through. This is repairable—one can likely derive the correct odd-j identities from the same contour-integral technique—but it is a load-bearing error, not a cosmetic one.\n\nMy recommendation: this paper deserves a serious referee, but not acceptance in present form. The referee should ask for a corrected Lemma 5.3 (or a restriction to even j with a separate argument for odd j), a detailed verification of all parity cases in Theorem 5.4, and a clearer statement of what is new relative to [9]. The special-functions community would benefit from the explicit formulas once the proof is fixed.","headline":"Central lemma is false as stated for odd j, so the main theorem's proof has a real hole; the explicit operator formulas and self-adjointness argument still make it worth a referee's time.","tokens_in":15649,"tokens_out":3996,"would_cite":false,"duration_ms":39695,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["33C45","42C05"],"pacs":[],"model":"deepseek-v4-flash","headline":"The sieved Jacobi polynomials—obtained when q is a root of unity—admit an explicit Dunkl-type eigenvalue equation, confirming their bispectrality.","keywords":["sieved Jacobi polynomials","Dunkl-type operators","bispectrality","orthogonal polynomials on the unit circle","CMV Laurent polynomials","cyclic reflections","eigenvalue equations","root-of-unity limits"],"falsifier":"Take a small sieving parameter such as N=2 or N=3 and low degrees n, substitute the explicit formulas into H(N)P_n(x(z);N)−Λ_n(N)P_n(x(z);N) symbolically, and check for a nonzero result; alternatively, evaluate the identities (5.8)-(5.9) directly for specific values of N, h, and j by contour integration or residue extraction, which would expose any hidden failure in the parity cases.","tokens_in":14551,"feed_emoji":"🧮","tokens_out":9605,"duration_ms":97371,"temperature":0.7,"pith_summary":"Sieved Jacobi polynomials—families of orthogonal polynomials obtained when the sieving parameter q is set to a primitive N-th root of unity—are eigenfunctions of a closed differential-difference operator of Dunkl type with cyclic reflections. The paper constructs this operator first on the unit circle for the associated CMV Laurent polynomials, then squares it with a linear correction to act on the real-line polynomials of the first and second kind, giving eigenvalues Λ_n(N)=n(n+N(α+β+1)) and their shifted companion values. If the construction is correct, it settles the long-standing conjecture, dating back to the introduction of these polynomials, that they admit an eigenvalue equation and are therefore bispectral. The first- and second-kind equations are coordinated: the same operator, up to a similarity transform, diagonalizes both, with the second-kind spectrum shifted by one.","feed_headline":"Sieved Jacobi polynomials diagonalize a Dunkl operator","feed_subtitle":"An explicit operator now gives the sieved polynomials eigenvalues n(n+N(α+β+1)).","key_machinery":"The load-bearing object is the first-order operator L(N)=z∂_z+Σ_{k=0}^{N−1} A_k(z;N)(R_k−I), where R_k f(z)=f(q^k/z) are cyclic reflections at the N-th roots of unity and the rational coefficients A_k are given by (5.15)-(5.16). Its eigenvalue equation on the sieved CMV Laurent polynomials ψ_n(z;N) (Theorem 5.4) is verified by expressing those Laurent polynomials through the ordinary N=1 Jacobi OPUC and using the summation identities (5.8)-(5.9) to cancel every unwanted reflection term; the remaining piece is the Dunkl operator K of the N=1 case. The passage from L(N) to H(N)=L(N)^2−N(α+β+1)L(N) then pairs the eigenvalues of ψ_{2n} and ψ_{2n−1}, which is exactly what lets the real-line polynomials P_n and Q_n be diagonalized by one quadratic expression.","core_discovery":"The paper's central claim is Theorem 6.1: for every positive integer N and real α, β, the sieved Jacobi polynomials of the first kind P_n(x(z);N) satisfy H(N)P_n(x(z);N)=Λ_n(N)P_n(x(z);N) and those of the second kind Q_n satisfy the companion equation with Λ_{n+1}(N), where x(z)=z+1/z, Λ_n(N)=n(n+N(α+β+1)), and H(N)=L(N)^2−N(α+β+1)L(N) with L(N) the first-order Dunkl-type operator of Theorem 5.4. This gives the sieved Jacobi families the same bispectral status as classical Jacobi and Bannai–Ito polynomials: a block recurrence relation plus an explicit difference-differential eigenvalue equation. The paper further shows (Theorem 6.2) that on symmetric Laurent polynomials H(N) takes the explicit second-order form $z^{2}$∂$_z^{2}$+C(z)∂_z plus a sum of reflection (equivalently rotation) difference terms, and that L(N) is self-adjoint on the unit circle with respect to the sieved Jacobi weight.","pith_inferences":["The method suggests a natural next step the paper leaves open: identify the algebra generated by H(N) and multiplication by x, extending the circle Jacobi algebra of the N=1 case to arbitrary sieving parameters.","A direct symbolic check of the parity cases and the two summation identities for small N (say N=2,3) would be a cheap, sharp test of the construction before extending it to other sieved families such as Pollaczek.","The pairing of eigenvalues via L(N)^2−cL(N) is likely a general mechanism: any family whose Laurent polynomials are eigenfunctions of a Dunkl operator with alternating eigenvalue signs will inherit a second-order eigenvalue equation on the real line.","One could test the rotation form of the operator on non-symmetric Laurent polynomials to see whether the reflection and rotation versions differ by a genuine invariant, possibly yielding additional commuting operators."],"forward_implications":["The sieved Jacobi polynomials of first and second kind are bispectral: besides their block recurrence relations they satisfy a closed eigenvalue equation in the variable x.","The spectrum Λ_n(N)=n(n+N(α+β+1)) is explicit and interpolates the classical Jacobi spectrum at N=1, providing a direct check of the construction.","In the ultraspherical case α=β, the operators simplify so that the first kind is diagonalized with eigenvalues n(n+N(2α+1)) and the second kind, after standardization, with n(n+N(2α+1)+2).","Because L(N) is self-adjoint with respect to the sieved Jacobi weight on the unit circle, the eigenvalue equations define genuine spectral problems rather than formal identities.","On symmetric Laurent polynomials the reflection operators R_k act like rotations T_{−k}, so the same operator can be written as a second-order differential operator plus rotation differences (Theorem 6.2)."],"supporting_citations":[{"why":"Introduces the sieved ultraspherical polynomials and raises the question of their differential equation.","marker":"[4]"},{"why":"Establishes the unit-circle version of sieved polynomials and the correspondence to the real-line first- and second-kind families.","marker":"[7]"},{"why":"Derives the CMV bispectrality of the Jacobi OPUC (the N=1 case) with the operator K and the circle Jacobi algebra.","marker":"[8]"},{"why":"Companion preprint that supplies the proof of the summation identities (Lemma 5.3) and the full parity-case verification of Theorem 5.4.","marker":"[9]"},{"why":"Provides the five-diagonal CMV formalism used to define the Laurent polynomials.","marker":"[10]"},{"why":"The survey in which the sieved Jacobi polynomials of the first and second kind were introduced.","marker":"[11]"},{"why":"Standard reference for the Jacobi differential equation and identities used in Lemma 5.1.","marker":"[20]"},{"why":"Classical source of the Szegő map relating OPUC to orthogonal polynomials on the real line.","marker":"[27]"}],"fun_headline_variants":["Askey's sieved Jacobi conjecture proven: Dunkl operator diagonalized","Sieved Jacobi polynomials: explicit eigenvalues from a Dunkl operator","New proof: Askey's sieved Jacobi polynomials are eigenfunctions","Dunkl operator eigenvalue equation settles Askey's sieved Jacobi question"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof relies on the two summation identities of Lemma 5.3, whose verification is deferred to the authors' companion preprint, and on the assertion that all omitted parity combinations of n, N, and j in Theorem 5.4 work out analogously; if any of these fails, the operator L(N) would not be diagonal on the sieved Laurent polynomials.","fun_headline_variants_meta":{"raw":{"variants":["Askey's sieved Jacobi conjecture proven: Dunkl operator diagonalized","Sieved Jacobi polynomials: explicit eigenvalues from a Dunkl operator","New proof: Askey's sieved Jacobi polynomials are eigenfunctions","Dunkl operator eigenvalue equation settles Askey's sieved Jacobi question"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00065,"raw_usage":{"total_tokens":2941,"prompt_tokens":865,"completion_tokens":2076,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":481,"completion_tokens_details":{"reasoning_tokens":1996}},"tokens_in":481,"tokens_out":2076,"duration_ms":15908,"temperature":1.0,"reasoning_tokens":1996,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T20:01:49.393451+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a small sieving parameter such as N=2 or N=3 and low degrees n, substitute the explicit formulas into H(N)P_n(x(z);N)−Λ_n(N)P_n(x(z);N) symbolically, and check for a nonzero result; alternatively, evaluate the identities (5.8)-(5.9) directly for specific values of N, h, and j by contour integration or residue extraction, which would expose any hidden failure in the parity cases.","supporting_citations":[{"cited_title":"Sieved ultraspherical polynomials","cited_arxiv_id":null,"evidence_quote":"Introduces the sieved ultraspherical polynomials and raises the question of their differential equation."},{"cited_title":"On sieved orthogonal polynomials","cited_arxiv_id":null,"evidence_quote":"Establishes the unit-circle version of sieved polynomials and the correspondence to the real-line first- and second-kind families."},{"cited_title":"The CMV bispectrality of the Jacobi polynomials on the unit circle","cited_arxiv_id":null,"evidence_quote":"Derives the CMV bispectrality of the Jacobi OPUC (the N=1 case) with the operator K and the circle Jacobi algebra."},{"cited_title":"Bispectrality of the sieved Jacobi polynomials","cited_arxiv_id":"2501.12806","evidence_quote":"Companion preprint that supplies the proof of the summation identities (Lemma 5.3) and the full parity-case verification of Theorem 5.4."},{"cited_title":"Five-diagonal matrices and zeros of orthogonal polyno- mials on the unit circle","cited_arxiv_id":null,"evidence_quote":"Provides the five-diagonal CMV formalism used to define the Laurent polynomials."},{"cited_title":"Orthogonal polynomials old and new, and some combinatorial connections","cited_arxiv_id":null,"evidence_quote":"The survey in which the sieved Jacobi polynomials of the first and second kind were introduced."},{"cited_title":"Hypergeometric orthogonal polynomials","cited_arxiv_id":null,"evidence_quote":"Standard reference for the Jacobi differential equation and identities used in Lemma 5.1."},{"cited_title":"American Mathematical Soc., 1939","cited_arxiv_id":null,"evidence_quote":"Classical source of the Szegő map relating OPUC to orthogonal polynomials on the real line."}],"review_version":1}