{"id":"63ae2810-751e-4e6d-afa4-494d254b58eb","arxiv_id":"2508.15113","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Closed-form bivariate generating functions are proved for two-rowed tight cylindric partitions, along with a bijection to Dousse-Hardiman-Konan partitions.","lead":"This paper studies tight cylindric partitions, a restricted subclass tied to affine Lie algebra characters, and derives two-variable generating functions for them. The main result gives explicit formulas for the two-row case and a bijection with another family of colored partitions.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 14's proof rests on unstated q-difference relations and an implicit uniqueness step; both are fillable but currently make the central closed form conditional.","rationale":"The reader's weakest_assumption identifies the unproved q-difference relations rel_j and rel_0, and that is precisely the most load-bearing point: Theorem 14's proof is a sequence of linear combinations of these relations, so an error in their signs or exponents would invalidate the central closed form for two-rowed tight cylindric partitions. My independent check of the definition of S indicates the relations are plausible and the algebra in the later linear combinations is internally consistent, but the manuscript gives no derivation, so a referee cannot verify the central claim without reconstructing this algebra. A second necessary ingredient, uniqueness of solutions to the diamond relations, is also not made explicit, although it follows by induction on q-degree. The ℓ=1 edge case in the statement of Theorem 14 is a minor indexing defect. None of this suggests the theorem is false; it does mean acceptance should be conditional on supplying the missing derivations and uniqueness argument. The reader's verdict of CONDITIONAL with moderate confidence remains appropriate, so no verdict change is needed.","tokens_in":15469,"tokens_out":31586,"duration_ms":342010,"concrete_test":"Derive rel_j directly from the defining sum: in S(t;v)-S(t;v+e_j), use (1-q^{n_j})/(q;q)_{n_j}=1/(q;q)_{n_j-1}, shift n_j -> n_j-1, and track the factor (q;q)_{N_1+1}/(q;q)_{N_1}=1-q^{N_1+1}; confirm that the surviving terms are exactly -z q^{v_j+j} S(t+1;v+Δ_j) + z q^{v_j+j+1} S(t+1;v+δ_1+Δ_j). Derive rel_0 similarly. Then verify Proposition 15 and equations (11)-(14) symbolically for ℓ=2,3,4 by truncating S(0;-η_i) at q^N and comparing with a direct enumeration of tight cylindric partitions from Definition 3 up to weight 10. Also write the one-paragraph q-degree induction showing the diamond relations have a unique formal solution with the stated initial conditions.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central identity, Theorem 13/14, is proved by showing that S(0;-η_i) satisfies the diamond relations. That verification is entirely carried out with the relations rel_j and rel_0 in Section 4.1, which are stated without derivation: rel_j is justified only by 'obtained routinely' and rel_0 by 'obtained easily.' Every cancellation in the proof of Proposition 15 and in equations (11)-(14) depends on the exact signs and exponents of these relations, e.g. the term z q^{v_j+j+1} S(t+1; v+δ_1+Δ_j). A sign error there would propagate through the linear combinations and break the claimed identity. Separately, even if S is shown to satisfy the diamond relations, the paper never states or proves that these recurrences plus the initial conditions R_i(0,q)=R_i(z,0)=1 have a unique formal solution; this is needed to conclude T equals the displayed multisum. The missing uniqueness is a routine induction on the q-degree, but it is not present. There is also a small edge-case omission: for ℓ=1 the displayed system in Theorem 14 refers to R_1 or R_{-1}, which are not defined. These are fillable gaps rather than evidence of falsity, but they are load-bearing in the current write-up.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper initiates the study of bivariate generating functions for tight cylindric partitions, refining the usual cylindric-partition framework with the maximum-part statistic. For general r-rowed tight cylindric partitions of profile c, the authors prove analogs of the Corteel–Welsh functional equations (Theorem 10). In the two-rowed case they specialize these to the 'diamond relations' (Proposition 11) and to a single recurrence without inclusion–exclusion (Proposition 12). The main result, Theorem 13/14, gives a closed multisum formula for the bivariate generating function T_{(ℓ-b,b)}(z,q) for 0 ≤ b ≤ ⌊ℓ/2⌋, and the authors show that this multisum satisfies the diamond relations. They also give a bijection between two-rowed tight cylindric partitions of profile (ℓ-a,a) and Dousse–Hardiman–Konan partitions of class DHK_{a,ℓ}, sending the maximum part to the number of nonzero parts, and use this to give an alternate proof of the DHK product identity. A level-one result for r rows is included as Proposition 17.","tokens_in":15792,"tokens_out":7283,"duration_ms":76130,"significance":"If the main identity is correct, this is the first closed form for two-rowed tight cylindric partitions with the maximum-part statistic, and it connects these objects to known Andrews and Kim–Yee sums and to Dousse–Hardiman–Konan partitions. The paper contains several genuinely useful ingredients: explicit functional equations, a product formula, a concrete bijection, and a new level-one formula. The approach is largely self-contained and the combinatorial recurrences are clearly motivated. However, the proof of the central theorem currently rests on q-difference relations that are only sketched and on an unstated uniqueness assertion for the recurrence system; these gaps make the main result conditional in the present write-up.","major_comments":[{"comment":"The proof of Theorem 14 is entirely carried out through the relations rel_j(t;v) and rel_0(t;v). rel_j is justified only by 'obtained routinely' and rel_0 by 'obtained easily'. Every cancellation in Proposition 15 and in equations (11)–(14) depends on the exact signs and shifts in these relations, for instance the term z q^{v_j+j} S(t+1; v+Δ_j) in rel_j. A sign or exponent error would propagate through the linear combinations and invalidate the verification that S(0; -η_i) satisfies the diamond relations. Please provide a complete derivation of both relations, or at minimum spell out the index shift for rel_j, including the treatment of the n_j = 0 boundary terms.","section":"§4.1, rel_j and rel_0"},{"comment":"The proof shows that the multisum S(0; -η_i) satisfies the diamond relations and the initial conditions R_i(0,q)=R_i(z,0)=1. It is never stated or proved that these recurrences together with the initial conditions have a unique formal power series solution. Since the actual tight-cylindric generating functions T_{(ℓ-b,b)} also satisfy the same recurrences by Proposition 11, the equality T_{(ℓ-b,b)} = S(0; -η_b) requires a uniqueness lemma. Such a lemma is standard—one may induct on q-degree, using that R_i(zq^k) shifts the q-degree when k>0—but it is not present. The same missing uniqueness is asserted in §6.2, where Propositions 12 and 21 are said to define the same unique solution.","section":"§4.2, proof of Theorem 14"},{"comment":"Theorem 13 states ℓ ≥ 1, but the system displayed in Theorem 14 is not defined for ℓ = 1: the odd case refers to R_{(ℓ-3)/2} = R_{-1}, and the initial relation refers to R_1, while the allowed range is 0 ≤ i ≤ ⌊ℓ/2⌋ = 0. Please either restrict Theorem 14 (and Proposition 11, if needed) to ℓ ≥ 2 and treat ℓ = 1 separately, or define the missing boundary values appropriately. This is a small but genuine gap in a theorem statement that claims to cover all ℓ ≥ 1.","section":"Theorem 14, ℓ = 1 edge case"}],"minor_comments":[{"comment":"The two displayed multisum expressions are asserted to be equal. The equality follows from n_j = N_j - N_{j+1}, but the index manipulation is not shown; a sentence indicating this would improve readability.","section":"Theorem 13"},{"comment":"The abacus diagram in Example 5 is difficult to parse as printed. Adding explicit labels for the zeroth yoke and the first few yokes, or drawing the yokes as arcs, would make the example much easier to follow.","section":"Example 5"},{"comment":"The claim that the number of vacancies between two adjacent yokes of shapes α and β is |α−β| is cited to [9, Eq. (31)]. Since this equality is the heart of the bijection with DHK partitions, a short derivation or a more precise quotation of the relevant statement would be helpful.","section":"§6.3, bijection"},{"comment":"The q argument is suppressed in R_i(z) throughout Theorem 14. It is clear from context, but explicitly writing R_i(z,q) in the display would avoid ambiguity, especially because the recurrences involve R_i(zq) and R_i(zq^2).","section":"Notation in Theorem 14"}],"recommendation":"major_revision","confidential_remarks":"The concerns raised in the major comments are fillable within the scope of the manuscript: the q-difference relations can be verified by explicit computation, the uniqueness step is a standard induction, and the ℓ = 1 case can be separated. I found no evidence that the main identity is false. I would recommend requesting a revision that supplies these missing proofs before publication. The paper is a good fit for the journal and the main result, once the proof is made fully rigorous, would be a solid contribution."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First, the useful facts. Kanade and Russell do what the title says: they initiate the bivariate theory of tight cylindric partitions, prove the tight analog of the Corteel–Welsh recurrences (Thm 10), then solve the two-row case with explicit multisums (Thm 13/14). The sums are new, and the authors are honest that setting z=1 recovers known identities of Andrews and of Kim–Yee. The DHK bijection in Section 6.3 is a nice addition and gives a combinatorial proof of the Dousse–Hardiman–Konan identity. Proposition 17, the level-1 r-rowed formula, is a clean bonus.\n\nThe proofs are mostly from first principles and I saw no sign of parameter-fitting or circularity. The derivation of Theorem 10 is involved but coherent, with the inclusion–exclusion handled carefully.\n\nThe soft spots are real but localized. The proof of Theorem 14 is a verification that S(0;−η_i) satisfies the diamond relations, and every cancellation in Proposition 15 and (11)–(14) uses rel_j and rel_0 from Section 4.1. The text says these are obtained 'routinely' and 'easily', and indicates the mechanism (multiply by (1−q^{n_j}), shift the sum), but it does not display the derivations. Since a single sign or exponent error would propagate through the linear combinations, this is load-bearing. A referee should ask for those computations in full. Second, the paper never states that the diamond recurrences plus the initial conditions R_i(0,q)=R_i(z,0)=1 have a unique formal power series solution. The uniqueness is a routine induction on the degree in z, but it should be said. Third, the ℓ=1 case of the system in Theorem 14 refers to R_1, which is not defined; that edge case needs a sentence.\n\nNone of these look fatal to me; they are fillable gaps in an otherwise credible note. The main claims are likely correct. I would send this to a serious referee, and I would be surprised if the referee found a counterexample. The paper is for q-series and partition theorists, and for anyone who cares about crystal-theoretic identities. I'd bring it to a reading group focused on q-series; the proof gap makes a good discussion.","headline":"A credible, genuinely new start on bivariate tight cylindric partitions; the two-row closed forms are likely right but the proof rests on q-difference relations that need to be written out.","tokens_in":16242,"tokens_out":3267,"would_cite":true,"duration_ms":38000,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["05A17","11P84","17B67"],"pacs":[],"model":"deepseek-v4-flash","headline":"Two-rowed tight cylindric partitions have a closed-form bivariate generating function.","keywords":["tight cylindric partitions","bivariate generating functions","largest part statistic","q-hypergeometric multisums","functional equations","colored partitions","crystal bases","partition identities"],"falsifier":"Compute the coefficient of z^M q^N on both sides of Theorem 13 for, say, ℓ=3, b=1 and N ≤ 12 by enumerating the finitely many two-rowed tight cylindric partitions of profile (2,1); if any coefficient disagrees with the multisum, the closed form fails, and the same check can be run independently by substituting the proposed multisums into the diamond relations to high order in q.","tokens_in":15371,"feed_emoji":"🧮","tokens_out":7535,"duration_ms":85057,"temperature":0.7,"pith_summary":"This paper initiates the study of bivariate generating functions for tight cylindric partitions—cylindric partitions whose abacus yokes are packed as closely as possible, the objects that match affine Lie algebra characters directly. The main result, Theorem 13, is a closed form: for every level ℓ and every 0 ≤ b ≤ ⌊ℓ/2⌋, the generating function that tracks both the total weight and the largest part of two-rowed tight cylindric partitions of profile (ℓ−b,b) is a finite ℓ-fold basic hypergeometric multisum. The proof passes through new functional equations, called the diamond relations, that generalize the classical functional equations for cylindric partitions to the tight setting, and solves them with a family of multisums S(t;v;z,q). The same functional equations and a direct bijection connect these partitions to a class of colored partitions whose generating function was previously known from crystal theory, giving an independent route to that identity. The closed form matters because it is the first exact bivariate description of tight cylindric partitions with the largest-part statistic, a statistic tied to the representation theory behind the objects.","feed_headline":"A multisum counts two-row tight cylindric partitions","feed_subtitle":"New bivariate generating functions track largest part and weight; a bijection explains a known partition identity.","key_machinery":"The workhorse is the family of multisums S(t; v; z,q) together with the two q-difference relations, rel_j(t;v) and rel_0(t;v), that they satisfy. The first relation follows by multiplying each summand by (1−q^{n_j}) and shifting the summation index; the second by subtracting adjacent t-slices. The proof of Theorem 14 is a carefully chosen linear combination of these relations, with telescoping cancellations, that collapses exactly to the diamond relations; since the multisums share the same initial conditions as the tight cylindric partition generating functions, uniqueness identifies them.","core_discovery":"On the paper's own terms, the central discovery is that two-rowed tight cylindric partitions are governed exactly by a finite multisum with no free parameters. For level ℓ and 0 ≤ b ≤ ⌊ℓ/2⌋, let T(ℓ−b,b)(z,q) be the generating function for tight cylindric partitions of profile (ℓ−b,b), with z marking the largest part and q the sum of entries. Theorem 13 states T(ℓ−b,b)(z,q)=T(b,ℓ−b)(z,q) equals Σ_{n1,...,nℓ≥0} z^{N1} q^{(N1^2+...+Nℓ^2)/2 + (n1+n3+...+n_{2b-1})/2 + (N_{2b+1}+...+Nℓ)/2} (q)_{N1} / ((zq;q)_{N1}(q)_{n1}...(q)_{nℓ}), where N_j=n_j+...+n_ℓ. The proof route is Theorem 14: each R_i(z,q)=S(0;−η_i;z,q) solves the diamond relations—the two-row case of the paper's new functional equatio","pith_inferences":["If Theorem 13 is correct, the largest-part statistic likely has a representation-theoretic reading as a degree or energy grading on the crystal of the level-ℓ affine sl_2 module, and the multisum should be matchable to known crystal-energy generating functions.","The bijection to colored partitions suggests transferring statistics: the color sequence of the colored partition may translate to a statistic on tight cylindric partitions, such as yoke-shape counts, yielding a refinement of Theorem 13 with more than two variables.","The linear-combination-of-q-difference-relations strategy used for two rows is a plausible template for three-row tight cylindric partitions, a case the paper leaves open; the same rel_j machinery may produce conjectural multisums.","The paper reports computer experiments suggesting strict unimodality of the z-coefficients for fixed q-degree; proving this could follow from the bijection with colored partitions or from an sl_2 action, and low-level cases are now testable directly from Theorem 13."],"forward_implications":["At z=1, the formula reproduces the two known families of finite sums that had appeared in earlier partition identities for odd and even ℓ, giving product forms for the univariate generating function.","Together with the new functional equations, the closed form yields an alternate proof of the sum-to-product identity for the colored partitions of the same profile, since both families satisfy the same recurrences with the same initial data; the direct bijection makes the match combinatorial.","For arbitrary r rows at level 1, the paper obtains an explicit single-sum formula for the bivariate generating function in terms of the largest part.","For r ≥ 2, the new functional equations provide an analog of the classical recurrences, so future bivariate studies of tight cylindric partitions can start from a recurrence rather than from scratch."],"supporting_citations":[{"why":"Supplies the abacus model, the characterization of tight cylindric partitions by tight abaci, the product evaluation for univariate tight generating functions, and the vacancy-shape relation used in the bijection.","marker":"[9]"},{"why":"Provides the original functional-equation method for cylindric partitions that the paper adapts to the tight case; the diamond relations are the two-rowed instance of this adaptation.","marker":"[5]"},{"why":"Defines the colored partitions and proves their sum-to-product identity by crystal theory; the paper proves an alternate derivation via functional equations and gives a bijection to these partitions.","marker":"[7]"},{"why":"One of the two known z=1 sums that Theorem 13 reproduces for odd ℓ, used as a check and as a bridge to earlier partition identities.","marker":"[13]"},{"why":"The other known z=1 sum, for even ℓ, used in the same way.","marker":"[1]"}],"fun_headline_variants":["Two-row tight cylindric partitions yield closed form","Finite multisum solves 2-row tight cylindric partitions","Bijection links tight cylindric to known partition class","Closed form for 2-row tight cylindric partition counts","Tight cylindric partitions: bivariate generating function"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The proof leans on two families of q-difference identities, rel_j(t;v) and rel_0(t;v), that are declared to be obtained routinely or easily without derivation; if any sign or exponent in them is off, the multisum verification of the diamond relations fails.","fun_headline_variants_meta":{"raw":{"variants":["Two-row tight cylindric partitions yield closed form","Finite multisum solves 2-row tight cylindric partitions","Bijection links tight cylindric to known partition class","Closed form for 2-row tight cylindric partition counts","Tight cylindric partitions: bivariate generating function"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000112,"raw_usage":{"total_tokens":875,"prompt_tokens":702,"completion_tokens":173,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":446,"completion_tokens_details":{"reasoning_tokens":109}},"tokens_in":446,"tokens_out":173,"duration_ms":2688,"temperature":1.0,"reasoning_tokens":109,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T18:06:11.996205+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the coefficient of z^M q^N on both sides of Theorem 13 for, say, ℓ=3, b=1 and N ≤ 12 by enumerating the finitely many two-rowed tight cylindric partitions of profile (2,1); if any coefficient disagrees with the multisum, the closed form fails, and the same check can be run independently by substituting the proposed multisums into the diamond relations to high order in q.","supporting_citations":[{"cited_title":"Foda and T","cited_arxiv_id":null,"evidence_quote":"Supplies the abacus model, the characterization of tight cylindric partitions by tight abaci, the product evaluation for univariate tight generating functions, and the vacancy-shape relation used in the bijection."},{"cited_title":"Corteel and T","cited_arxiv_id":null,"evidence_quote":"Provides the original functional-equation method for cylindric partitions that the paper adapts to the tight case; the diamond relations are the two-rowed instance of this adaptation."},{"cited_title":"Dousse, L","cited_arxiv_id":null,"evidence_quote":"Defines the colored partitions and proves their sum-to-product identity by crystal theory; the paper proves an alternate derivation via functional equations and gives a bijection to these partitions."},{"cited_title":"Kim and A","cited_arxiv_id":null,"evidence_quote":"One of the two known z=1 sums that Theorem 13 reproduces for odd ℓ, used as a check and as a bridge to earlier partition identities."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The other known z=1 sum, for even ℓ, used in the same way."}],"review_version":1}