{"id":"f0966300-3697-4415-b281-bbf3e6d26b48","arxiv_id":"2506.03432","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A general explicit first and second order differential equation, plus a Verblunsky-coefficient difference equation, is established for semiclassical orthogonal polynomials on the unit circle with Pearson weights of degree at most three.","lead":"These authors derive explicit differential equations for orthogonal polynomials on the unit circle whose weight obeys a Pearson-type equation of degree at most three. The result extends earlier degree-two work and links the polynomials' Verblunsky coefficients to a discrete Painlevé II equation.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Boundary condition A(1)[w(2π)-w(0)]=0 is load-bearing and not automatic: weights like e^{-ηθ} (A=1, B=-η) satisfy (1.5) with deg≤3 but violate it, so the results do not cover the whole class (p,q), p,q≤3.","rationale":"The reader's weakest assumption correctly identifies the boundary condition A(1)[w(2π)-w(0)]=0 as the load-bearing premise. Reading the paper in good faith, the main derivation is internally coherent: Theorem 2.1 is explicitly conditional on this boundary hypothesis, and the examples all satisfy it. The concern is not an internal inconsistency but an overstatement of scope in the abstract and in the reader's paraphrase 'the whole class (p,q) with p,q≤3'. The provided concrete counterexample shows the condition is genuinely restrictive and not automatic for the Pearson class, so without adding the condition to the class definition the central claim is only proven for a proper subclass. This does not invalidate the theorems as stated; it clarifies that the advertised generality should be narrowed. Hence the reader's CONDITIONAL verdict remains appropriate, with no change required.","tokens_in":22961,"tokens_out":27408,"duration_ms":268170,"concrete_test":"For the weight w(θ)=e^{-θ} (η=1), take A(z)=1, B(z)=-1, so (1.5) holds with deg≤3 and A(1)[w(2π)-w(0)]=e^{-2π}-1≠0. Compute the MOPUC Φ_0,...,Φ_4 by Gram-Schmidt against the inner product ∫_0^{2π} f(e^{iθ}) overline{g(e^{iθ})} e^{-θ} dθ. Evaluate (2.2) numerically for n=4 and k=2: ⟨Φ'_4,z^2⟩ should equal -iΦ_4(1)(e^{-2π}-1) ≠ 0. Then repeat the coefficient-elimination step in the proof of Theorem 2.1 for n=4 and check whether s_{4,2} vanishes; it will not, confirming that the boundary condition is essential and the theorem does not apply to this weight.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The proof of Theorem 2.1 uses (2.2) to annihilate the coefficients s_{n,k} for k=2,...,n-2: the boundary term -iΦ_n(1)A(1)[w(2π)-w(0)] in (2.1) must vanish. This condition is satisfied by all five examples (mostly through A(1)=0, or in Examples 4-5 through periodicity), but it is not a consequence of the Pearson equation (1.5) with deg A, deg B≤3. A concrete counterexample in the class is w(θ)=e^{-ηθ}, η≠0, with A(z)=1, B(z)=-η: it satisfies (1.5) with degrees 0, yet A(1)[w(2π)-w(0)]=e^{-2πη}-1≠0. For this weight the right-hand side of (2.2) is nonzero, so the elimination of s_{n,2},...,s_{n,n-2} in Theorem 2.1 fails; consequently the formulas for U_n,V_n,W_n,Y_n in Theorem 3.1 do not hold. The theorem statements themselves include the boundary hypothesis, but the abstract's unqualified 'a class of semiclassical orthogonal polynomials' and the reader's paraphrase 'the whole class (p,q) with p,q≤3' overstate the proven scope unless the class is explicitly defined by the boundary condition.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies semiclassical orthogonal polynomials on the unit circle whose weight satisfies the Pearson-type equation (1.5) with A and B polynomials of degree at most three. Under the hypothesis A(1)[w(2π)-w(0)]=0, the authors derive structure relations for the monic orthogonal polynomials (Theorem 2.1), first-order differential equations for Φ_n and Φ*_n (Theorem 3.1), second-order equations (Theorems 3.6 and 3.8), and difference equations for the Verblunsky coefficients (Theorem 2.3). Five families of weights are then treated as applications, including generalizations of circular Jacobi and modified Bessel polynomials, and a discrete Painlevé II equation is exhibited for the modified circular Jacobi case. The proofs are detailed and proceed by standard orthogonality and integration-by-parts arguments.","tokens_in":23260,"tokens_out":6266,"duration_ms":73196,"significance":"If the scope is stated precisely, the paper gives useful and explicit structure relations and differential equations for a substantial family of semiclassical OPUC. The coefficient formulas are concrete, the difference equations for Verblunsky coefficients are of independent interest, and the applications to known families and to discrete Painlevé II are valuable. The comparison with prior work [4] and the numerical verification in Section 5 are appropriate. The main theorems appear correct under the stated boundary hypothesis, and I found no circularity.","major_comments":[{"comment":"The boundary hypothesis A(1)[w(2π)-w(0)]=0 is load-bearing for the whole development, since Eq. (2.2) is used in the proof of Theorem 2.1 to force s_{n,k}=0 for k=2,...,n-2. This condition is not a consequence of the Pearson equation (1.5) with deg A, deg B ≤ 3. For example, w(θ)=e^{-ηθ}, η≠0, with A(z)=1 and B(z)=-η satisfies (1.5) and belongs to the class (0,0), but A(1)[w(2π)-w(0)]=e^{-2πη}-1≠0, so Eq. (2.2) fails and the structure relation formulas are not established for this weight. The theorems themselves state the hypothesis, but the Abstract and the final paragraph ('we provided explicit first and second order differential equations for semiclassical MOPUC belonging to the class (p,q), with p≤3 and q≤3') overstate the proven scope. Please redefine the class under study to include the boundary condition, or explicitly state that all results are for weights satisfying (1.5) and A(1)[w(2π)-w(0)]=0, in the Abstract, Introduction, and Section 5.","section":"§2, Eq. (2.2), Theorem 2.1, and §5 final paragraph"},{"comment":"The final step of the proof, in which the coefficient r_n is set to zero, is too terse. The displayed identity equates a polynomial to (r_n z)/(r A(z)) Φ_n(z); to conclude r_n=0 one must argue that A(z)=(z-r)(z-1/r) has exactly one zero outside the unit disk, that the other zero lies inside, that z=0 is not a zero of A, and that all zeros of Φ_n lie inside the unit disk, so no cancellation with Φ_n(z) can remove the pole at the outside zero unless r_n=0. Please spell out this divisibility argument explicitly, since this step is needed for the claimed homogeneous linear second-order equation in Corollary 4.27.","section":"§4.5, proof of Corollary 4.27"}],"minor_comments":[{"comment":"The phrase 'a weight function that satisfy a Pearson-type differential equation' should read 'that satisfies'; also consider mentioning the boundary condition in the Abstract to avoid the scope overstatement noted above.","section":"Abstract"},{"comment":"The displayed formula for ℓ_{n,n-1} contains the term α_{n−1}α_n − α_{n−1}α_n, which cancels identically; this is likely a typographical error, and the intended expression involving complex conjugates should be corrected.","section":"§4.2, display for ℓ_{n,n-1}"},{"comment":"The phrase 'From Proposition 4.2 and (4.11) of [4]' is ambiguous because the current paper has no Proposition 4.2; please cite the result as 'Proposition 4.2 of [4]' or restate the needed identity.","section":"§4.1, proof of Corollary 4.3"},{"comment":"The text contains the typo 'discrete Painvel´ e equation' where 'Painlevé' is intended.","section":"§4.4.1"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of the journal and the main mathematical development appears sound. The decision should rest on whether the authors can fix the class definition to include the boundary hypothesis explicitly and supply the missing justification in Corollary 4.27; I do not see grounds for rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things up front. The paper is a real extension of the authors' earlier degree-two program: Theorem 2.1 gives structure relations for degree-three Pearson weights, Theorem 2.3 gives difference equations for the Verblunsky coefficients, and Theorems 3.1, 3.6, and 3.8 turn them into explicit first- and second-order ODEs for Phi_n and Phi*_n. The formulas are concrete, the proof route is standard--integration by parts plus orthogonality--and there is no circularity. The five worked examples, including the Jacobi-type and modified Bessel families and the discrete Painleve II connection, show the machinery working and are a useful check.\n\nThe soft spot is the boundary condition. Equation (2.1) carries the boundary term -i Phi_n(1) A(1)[w(2pi)-w(0)], and it must vanish for the elimination of the middle s_{n,k} in Theorem 2.1. The theorem statements do include that hypothesis, but the abstract and the phrase 'class (p,q), p,q <= 3' invite the reader to think the results cover every Pearson weight with degree at most three. They do not. For example, w(theta)=e^{-eta theta}, eta != 0, satisfies (1.5) with A=1, B=-eta, yet A(1)[w(2pi)-w(0)] = e^{-2pi eta}-1 != 0. The results simply do not apply to that weight. This is not fatal; the condition is natural and all the examples satisfy it. Still, the authors should describe the scope as 'class (p,q), p,q <= 3, plus the boundary condition,' and ideally say what the condition means in practice, such as periodic weights or A(1)=0.\n\nMinor point: Corollary 4.27 sets r_n=0 with a one-line divisibility and zero-location argument. The conclusion is right--if A has a zero outside the unit disk and Phi_n has all zeros inside, the rational term can only be polynomial if r_n=0--but the printed text compresses the step more than the rest of the paper does.\n\nThe citations to [4] are legitimate: that paper is the degree-two base case, and the present work builds on it openly rather than hiding it. The numerical generation of Verblunsky coefficients in Section 5 is a small but honest extra.\n\nBottom line: this deserves a serious referee. It is a careful, useful reference for anyone working on semiclassical OPUC. My own verdict is accept after a minor revision that fixes the scope statement and expands the r_n step.","headline":"A genuine degree-three extension with explicit structure relations and differential equations; the main theorems are correctly hedged, but the advertised class needs a boundary-condition qualifier and one coefficient elimination needs more detail.","tokens_in":23803,"tokens_out":3585,"would_cite":true,"duration_ms":41679,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["42C05","33C47"],"pacs":[],"model":"deepseek-v4-flash","headline":"A degree-three Pearson-type weight equation forces explicit differential equations for unit-circle orthogonal polynomials.","keywords":["orthogonal polynomials on the unit circle","semiclassical weight functions","Pearson-type differential equation","structure relations","difference equations","discrete Painlevé II","Verblunsky coefficients","Jacobi polynomials on the unit circle"],"falsifier":"Take a weight satisfying all hypotheses, such as $w(\\theta)=e^{-\\eta\\theta}[\\sin^2(\\theta/2)]^{\\lambda}[\\cos^2(\\theta/2)]^{\\beta}$ with $\\eta,\\lambda,\\beta$ nonzero, construct $\\Phi_3$ explicitly from the Szegő recurrences and the known initial Verblunsky coefficients, and check whether $zA(z)\\Phi_3'(z)-U_3(z)\\Phi_3(z)-V_3(z)\\Phi_3^*(z)$ is identically zero with $U_3,V_3$ as in Theorem 3.1. Any nonzero polynomial coefficient would disprove the structure relation; the same check with $\\beta=0$ should reproduce Corollary 4.3 as a control.","tokens_in":22767,"feed_emoji":"📐","tokens_out":10321,"duration_ms":106284,"temperature":0.7,"pith_summary":"The paper shows that a single differential assumption on the weight—a Pearson-type equation $d[A(e^{i\\theta})w(\\theta)]/d\\theta = B(e^{i\\theta})w(\\theta)$ with $A,B$ complex polynomials of degree at most three—controls the whole orthogonal-polynomial family on the unit circle. Once the boundary condition $A(1)[w(2\\pi)-w(0)]=0$ holds, the monic orthogonal polynomials $\\Phi_n$ and their reversed polynomials $\\Phi_n^*$ satisfy explicit first-order differential systems, and the same machinery yields second-order scalar equations for both. The same structure relation produces nonlinear difference equations for the Verblunsky coefficients, including a discrete Painlev\\'e II equation in symmetric cases. Applications give concrete differential equations for families generalizing Jacobi polynomials on the unit circle and for modified Bessel polynomials, recovering several known equations as special cases. The value of the result is that it converts a hypothesis about the weight into ready-to-use formulas in the coefficients of the polynomials.","feed_headline":"Degree-three semiclassical weights pin down circle polynomials","feed_subtitle":"A boundary condition on the weight equation unlocks explicit first- and second-order equations and new Verblunsky recurrences.","key_machinery":"The engine of the argument is the vanishing identity (2.2): under the boundary hypothesis, $\\langle A(z)\\Phi_n'(z), z^k\\rangle = 0$ for the middle range $k=2,\\dots,n-2$. This forces the expansion of $A(z)\\Phi_n'(z)-n a_3\\Phi_{n+2}(z)$ into the orthogonal basis to collapse to five terms, producing the structure relation of Theorem 2.1 with coefficients $s_{n,n+1},s_{n,n},s_{n,n-1},p_{n,n},t_{n,n}$ computed from $A,B,\\alpha_n,\\ell_{n,k}$. Szeg\\H{o}'s recurrences for the reversed polynomial $\\Phi_n^*(z)=z^n\\Phi_n(1/z)$ then convert that relation into the first-order systems (3.1)–(3.2), and elimination via an integrating factor yields the second-order equations. The named objects carrying the proof are the monic orthogonal polynomials on the unit circle (MOPUC), the reversed polynomial, and the Verblunsky coefficients—the constants $\\alpha_n$ that parametrize the Szeg\\H{o} recurrences.","core_discovery":"The central claim is that whenever a positive weight $w$ satisfies (1.5) with $\\deg A,\\deg B \\le 3$ and $A(1)[w(2\\pi)-w(0)] = 0$, the associated MOPUC obey the structure relation of Theorem 2.1, from which the paper derives the first-order systems (3.1)–(3.2) for $\\Phi_n$ and $\\Phi_n^*$ and then explicit second-order differential equations in Theorems 3.6 and 3.8. All coefficients $U_n,V_n,W_n,Y_n$ are written explicitly in terms of $A$, $B$, the Verblunsky coefficients $\\alpha_n$, and the sub-leading coefficients $\\ell_{n,k}$. Theorem 2.3 gives a difference equation for the $\\alpha_n$; for the symmetric weight $e^{t\\cos\\theta}[\\sin^2(\\theta/2)]^\\lambda$ this becomes the discrete Painlev\\'e II equation (1.7). The applications show that modified Bessel polynomials, circular Jacobi polynomials, Jacobi polynomials on the unit circle, and a further two-parameter family all fall under the same formulas, with earlier differential equations recovered as limiting cases.","pith_inferences":["If the boundary condition $A(1)[w(2\\pi)-w(0)]=0$ is dropped, carrying the extra term $-i\\Phi_n(1)A(1)[w(2\\pi)-w(0)]$ through the same projection argument should produce a nonhomogeneous correction in the structure relation and the differential systems; this is a natural next test for quasi-semiclassical circle weights.","The pattern of the coefficients suggests a plausible extension to $A,B$ of arbitrary degree $d$, with a structure relation whose number of terms grows linearly in $d$; trying $d=4$ would reveal whether the coefficient construction remains closed or new obstructions appear.","Because the symmetric-case recurrence is a discrete Painlev\\'e II equation, the numerical scheme of Section 5 could be pushed further to scan the $(\\lambda,t)$ parameter plane and locate transitions in $|\\alpha_n|$; the paper only displays a few parameter choices."],"forward_implications":["Every monic orthogonal polynomial in the class with $p,q\\le 3$ comes with explicit first- and second-order differential equations expressed directly through $A$, $B$, the Verblunsky coefficients, and the sub-leading coefficients $\\ell_{n,k}$.","Evaluating the structure relation at $z=0$ yields nonlinear difference equations for the Verblunsky coefficients; for symmetric weights such as $e^{t\\cos\\theta}[\\sin^2(\\theta/2)]^\\lambda$ these reduce to the discrete Painlev\\'e II equation (1.7).","Known families—modified Bessel polynomials, circular Jacobi polynomials, Jacobi polynomials on the unit circle, and the weight of Example 2—are covered by one set of formulas, and the differential equations recover earlier results as special cases.","The second-order equation for $\\Phi_n$ is linear when $V_n'(z)=0$; when $V_n'\\ne 0$, Corollary 3.7 supplies an explicitly linear second-order equation under the condition $V_n\\ne 0$, with the analogous statement for $\\Phi_n^*$ via $W_n$.","The discrete recurrences give a practical way to generate the Verblunsky coefficients numerically, avoiding the increasingly costly determinant evaluations of Heine's formula, as demonstrated in Section 5."],"supporting_citations":[{"why":"Supplies the degree-two structure relations and difference equations that this paper extends to degree three.","marker":"[4]"},{"why":"Provides the semiclassical functional framework and the integration-by-parts identity behind (2.1).","marker":"[20]"},{"why":"Gives the standard MOPUC definitions, Szegő recurrences, and Heine formula used throughout.","marker":"[25]"},{"why":"Supplies the earlier differential-equation method and comparison cases for Jacobi and circular Jacobi polynomials.","marker":"[16]"},{"why":"Provides the hypergeometric weight example whose differential equation is recovered in Example 1.","marker":"[26]"},{"why":"Derives the discrete Painlevé II relation for Verblunsky coefficients used in the Bessel case.","marker":"[24]"},{"why":"Documents the discrete Painlevé II connection and the modified Bessel polynomial context.","marker":"[28]"},{"why":"Gives equivalent first-order equations in a Riemann-Hilbert setting, used to cross-check the modified Bessel equations.","marker":"[5]"},{"why":"Supplies the explicit Jacobi-on-unit-circle expressions and Verblunsky coefficients used in Example 3's special case.","marker":"[2]"}],"fun_headline_variants":["Degree-three semiclassical weights yield explicit ODEs","Circle polynomials: first- and second-order equations uncovered","Pearson-type weights produce new circle polynomial equations","Explicit differential equations for semiclassical circle polynomials","Discrete Painlevé II emerges from degree-three weight condition"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is the boundary condition $A(1)[w(2\\pi)-w(0)] = 0$; if a weight in this Pearson class has $w(2\\pi)\\ne w(0)$ with $A(1)\\ne 0$, the key identity (2.2) fails and the structure relation must be modified before any of the differential equations can be derived.","fun_headline_variants_meta":{"raw":{"variants":["Degree-three semiclassical weights yield explicit ODEs","Circle polynomials: first- and second-order equations uncovered","Pearson-type weights produce new circle polynomial equations","Explicit differential equations for semiclassical circle polynomials","Discrete Painlevé II emerges from degree-three weight condition"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000222,"raw_usage":{"total_tokens":1426,"prompt_tokens":889,"completion_tokens":537,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":505,"completion_tokens_details":{"reasoning_tokens":459}},"tokens_in":505,"tokens_out":537,"duration_ms":6329,"temperature":1.0,"reasoning_tokens":459,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T11:03:56.627004+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a weight satisfying all hypotheses, such as $w(\\theta)=e^{-\\eta\\theta}[\\sin^2(\\theta/2)]^{\\lambda}[\\cos^2(\\theta/2)]^{\\beta}$ with $\\eta,\\lambda,\\beta$ nonzero, construct $\\Phi_3$ explicitly from the Szegő recurrences and the known initial Verblunsky coefficients, and check whether $zA(z)\\Phi_3'(z)-U_3(z)\\Phi_3(z)-V_3(z)\\Phi_3^*(z)$ is identically zero with $U_3,V_3$ as in Theorem 3.1. Any nonzero polynomial coefficient would disprove the structure relation; the same check with $\\beta=0$ should reproduce Corollary 4.3 as a control.","supporting_citations":[{"cited_title":"Ifd= 3 we obtain ⟨A(z)Φ′ n, zk⟩=−iΦ n(1)A(1)[w(2π)−w(0)], k= 2,","cited_arxiv_id":null,"evidence_quote":"Supplies the degree-two structure relations and difference equations that this paper extends to degree three."},{"cited_title":"de Jesus and J","cited_arxiv_id":null,"evidence_quote":"Provides the semiclassical functional framework and the integration-by-parts identity behind (2.1)."},{"cited_title":"Ismail and N.S","cited_arxiv_id":null,"evidence_quote":"Gives the standard MOPUC definitions, Szegő recurrences, and Heine formula used throughout."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the earlier differential-equation method and comparison cases for Jacobi and circular Jacobi polynomials."},{"cited_title":"Jones, O","cited_arxiv_id":null,"evidence_quote":"Provides the hypergeometric weight example whose differential equation is recovered in Example 1."},{"cited_title":"Further results involving orthogonal polynomials and Painlev´ e equations (continuous and discrete) can be found for instance in [12, 18, 19, 28] and references therein","cited_arxiv_id":null,"evidence_quote":"Derives the discrete Painlevé II relation for Verblunsky coefficients used in the Bessel case."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Documents the discrete Painlevé II connection and the modified Bessel polynomial context."},{"cited_title":"As a consequence, we obtained difference equations for the Verblunsky coefficients","cited_arxiv_id":null,"evidence_quote":"Gives equivalent first-order equations in a Riemann-Hilbert setting, used to cross-check the modified Bessel equations."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the explicit Jacobi-on-unit-circle expressions and Verblunsky coefficients used in Example 3's special case."}],"review_version":1}