{"id":"31d394e2-e14d-4226-971d-c9b67da8b086","arxiv_id":"1908.02925","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"Open Richardson varieties in Grassmannians have trivial divisor class group over any field and over the integers.","lead":"This paper proves that every open Richardson variety in a Grassmannian has a trivial divisor class group: every divisor is principal. The proof shows the coordinate ring is a unique factorization domain, using Plücker coordinates and positroid varieties.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The Z-extension in Lemma 5 rests on a one-sentence Hilbert-function argument that is not fully justified; the field-case proof is otherwise coherent.","rationale":"I read the paper in good faith and traced the main argument. The proof over an arbitrary field is coherent: the parameterization in Theorem 7 is explicit, the localization at the Plücker coordinates is handled correctly, the positroid identification in Theorem 12 is checkable, and the cited KLS integrality result supplies the needed primality. I do not find an internal inconsistency in the field case. The only load-bearing concern is the extension to the integers, exactly as the reader identified. Lemma 5 is a one-sentence proof of integrality over Z, and the authors themselves note that the analogous Hilbert-function statement for positroid varieties over Z is not proved. Since the abstract explicitly claims the result over the integers, this is the place where the paper is most exposed. The concern is not that the theorem is false, but that the Z-case is underproved relative to the paper's own standards; a standard monomial basis over Z would close the gap. This does not move the reader's conditional verdict, but it confirms that the conditional is appropriate.","tokens_in":8941,"tokens_out":45875,"duration_ms":477309,"concrete_test":"Prove Lemma 5 via standard monomial theory: exhibit, for every degree d, a free Z-basis of A^γ_β[d] consisting of standard monomials in the Plücker coordinates. If such a basis exists, the Hilbert function is visibly field-independent, each graded piece is Z-free, and the flatness implication in Lemma 5 goes through. If the basis can only be constructed over Q, or if a Hilbert-function comparison over Q and F_p for a small example such as Gr(3,6) with β={1,2,4}, γ={3,5,6} reveals a discrepancy, then the Z-case is a genuine gap.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The theorem is claimed over any field and over the integers. The field case is well supported: W^γ_β is a product of affine space and a torus (Theorem 7), so B^γ_β is a UFD; Lemma 9 and Theorem 12 show that the relevant set of Plücker coordinates is a positroid; KLS integrality (Theorem 13) gives Corollary 15; Nagata's Criterion then yields Proposition 2. The load-bearing soft spot is the integer case, handled solely by Lemma 5 in Section 2.1. Its entire proof is: if X^γ_β over Z were not integral, A^γ_β would have p-torsion; this is impossible because the Hilbert function is the same over any field. Two subclaims are needed without proof: (i) that the coordinate ring is flat over Z so that p-torsion is the only obstruction to being a domain given a domain generic fiber, and (ii) that equality of Hilbert functions over Q and over every F_p forces each graded piece to be Z-free. The paper itself flags the analogous Hilbert-function statement for positroid varieties over Z as unproved, but Lemma 5 is the corresponding unstated step for Richardson varieties. If subclaim (ii) fails for some β, γ, and p, then Proposition 2 and Theorem 1 are not established over Z, although the field case would remain intact.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves Theorem 1, which states that the divisor class group of any open Richardson variety X^γ_β in the Grassmannian is trivial, equivalently that its coordinate ring is a unique factorization domain. The proof is via Nagata's criterion: the authors define an open subscheme W^γ_β, show it is isomorphic to a product of affine space and a torus, and conclude that the localized ring B^γ_β is a UFD (Theorem 7 and Corollary 8). They then show that each of the Plücker coordinates Δ_{δ_t} is either a unit or a prime element of A^γ_β (Corollaries 10 and 15), using Lemma 9 to identify the relevant set of Plücker coordinates as a positroid and Theorem 13 (Knutson–Lam–Speyer) to obtain integrality of the corresponding positroid variety. Nagata's criterion then gives Proposition 2 and hence Theorem 1. The theorem is claimed over an arbitrary field and over the integers.","tokens_in":9160,"tokens_out":25817,"duration_ms":301695,"significance":"If correct, this is a clean and useful result: it settles the divisor class group, Picard group, and UFD property for all open Richardson varieties in the Grassmannian, over any field and over the integers. The method is elegant and largely self-contained, reducing a geometric statement about divisors to the combinatorial theory of positroids and to standard theorems of Postnikov and Knutson–Lam–Speyer. The explicit parameterization in Theorem 7 is a nice contribution in its own right. The field-case argument appears sound and the external inputs are clearly identified. The main weakness is the integer-case extension, where the proof is considerably more terse and, in my assessment, incomplete as written.","major_comments":[{"comment":"The proof that X^γ_β is integral over Spec Z is not adequate as written. The sentence 'This cannot happen since e.g. the Hilbert function of A^γ_β is the same over any field' requires two unproved assertions: (i) the Hilbert function of the Richardson variety is independent of the field, and (ii) equality of the Q-valued and F_p-valued Hilbert functions forces each graded piece of A^γ_β to be torsion-free over Z. Assertion (ii) is not automatic and needs an argument controlling the reduction of a finite free resolution, and assertion (i) needs a proof or a precise citation such as standard monomial theory. The further implication 'if not integral over Z, then A has p-torsion' relies on knowing that the only obstruction to A being a domain, given that Q⊗A is a domain, is torsion; this is true only after one knows the kernel of A→Q⊗A is the torsion submodule, which is exactly what the Hilbert-function statement is supposed to establish. Since Theorem 1 and Proposition 2 are explicitly claimed over the integers, this gap is load-bearing. The manuscript itself notes after Lemma 6 that the analogous integrality statement for positroid varieties over Z is not proved; the same standard is needed here.","section":"§2.1, Lemma 5"}],"minor_comments":[{"comment":"The sentence 'Proj A^γ_β/⟨Δ_{δ_t}⟩ is a closed integral subscheme of Proj A^γ_β, which is equivalent' should be justified: integrality of the Proj of a Z-graded ring does not in general imply primality of the ideal unless one knows the ring is a Laurent polynomial extension of its degree-zero part. Here this is true because Δ_β is a unit of degree one, so A^γ_β ≅ A_0[Δ_β, Δ_β^{-1}], but the step should be stated explicitly.","section":"§6, Corollary 15"},{"comment":"There is a typo: 'an unique factorization domain' should read 'a unique factorization domain'.","section":"§3, Corollary 8"},{"comment":"The notation for ℙ_t and ℙ^t is visually indistinguishable in the typeset version of the manuscript; please use clearly distinct symbols and define them once at the start of Section 4.","section":"§4–§6"}],"recommendation":"major_revision","confidential_remarks":"The field-case proof is solid and the paper is worth publishing once the integer-case argument in Lemma 5 is repaired. If the authors cannot supply the missing Hilbert-function/torsion argument, they could instead restrict Theorem 1 to fields and state the integer case conditionally; as written, however, the stated theorem over the integers is not fully proved."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The headline: this is a genuinely new result, and the proof is mostly clean. The field-case argument is solid; the integer-case extension is under-justified and needs work before the paper is fully rigorous.\n\nWhat's new: the statement that every open Richardson variety in the Grassmannian has trivial divisor class group (equivalently, its coordinate ring is a UFD). Prior work covered Schubert cells and Schubert varieties, but not Richardson varieties, and the paper's route through positroid varieties is not just a repackaging. The parametrization of the localized subscheme W^γ_β as affine space times a torus (Theorem 7) is elegant, and the application of Nagata's criterion is well organized. Lemma 9 and Theorem 12, which identify the relevant Plücker coordinates as a positroid, are clear and check out. For anyone working in Schubert calculus, this is a useful tool, not a marginal curiosity.\n\nThe soft spot is the integer case. Lemma 5 claims integrality of X^γ_β over Spec Z with a one-sentence proof: 'the Hilbert function is the same over any field.' The stress-test note is right that this needs more. To rule out p-torsion, you need the graded pieces over Z to be torsion-free, and equality of dimensions over Q and over every F_p gives you that only if you already know the Hilbert function over F_p is computed from F_p ⊗ A_Z (which it is) and that rank_Z(A_d) = dim_Q(A_d ⊗ Q) is actually the dimension over Q. That is all fine, but the equality of Hilbert functions across all fields is a real fact about Richardson varieties that is not proved or cited here. The paper even flags the analogous integrality statement for positroid varieties over Z as unproved. As written, the theorem over Z is not fully established. This is repairable: one can either give a proper proof or cite a reference for flatness of Richardson varieties over Z, and the abstract should then match the proof. For the main application in the companion paper, the field case suffices, so this gap doesn't sink the paper.\n\nThe citation pattern looks fine; the companion paper [10] appears only as motivation. Overall I'd take this paper seriously and send it to a referee. My recommendation: accept conditionally, with the Z-extension either expanded or trimmed.","headline":"A clean new result for open Richardson varieties over fields; the integer-case proof is too terse to be fully rigorous as written.","tokens_in":9702,"tokens_out":4733,"would_cite":true,"duration_ms":53030,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["14M15","14C20","13F15"],"pacs":[],"model":"deepseek-v4-flash","headline":"Open Richardson varieties have trivial divisor class groups","keywords":["divisor class group","open Richardson variety","Grassmannian","unique factorization domain","Nagata's criterion","Plücker coordinates","positroid variety","Picard group"],"falsifier":"Compute the divisor class group of a concrete open Richardson variety, for instance $\\mathrm{Gr}(2,4)$ with $\\beta=\\{1,2\\}$ and $\\gamma=\\{3,4\\}$; the theorem predicts it is zero, so any non-principal divisor would disprove it. A second check is to search for $p$-torsion in the coordinate ring of some positroid variety over $\\mathbb{Z}$; the paper's integer-case argument would break if such torsion existed.","tokens_in":8710,"feed_emoji":"","tokens_out":12351,"duration_ms":104205,"temperature":0.7,"pith_summary":"The paper proves a structural fact about every open Richardson variety $X^{\\gamma}_{\\beta}$ in a Grassmannian $\\mathrm{Gr}(k,n)$, over any field or over the integers: its divisor class group is trivial. That is, every Weil divisor on $X^{\\gamma}_{\\beta}$ is principal, and equivalently the coordinate ring is a unique factorization domain and the Picard group vanishes. This matters because smooth, affine, rational varieties need not have trivial divisor class groups, so the result supplies a genuinely new algebraic property for a large family of varieties that appear throughout Schubert calculus. The proof works without assuming the field is algebraically closed, which the authors need for an application over the real numbers.","feed_headline":"Open Richardson varieties have trivial divisor class groups","feed_subtitle":"Every divisor is principal, so each coordinate ring is a unique factorization domain, over any field or the integers.","key_machinery":"The argument is carried by Nagata's criterion, the standard lemma that a Noetherian domain is a unique factorization domain whenever a localization at a set of prime elements is a UFD. It is applied to the coordinate ring $A^{\\gamma}_{\\beta}$ with the multiplicative set generated by $\\Delta_{\\delta_1},\\ldots,\\Delta_{\\delta_{k-1}}$. Three objects do the work. The localized ring $B^{\\gamma}_{\\beta}$ and its Proj, the open subscheme $W^{\\gamma}_{\\beta}$, are shown to be parametrized by explicit $k\\times n$ matrices with prescribed zero, one, free, and invertible entries, making $W^{\\gamma}_{\\beta}$ a product of affine space and a torus and hence $B^{\\gamma}_{\\beta}$ a UFD. For each $t$, the set $\\mathcal{P}_t=\\{\\alpha\\in[\\beta,\\gamma]:\\alpha\\cap[\\beta(t+1),\\gamma(t)]\\neq\\varnothing\\}$ is shown to be a positroid, so the associated positroid variety is an integral scheme; Lemma 9 uses the Plücker relations to prove that every $\\Delta_{\\alpha}$ with $\\alpha\\in\\mathcal{P}_t$ lies in the ideal $\\langle\\Delta_{\\delta_t}\\rangle$, which identifies the quotient by $\\langle\\Delta_{\\delta_t}\\rangle$ with an integral subscheme. That makes $\\Delta_{\\delta_t}$ prime.","core_discovery":"The central claim is Theorem 1: for any interval $[\\beta,\\gamma]$ in the Bruhat order on $k$-element subsets of $[n]$, the open Richardson variety $X^{\\gamma}_{\\beta}$ has trivial divisor class group. Equivalently, its coordinate ring $A^{\\gamma}_{\\beta}$ is a unique factorization domain. The proof isolates the Plücker coordinates $\\Delta_{\\delta_0},\\ldots,\\Delta_{\\delta_k}$, where $\\delta_t=\\{\\beta(1),\\ldots,\\beta(t),\\gamma(t+1),\\ldots,\\gamma(k)\\}$, and shows two things. First, localizing $A^{\\gamma}_{\\beta}$ at all of them produces a ring $B^{\\gamma}_{\\beta}$ whose Proj is explicitly parametrized by matrices of a fixed block form, hence is a product of affine space and a torus and therefore has a UFD coordinate ring. Second, each intermediate $\\Delta_{\\delta_t}$ is either a unit of $A^{\\gamma}_{\\beta}$ or a prime element; the prime case is proved by identifying the vanishing locus of $\\Delta_{\\delta_t}$ with an open subscheme of a positroid variety, which is integral. Nagata's criterion then upgrades the localization statement to the UFD property of $A^{\\gamma}_{\\beta}$ itself.","pith_inferences":["Editorial inference: the same localization at the $\\delta_t$ Plücker coordinates should give explicit equations for the boundary divisors of any open Richardson variety, because a trivial class group means each boundary divisor is the zero set of a single regular function.","Editorial inference: the paper's integer-case argument would become uniform if positroid varieties were known to be torsion-free over $\\mathbb{Z}$; the Hilbert-function argument used for Richardson varieties is explicitly not available there, so proving integrality of positroid varieties over $\\mathbb{Z}$ is a natural next step.","Editorial inference: because $W^{\\gamma}_{\\beta}$ is a product of affine space and a torus, the localized ring $B^{\\gamma}_{\\beta}$ is a UFD with an explicit unit group, which may make the factorization structure of $A^{\\gamma}_{\\beta}$ visible through combinatorial tools such as valuations or Newton polytopes.","Editorial inference: a flag-variety analogue would need a replacement for the positroid integrality step, since that step is specific to the Grassmannian; the paper's description of each $X_{\\mathcal{P}_t}$ as an intersection of shifted Schubert varieties suggests where to look."],"forward_implications":["Every divisor on $X^{\\gamma}_{\\beta}$ is principal, and the Picard group of $X^{\\gamma}_{\\beta}$ is trivial.","The coordinate ring $A^{\\gamma}_{\\beta}$ is a unique factorization domain, so irreducible regular functions factor uniquely up to units.","The divisor class group of the closed Richardson variety $X^{\\gamma}_{\\beta}$ is generated by the boundary divisors of the open Richardson variety.","The result holds over any field and over the integers, so no algebraic closure hypothesis is needed for applications.","The open subscheme $W^{\\gamma}_{\\beta}$ admits an explicit description as the complement of a finite union of positroid varieties inside $X^{\\gamma}_{\\beta}$."],"supporting_citations":[{"why":"supplies Nagata's criterion, the lemma that reduces UFD-ness of $A^{\\gamma}_{\\beta}$ to a localization being a UFD and the localized-at elements being prime.","marker":"[4]"},{"why":"supplies the theorem that positroid varieties are integral schemes, which makes the vanishing locus of $\\Delta_{\\delta_t}$ integral and hence $\\Delta_{\\delta_t}$ prime.","marker":"[9]"},{"why":"characterizes positroids as images of Bruhat intervals, which is used to prove each set $\\mathcal{P}_t$ is a positroid.","marker":"[12]"}],"fun_headline_variants":["Every divisor on open Richardson variety is principal","Open Richardson varieties: all divisors principal","Open Richardson varieties: UFD coordinate rings","Class group vanishes for open Richardson varieties","Open Richardson varieties: class group trivial"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof that the elements $\\Delta_{\\delta_t}$ are prime depends on the external theorems that positroid varieties are integral and that each set $\\mathcal{P}_t$ is a positroid; for the integer version it also assumes Richardson varieties over $\\mathbb{Z}$ are integral, while the analogous integrality statement for positroid varieties over $\\mathbb{Z}$ is explicitly left unproved in the paper.","fun_headline_variants_meta":{"raw":{"variants":["Every divisor on open Richardson variety is principal","Open Richardson varieties: all divisors principal","Open Richardson varieties: UFD coordinate rings","Class group vanishes for open Richardson varieties","Open Richardson varieties: class group trivial"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000766,"raw_usage":{"total_tokens":3355,"prompt_tokens":862,"completion_tokens":2493,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":478,"completion_tokens_details":{"reasoning_tokens":2430}},"tokens_in":478,"tokens_out":2493,"duration_ms":18892,"temperature":1.0,"reasoning_tokens":2430,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:29:30.050335+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the divisor class group of a concrete open Richardson variety, for instance $\\mathrm{Gr}(2,4)$ with $\\beta=\\{1,2\\}$ and $\\gamma=\\{3,4\\}$; the theorem predicts it is zero, so any non-principal divisor would disprove it. A second check is to search for $p$-torsion in the coordinate ring of some positroid variety over $\\mathbb{Z}$; the paper's integer-case argument would break if such torsion existed.","supporting_citations":[{"cited_title":"Eisenbud, Commutative algebra with a view toward algebraic geometry , Graduate T exts in Math","cited_arxiv_id":null,"evidence_quote":"supplies Nagata's criterion, the lemma that reduces UFD-ness of $A^{\\gamma}_{\\beta}$ to a localization being a UFD and the localized-at elements being prime."},{"cited_title":"Knutson, T","cited_arxiv_id":null,"evidence_quote":"supplies the theorem that positroid varieties are integral schemes, which makes the vanishing locus of $\\Delta_{\\delta_t}$ integral and hence $\\Delta_{\\delta_t}$ prime."}],"review_version":1}