{"id":"1c833961-5647-49cf-9cfb-4077b73b2dec","arxiv_id":"2506.17706","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The paper proves the KKLS conjecture that averaged gl-weight systems equal combinations of Schur functions with Bernoulli polynomial coefficients, and establishes the analogous q-deformed formula for quantum gl-weight systems on Hecke algebras.","lead":"Zaitsev proves a conjecture of Kazarian, Krasilnikov, Lando and Shapiro: the average value of the universal gl-weight system on permutations has an explicit closed form as a combination of Schur functions with Bernoulli polynomial coefficients. To prove it he introduces a quantum deformation of the weight system and shows its average has a parallel q-Bernoulli polynomial formula.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central identity is only as reliable as the imported RE-algebra Harish-Chandra isomorphism and the singular change of variables used to take q to 1; neither is justified in the text.","rationale":"The paper's central claim is Theorem 4, and the proof chain is: RE-algebra Harish-Chandra identification (4.7)-(4.8), Proposition 24 for chi_m(h_m), Corollary 18/Proposition 25 for Omega_m, Corollary 27 with the q-Bernoulli limit, and Theorem 21 for the limit to W_m. Every step after the Harish-Chandra identification is algebraic manipulation, but that identification is the place where the quantum objects are converted into the symmetric polynomials S_k in which the final formula is stated. The paper imports this from [GPS] and gives no independent normalization check. That would be acceptable if the subsequent limit were transparent, but it is not: the variables x_i differ from the xi_i in which chi_m is naturally expressed by a factor q^{1-N-xi_i}/(q^2-1), which is singular at q=1. Remark 3's coefficient-wise limit prescription is an assertion, not a proof; if the correct Harish-Chandra normalization contains an extra q-power, the Schur-function form of Omega_m and its limit would shift. The undefined index l in Eq. (6.2) and the disagreement between the Bernoulli orders in (2.1) and (6.1) are exactly the kind of normalization inconsistency that would arise from such an error. I am not claiming the theorem is false; the external evidence ([GPS], [JLM], [Ok], and the agreement with the KKLS conjecture) makes it plausible. But the printed proof is not checkable as written at its most load-bearing step, so the verdict should remain conditional pending an independent small-case verification.","tokens_in":19927,"tokens_out":20658,"duration_ms":192377,"concrete_test":"Use a symbolic CAS to verify the displayed formulas from first principles for N=2,3 and m=1,2,3. Build the R-matrix (4.2), the Reflection Equation generators M via (4.4), L=(1-M)/(q-q^{-1}), and the q-symmetrizer H(m); compute Omega_m(N)=<L_1...L_m H(m)> directly from Definition 14. Reduce the result to symmetric polynomials in x_i using (4.8), compare it with Eq. (6.1), and compare the q to 1 limit with W_m(N) from Definition 1/Proposition 2. If a mismatch appears, test the two candidate Bernoulli orders from (2.1) and (6.1) to determine which is correct. Also compute p_m from (4.5) and verify Eq. (4.7) for the same small N and m, checking whether q^{-2xi_i} and x_i satisfy the stated relation as formal Laurent series in q-1.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The proof of Theorem 4 routes everything through the identification of the center of M(N) with symmetric polynomials in x_i via the RE-algebra Harish-Chandra isomorphism, Eqs. (4.7)-(4.8), imported from [GPS]. Proposition 24, Proposition 25/Corollary 27, and the limit step in Theorem 21 all use this identification. The weakest point is not the citation itself but how the limit q to 1 is taken after it: the paper sets x_i = q^{1-N-xi_i}/(q^2-1) (Section 4.2), a q-dependent, singular reparameterization under which the characteristic values S_k(xi) do not have a plain coefficient-wise limit; Remark 3 asserts that the limit can be taken coefficient-wise in x_i, but no proof is given that the Harish-Chandra image in these variables converges coefficient-wise to the classical Harish-Chandra image. If the normalization or domain of (4.7)/(4.8) is off by a q-power, or if the identification is only valid for N > m, the Schur-function form of Omega_m and its classical limit would fail. This is not merely hypothetical: Eq. (6.2) contains an undefined index l in the factor ([nu+m+l]_q)_{m-k}/([m+l]_q)_{m-k}, and Eq. (2.1) and Eq. (6.1) disagree on the order of the q-Bernoulli polynomial (beta^{(m-nu+1,nu)}_l vs beta^{(nu+m-l,nu)}_l). These are likely typos, but they make the printed derivation impossible to check as written and leave open the possibility of a systematic normalization error.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper claims to prove a conjecture of Kazarian, Krasilnikov, Lando, and Shapiro giving an explicit formula for the average value of the universal gl-weight system on permutations: the generating function identity of Theorem 4, equivalently Eq. (2.2), expressing W_m(N) as a linear combination of one-part Schur functions with Bernoulli-polynomial coefficients. The proof introduces a universal quantum gl-weight system on Hecke algebras of type A, constructed through the Reflection Equation algebra, and derives a q-analogue formula (2.1)/(6.1) for the average value on the q-symmetrizer in terms of q-Bernoulli polynomials and Schur functions. The classical conjecture is then recovered by taking the q to 1 limit, with a separate proof of the limit of the q-Bernoulli polynomials.","tokens_in":20077,"tokens_out":16848,"duration_ms":158380,"significance":"The paper targets a recent conjecture and, if the proof is sound, would provide both a proof and a one-parameter quantum deformation of the average universal gl-weight system. The construction has a natural algebraic framework, uses independent external results ([GPS], [JLM], [Ok], [C]), and includes explicit low-degree computations as cross-checks. The main limitations are presentation and rigor gaps in the central formulas and the q to 1 limit, which are fixable but currently prevent verification.","major_comments":[{"comment":"Eq. (6.2) uses an index l in the ratio ([nu+m+l]_q)_{m-k}/([m+l]_q)_{m-k} without defining it or summing over it, so the central formula for Omega_m(N) is not well-formed as printed. From Corollary 18 and Lemma 22 the factor should presumably be ([nu+m]_q)_{m-k}/([m]_q)_{m-k} (possibly with a q-power), and the correction is needed before the derivation can be checked.","section":"§6.1, Eq. (6.2)"},{"comment":"The two displayed formulas for Omega_m(N), both presented as the main result, disagree on the order of the q-Bernoulli polynomial: Eq. (2.1) has beta^{(m-nu+1,nu)}_l(nu/2) while Eq. (6.1) has beta^{(nu+m-l,nu)}_l(nu/2). This is not a harmless difference, because the order enters the coefficient and its q to 1 limit; the inconsistency must be resolved.","section":"§2, Eq. (2.1); §6, Eq. (6.1)"},{"comment":"The claim that the q to 1 limit can be taken coefficient-wise in the variables x_i is not justified. The change of variables x_i = q^{1-N-xi_i}/(q^2-1) is singular at q=1: if x_i are fixed, xi_i diverges, while if xi_i are fixed, x_i diverges. Corollary 27 substitutes u=q^{-nu} into Eq. (6.3) without explaining how the variables u-xi_i are related to the x_i appearing in Eq. (6.1). A precise statement of the ring in which Eq. (6.1) is an identity, and a proof that the Schur-function expansion is compatible with the classical Harish-Chandra image under this reparameterization, are needed for the limit step in Theorem 4.","section":"§4.2, Remark 3; §6.1, Corollary 27"},{"comment":"The passage from the triangular relations (5.4)-(5.5) to the conclusion that lim_{q to 1} Omega_m(N) = W_m(N) is compressed into a single sentence. The paper should state explicitly that the inverse of a unitriangular matrix whose entries are regular at q=1 is regular at q=1 (or give the ring of functions in which the inversion is performed), and it should identify the limit of the inverse with the inverse of the classical limit; the current wording leaves the regularity of the inverse entries as an unstated assumption.","section":"§5, proof of Theorem 21"}],"minor_comments":[{"comment":"The same symbols p_i and C_i denote generators of M(N), central elements of U(gl(N)), and universal variables; a short clarification of these identifications would improve readability.","section":"§1.2 and throughout"},{"comment":"In the proof of Lemma 6, the sentence 'The element gm does not affect the calculation' should refer to g_{m-1}; as written it is a typo.","section":"§4.2, Lemma 6"},{"comment":"In Example 1, item 5, the expression '1/q^2N' should be typeset as q^{-2N}.","section":"§4.3, Example 1"},{"comment":"The sentence 'Collecting these equalities for all n in a generating series' uses n where m is meant.","section":"§6"},{"comment":"Reference [Ok] has a typo in the title ('ans' should be 'and').","section":"References"},{"comment":"The phrase 'This mapping may not satisfy the definition of a quantum weight system and may even be non-linear' is confusing in a proof of well-definedness and should be rephrased.","section":"§4.2, Proposition 15"}],"recommendation":"major_revision","confidential_remarks":"The paper depends heavily on the normalization of the Harish-Chandra isomorphism for the Reflection Equation algebra imported from [GPS]; a referee with access to [GPS] should check that Eqs. (4.7)-(4.8) match the notation used here, since a q-power normalization error would invalidate the Schur-function form. The typos in (2.1), (6.1), and (6.2) should be brought to the author's attention before resubmission, as they currently make the main formulas unverifiable. The paper is otherwise within scope and likely publishable after revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this is a proof of the KKLS conjecture, not a restatement, and it adds a genuinely new q-deformed result. The main formula is not fitted; it emerges from a structured computation in the Reflection Equation algebra, and the classical limit is benchmarked against independent results. I think it should be refereed.\n\nWhat is new: Theorem 4 was open in [KKLS]. The proof goes through the universal quantum gl-weight system on Hecke algebras, uses the characteristic map to reduce the average of the q-symmetrizer, and lands on q-Bernoulli polynomials times one-part Schur functions. The universality arguments in Propositions 12–16 are a real piece of work: they establish the weight system by polynomial identities checked at infinitely many N. Section 6.2 on q-Bernoulli limits is the cleanest part; the recursion there is explicit and the limit is Carlitz's theorem applied through legitimate identities. There are no free parameters anywhere.\n\nSoft spots, in size order. First, the manuscript has concrete typos in central formulas: Eq. (2.1) and Eq. (6.1) disagree on the order parameter of the q-Bernoulli polynomial, and Eq. (6.2) contains an undefined l. These are likely copy-paste errors, but they make the printed derivation impossible to verify as written. Second, the limit q→1 after the Harish-Chandra substitution is asserted more quickly than I would like. Remark 3 says the limit is coefficient-wise in the xi variables, but the relation xi = q^{1-N−ξ_i}/(q^2−1) is singular at q=1, and the text does not justify that the image converges coefficient-wise. This is the one mathematically substantive gap. I would not call it fatal: the computation is checked against independent results of [JLM] and [Ok], and the conjecture itself is not assumed, so there is no circularity. But a referee should ask for a direct justification of the limit or a cleaner change of variables. The paper's own admission that it cannot yet prove specialization on arbitrary Hecke elements is honest and does not affect the averaged statement.\n\nThis is for people in weight systems, Vassiliev invariants, and Hecke-algebra centers. I would send it to a serious referee; after the typos and the limit justification are addressed, I would cite it.","headline":"Genuine proof of the KKLS conjecture with a new q-deformation, worth refereeing despite typos and a terse q→1 limit step.","tokens_in":20799,"tokens_out":4382,"would_cite":true,"duration_ms":47094,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["05E05","05E10","20C08","17B37"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves that the average value of the universal gl-weight system on permutations is a finite sum of one-part Schur functions with Bernoulli-polynomial coefficients, obtained as the classical limit of a new quantum deformation.","keywords":["weight systems","universal gl-weight system","Hecke algebra","Reflection Equation algebra","one-part Schur functions","q-Bernoulli polynomials","quantum weight system","Harish-Chandra isomorphism"],"falsifier":"Directly evaluate both sides of Eq. (2.2) for a small case such as $m=3$, $N=2$: compute the left side from the definition of the $\\mathfrak{gl}(2)$-weight system on the six permutations of $S_3$, and compute the right side as a polynomial in $x_1,x_2$ using the Schur functions $S_0,S_1,S_2,S_3$ and Bernoulli polynomials $B_l^{(1)}(1/2)$. Any mismatch between these two explicit polynomials would refute the theorem.","tokens_in":19515,"feed_emoji":"🧮","tokens_out":15677,"duration_ms":139551,"temperature":0.7,"pith_summary":"The paper proves a conjectured closed formula for the average value of the universal $\\mathfrak{gl}$-weight system, a family of polynomial invariants attached to permutations that encodes finite-type knot invariants and unifies all $\\mathfrak{gl}(N)$-weight systems. The formula says that the average $W_m$ is a finite linear combination of one-part Schur functions $S_k$ (Schur polynomials indexed by partitions with a single part) with explicit coefficients built from Bernoulli polynomials of order $\\nu=N-1$: $$W_m(N)=\\sum_{l=0}^{m}\\frac{(\\nu+m)_l}{l!}\\,$B_l^{{(\\nu)}}$\\left(\\frac{\\nu}{2}\\right)S_{m-l},$$ with the equivalent generating-function identity of Theorem 4. To get there, the paper builds a quantum analogue of the weight system on Hecke algebras of type $A$ and shows that its average value $\\Omega_m(N)$ on the $q$-symmetrizer is the same kind of combination with $q$-Bernoulli coefficients; the classical formula is recovered by taking $q\\to 1$ coefficient-wise. The proof makes the averaged universal $\\mathfrak{gl}$-weight system explicitly computable for every $m$ and $N$ without summing over permutations, and it yields a one-parameter quantum deformation of the classical average.","feed_headline":"Average gl-weight values are explicit Schur-function sums","feed_subtitle":"A quantum q-deformation with Bernoulli coefficients settles the open conjecture and makes every N and m computable at once","key_machinery":"The central mechanism is the quantum $\\mathfrak{gl}$-weight system $\\omega_m$ on the type-$A$ Hecke algebra $H_m$, built from the Reflection Equation algebra $M(N)$ and its $R$-matrix representation. The center of $M(N)$ is identified with symmetric polynomials in auxiliary variables $\\xi_i$ (the quantum analogue of the Harish-Chandra isomorphism), and the quantum power sums $p_m$ and quantum Casimirs $C_m$ are expressed through those variables by the quantum Perelomov-Popov formulas. The averaged value $\\Omega_m(N)=\\omega_{m,N}(h_m)$ is computed with the characteristic mapping $\\chi_m$ and the $q$-symmetrizer $h_m$, using the trace formula $\\langle H^{(m)}\\rangle_{m,m}=q^{-1}\\frac{[\\nu+m]_q}{[m]_q}H^{(m-1)}$; the result is a finite sum of one-part Schur functions whose coefficients are $q$-Bernoulli polynomials. The classical Bernoulli polynomials $B_l^{(\\nu)}$ emerge as the $q\\to 1$ limits of these coefficients.","core_discovery":"The paper's central claim is Theorem 4: the averaged universal $\\mathfrak{gl}$-weight system satisfies the generating function identity $$\\sum_{m=0}^\\infty\\frac{W_m}{(m+\\nu)!}t^m = \\left(\\frac{$e^{{t/2}}$-$e^{{-t/2}}$}{t}\\right)^{-\\nu}\\sum_{k=0}^\\infty \\frac{S_k}{(k+\\nu)!}t^k,$$ where $\\nu=N-1$ and $S_k$ are the one-part Schur polynomials. Equivalently, $W_m(N)=\\sum_{l=0}^m \\frac{(\\nu+m)_l}{l!}B_l^{(\\nu)}(\\nu/2)S_{m-l}$, which is precisely the expression previously conjectured. The proof introduces the quantum $\\mathfrak{gl}$-weight system $\\omega_m$ on the Hecke algebra $H_m$ and demonstrates that its value on the $q$-symmetrizer $h_m$ is $$\\Omega_m(N)=\\frac{m!}{(\\nu+m)!}\\frac{[\\nu+m]_q!}{[m]_q!}\\sum_{l=0}^m \\frac{(\\nu+m)_l}{$q^{{2l}}$l!}\\$beta_l^{{(\\nu+m-l,\\nu)}}$(\\nu/2)S_{m-l},$$ where $\\beta_l^{(h,k)}$ are order-$k$ $q$-Bernoulli polynomials. The paper then proves that the termwise $q\\to 1$ limit of this quantum expression equals $W_m(N)$, establishing the conjectured classical formula as a specialization of the quantum invariant.","pith_inferences":["Beyond the paper, the same characteristic-mapping and $q$-Casimir machine should give explicit averaged formulas for other central idempotents of the Hecke algebra, not just the $q$-symmetrizer.","The appearance of $q$-Bernoulli polynomials points to a measure-theoretic reading of the average weight system as an expectation against a $q$-deformed distribution, in the spirit of $q$-Volkenborn integration.","The classical formula's Bernoulli-polynomial coefficients invite a connection to Todd classes and Hirzebruch genera; the averaged $\\mathfrak{gl}$-weight system could be a combinatorial shadow of a universal multiplicative genus, a direction the paper does not explore.","A direct numerical check of Eq. (2.1) for small $N$ and $m$ using the $R$-matrix definition of $\\omega_{m,N}$ would test the quantum formula independently of the derivations."],"forward_implications":["The averaged $\\mathfrak{gl}(N)$-weight system is explicitly computable by substituting $N$ and the variables $x_1,\\dots,x_N$ into Eq. (2.2); no summation over all $m!$ permutations is required.","The generating-function identity of Theorem 4 packages every average $W_m$ into a single expression, so information about all $m$ can be extracted from one expansion.","The quantum formula (2.1) provides a one-parameter deformation $\\Omega_m(N)$ whose $q\\to 1$ limit is the classical value, giving a bridge between Hecke-algebra computations and classical weight systems.","Since the coefficients are Bernoulli polynomials of order $\\nu$, the average weight system inherits the generating series, recurrences, and arithmetic properties of classical Bernoulli polynomials."],"supporting_citations":[{"why":"States the conjecture on the average value of the universal gl-weight system that the paper proves, and supplies the Bernoulli-polynomial form of the expected answer.","marker":"[KKLS]"},{"why":"Provides the Reflection Equation algebra, the quantum Harish-Chandra isomorphism, and the formula for the average quantum character used in Proposition 24.","marker":"[GPS]"},{"why":"Defines the quantum gl-weight system and the quantum characteristic mapping on Hecke algebras, which are the paper's main deformation tool.","marker":"[GS]"},{"why":"Supplies the explicit Harish-Chandra image of the auxiliary element bOmega_m(N) in Eq. (5.1), a key input to the q-to-1 limit argument.","marker":"[JLM]"},{"why":"Supplies the corresponding classical Harish-Chandra image of cW_m(N) in Eq. (5.2), which identifies the q-to-1 limit target.","marker":"[Ok]"},{"why":"Establishes that the q-Bernoulli numbers of order 1 converge to classical Bernoulli numbers, the analytic input for Proposition 28.","marker":"[C]"},{"why":"Introduces the universal gl-weight system on permutations and its recurrence, on which the classical statement of the theorem rests.","marker":"[KL]"},{"why":"Together with [KL], establishes the universal gl-weight system formalism and the recurrence that makes averaging over permutations meaningful.","marker":"[ZY]"}],"fun_headline_variants":["Quantum weight system yields explicit Schur-Bernoulli formula","Average gl-weight values now explicit via q-Bernoulli polynomials","Conjecture on gl-weight averages proven with quantum deformations","Weight system average becomes Schur sum with Bernoulli coefficients","Quantum gl-weight system: average values finally computed"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole argument assumes the algebraic quantities can be faithfully re-expressed as ordinary polynomials in some auxiliary variables, and that the classical limit can be taken piece by piece in those variables; if that re-expression or that limit is not valid, the formula collapses.","fun_headline_variants_meta":{"raw":{"variants":["Quantum weight system yields explicit Schur-Bernoulli formula","Average gl-weight values now explicit via q-Bernoulli polynomials","Conjecture on gl-weight averages proven with quantum deformations","Weight system average becomes Schur sum with Bernoulli coefficients","Quantum gl-weight system: average values finally computed"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000949,"raw_usage":{"total_tokens":4072,"prompt_tokens":988,"completion_tokens":3084,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":604,"completion_tokens_details":{"reasoning_tokens":3004}},"tokens_in":604,"tokens_out":3084,"duration_ms":24081,"temperature":1.0,"reasoning_tokens":3004,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T19:03:47.105951+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Directly evaluate both sides of Eq. (2.2) for a small case such as $m=3$, $N=2$: compute the left side from the definition of the $\\mathfrak{gl}(2)$-weight system on the six permutations of $S_3$, and compute the right side as a polynomial in $x_1,x_2$ using the Schur functions $S_0,S_1,S_2,S_3$ and Bernoulli polynomials $B_l^{(1)}(1/2)$. Any mismatch between these two explicit polynomials would refute the theorem.","supporting_citations":[],"review_version":1}