{"id":"ed8cd30e-6e4e-4cb1-9ed5-cd96bae36cc6","arxiv_id":"2501.07065","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Compatibility-degree weights of cluster variables lie in the Gröbner cone of any finite-type cluster algebra, yielding explicit circular term orders and, for classical types, complete ray and lineality descriptions.","lead":"This paper shows that for every finite-type cluster algebra, the compatibility degree with any fixed cluster variable produces a valid monomial order that degenerates the algebra's relation ideal to its cluster complex ideal. It then uses this construction to describe the full set of such orders, the Gröbner cone, for types A, B, C, and D, and to prove a conjecture from earlier work.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Exceptional-type proof rests on an unreproduced Sage computation; a flaw in the §4.3 exchange-relation reconstruction would invalidate Theorem 4.1.1 for E6–E8, F4, G2.","rationale":"The reader's weakest-assumption identifies the same point: the exceptional-type computer verification. I agree that this is the most load-bearing concern. The proof architecture is otherwise strong: Theorem 4.1.1 follows from Proposition 4.1.2 via Lemma 2.2.4, and the classical types have a detailed combinatorial argument. The 'by inspection' lemmas (4.2.1, 5.3.2, 5.4.1) are terse but appear correct; spot-checking Lemma 4.2.1 against degenerate line positions does not reveal a counterexample. The reduction to no frozen variables and indecomposable blocks is sound, and the derivation of Corollary 4.4.1 from the cone membership is a genuine proof of the INCT21 conjecture, independent of the exceptional computation except through Theorem 4.1.1. Thus the only serious risk to the central claim is the unreproduced computational verification for E6–E8, F4, G2, together with the unproven correctness of the exchange-relation reconstruction algorithm. Because the ancillary code is referenced and the algorithm is described in enough detail to reimplement, this is a conditional-accept concern rather than grounds for rejection. If the independent check reproduces the claimed inequalities and exchange relations, the concern is resolved and Theorem 4.1.1 is established for all finite types.","tokens_in":30531,"tokens_out":29599,"duration_ms":279040,"concrete_test":"Run the ancillary SageMath code [IS25] in a clean Sage 9.5 environment. For each of E6, E7, E8, F4, G2, independently cross-check the reconstructed exchange relations: take a sample of exchangeable pairs (compatibility degree 1), compute the exchange relation via direct mutation in Sage's cluster algebra module or an independent implementation of Theorem 2.3.4, and verify that the maximum-clique reconstruction yields the same two RHS monomials. Then re-run the full check that (4.1.3) holds for every cluster variable v and every exchangeable pair. If any mismatch is found, Theorem 4.1.1 is unsupported for the exceptional types.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim Theorem 4.1.1 is proved by reducing to Proposition 4.1.2 and then checking that proposition case-by-case. Types A_n, B_n, C_n, D_n receive a human-readable combinatorial proof in §4.2. The exceptional types E6, E7, E8, F4, G2 are handled only by the SageMath verification described in §4.3, with code relegated to the ancillary file [IS25]. That verification is load-bearing because no other argument covers these five types. The described algorithm has unstated correctness assumptions: it reconstructs each exchange relation by taking a maximum clique S_{λ,μ} in the graph of cluster variables compatible with both exchangeable weights, asserts this clique is the common cluster, and asserts that both RHS weights λ+μ and λ⊎μ have non-negative expansions on S_{λ,μ}∪{λ}. If the maximum clique is not unique or if the expansion fails for some pair, the script could silently omit or misstate an exchange relation, and the inequality (4.1.3) might fail for a type not covered by any written argument. The same computational dependency appears in Remark 4.5.4, which uses SageMath to determine T-orbit parities. Since the paper offers no independent confirmation of the script's output and no machine-checked proof, the universality of Theorem 4.1.1 is not yet self-contained.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the Gröbner cone C_A parametrizing weights whose initial degeneration of the relation ideal I_A of a finite-type cluster algebra A is the Stanley-Reisner ideal I^×_A generated by products of incompatible cluster variables. The main theorem (Theorem 4.1.1) asserts that for every cluster variable v, the compatibility-degree vector ω_v = [(v||y)]_{y ∈ V∪W} lies in C_A. The authors prove this by reducing to indecomposable finite types: a combinatorial cross-counting argument covers A_n, B_n, C_n, D_n (§4.2), and a SageMath verification covers E_6, E_7, E_8, F_4, G_2 (§4.3). From this they derive an interior point of C_A giving a circular term order (Corollary 4.5.1), prove Conjecture 6.2.9 of [INCT21] (Corollary 4.4.1), and give explicit ray and lineality-space generators for the cone in the classical types, both with the special frozen variables of §3 (Propositions 5.1.1–5.1.2, Theorems 5.2.1–5.2.2) and without frozen variables (Theorems 6.1.1–6.1.4).","tokens_in":30929,"tokens_out":5015,"duration_ms":52544,"significance":"If fully justified, the results are a substantial contribution to the combinatorial and computational algebra of cluster algebras. Theorem 4.1.1 is a clean universal statement with a short and natural proof once Proposition 4.1.2 is established, and it yields concrete consequences: an explicit circular term order, a proof of a published conjecture, and complete ray/lineality descriptions that recover Speyer and Sturmfels' tropical Grassmannian result as a special case. A particular strength of the paper is that the central theorem is not obtained by fitting parameters; it is a proof by reduction to finite types, with the computational part explicitly isolated and honestly identified. The explicit descriptions in §§5–6 are falsifiable and likely to be useful for future work on degenerations of cluster algebras. The main unresolved question is whether the computational verification for the exceptional types is sufficiently self-contained and independently checkable.","major_comments":[{"comment":"The universal assertion of Theorem 4.1.1 depends on the verification of Proposition 4.1.2 for E6, E7, E8, F4, and G2. The text describes an algorithm but the only evidence is the SageMath code relegated to the ancillary reference [IS25]; the correctness of the exchange-relation reconstruction is not independently established. In particular, taking S_{λ,μ} to be \"a maximum clique\" of the graph of variables compatible with λ and μ assumes that this clique is the common extended cluster; non-uniqueness, or failure of the expansion of λ+μ and λ⊎μ to have non-negative coefficients on S_{λ,μ}∪{λ}, would silently invalidate the inequality (4.1.3). The same auxiliary computation underlies Remark 4.5.4 and therefore Theorem 4.5.5 for F4. Please include the code or, at minimum, a precise certificate of its verified output for each type, and state explicitly the correctness conditions under which the clique-reconstruction algorithm recovers every exchange relation.","section":"§4.3, Proposition 4.1.2"},{"comment":"The proof of Proposition 5.1.2 and Theorem 5.2.2 in type D_n relies on the assertion that every extended exchange matrix in the D_n model has rank n−1, so that the lineality space of C_A has dimension one more than the number of frozen variables. The text says this is \"straightforward to verify\" but gives no argument or reference. Since the claimed basis of the lineality space and the ray-generation statements both depend on this rank fact, please supply a proof or an explicit citation.","section":"§5.3, Lemma 5.3.1"}],"minor_comments":[{"comment":"The proof of Lemma 4.2.1 is by inspection of three configurations. The combinatorial claim is plausible and the figure helps, but since Proposition 4.1.2 for all classical types rests on this lemma, a short coordinate-based proof or an explicit list of crossing inequalities would make the argument easier to verify.","section":"§4.2, Lemma 4.2.1"},{"comment":"There is a typo in the sentence \"it is straightforward to calculate that the Gröbner cone A has lineality space generated by\"; it should read \"the Gröbner cone C_A of A has lineality space generated by\".","section":"Example 3.2.3"},{"comment":"The manuscript contains several spacing and punctuation artifacts, such as \"Theorems 5.1.1 5.1.2\" in the Introduction and \"the coneCA\" in Definition 2.2.3. A careful copyedit would improve readability.","section":"Throughout"},{"comment":"The reference [INCT21] is cited as arXiv:2111.02566v1; if a later or published version exists, the authors should update the citation.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The mathematical architecture of the paper is convincing: the reductions are natural, the classical-type arguments are detailed, and the computational part is clearly separated. My recommendation is driven by two load-bearing gaps: the lack of an independently checkable certificate for the exceptional-type verification in §4.3 and the unproved rank assertion in the D_n case of Lemma 5.3.1. Both are fixable within the scope of a revision, so I do not recommend rejection. If the authors provide the code/certificates and a proof of the rank fact, I would be inclined to accept."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my read. The paper has two real achievements. First, Theorem 4.1.1: for any cluster variable v, the compatibility-degree vector ω_v lies in the Gröbner cone. That's a clean, general statement, and the proof via Proposition 4.1.2 is sensible: reduce to no-frozen, indecomposable, then handle classical types combinatorially and exceptional types by computation. Second, the explicit generators for rays and lineality spaces in types A–D, with and without frozen variables, are new and concrete, and they recover the Speyer–Sturmfels description for G(2,n). The proof of Conjecture 6.2.9 of INCT21 is a nice bonus and follows from the same circle of ideas.\n\nThe classical-type proofs are the real meat. Lemma 4.2.1 is asserted by inspection of three configurations; I believe it, but a referee should ask the authors to spell it out or justify it more algebraically. The type D case analysis in §4.2 is intricate, but the exchange quadrilaterals make it readable. The reduction to no-frozen and indecomposable blocks is sound.\n\nThe soft spot is the exceptional types, E6–E8, F4, G2. Proposition 4.1.2 there rests on a Sage verification, and the code is only in the ancillary file [IS25]. That is not fatal—the algorithm in §4.3 is described carefully, including how exchange relations are reconstructed from g-vectors and compatibility via maximum cliques. But it is load-bearing. If the clique reconstruction silently fails on some pair, Theorem 4.1.1 could be false for an exceptional type. I'd want the code to be run end-to-end and the output logged before treating the universal statement as fully checked. I'd be surprised if it fails, but surprise isn't proof. The same caveat applies to the parity claim in Remark 4.5.4, which feeds into the F4 pointedness result. None of this undermines the classical-type results, which stand on their own.\n\nThe citation pattern is fine: results from INCT21 are used in the direction in which they were proved, and the authors are explicit about what is new and what is a computer calculation. The paper is honest about its own limits.\n\nWho is this for? Anyone working on cluster algebra deformations, Gröbner bases, or the tropical Grassmannian. It's a within-subfield advance, not a breakthrough, but it's a solid one. I'd send it to a serious referee, and I'd want the ancillary code bundled with the final version, ideally with a short log showing the script ran to completion.","headline":"A solid, genuinely new description of Gröbner cones for finite type cluster algebras, with a load-bearing but plausibly correct computer check for the exceptional types.","tokens_in":31368,"tokens_out":3624,"would_cite":true,"duration_ms":32361,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["13F60","13P10"],"pacs":[],"model":"deepseek-v4-flash","headline":"Compatibility-degree vectors of any cluster variable lie in the Gröbner cone for every finite-type cluster algebra, and the cone is explicitly described in types A_n, B_n, C_n, D_n.","keywords":["cluster algebras","Gröbner cones","compatibility degree","finite type","term orders","initial ideals","associahedra","Dynkin types"],"falsifier":"Independently recompute, for type $E_8$, the complete list of cluster variables, all compatibility-degree vectors $\\omega_v$, and all primitive exchange relations from the Coxeter-element model, then test inequality (4.1.3) for every cluster variable and every primitive relation. Finding one $v$ and one primitive relation where the inequality fails would disprove Theorem 4.1.1; a clean independent pass would convert the asserted computational verification into reproducible evidence.","tokens_in":30329,"feed_emoji":"📐","tokens_out":10749,"duration_ms":93043,"temperature":0.7,"pith_summary":"Every cluster algebra of finite cluster type carries two natural ideals: the relation ideal $I_A$ among its cluster and frozen variables, and the monomial ideal $I^\\times_A$ generated by products of incompatible cluster variables. The paper proves that for any cluster variable $v$, the weight vector $\\omega_v = [(v||y)]_{y \\in V \\cup W}$ whose entries are compatibility degrees belongs to the Gröbner cone $C_A$ of weights that degenerate $I_A$ to $I^\\times_A$. It follows that the sum of all such vectors lies in the interior of $C_A$, producing an explicit circular term order for every finite-type cluster algebra. For types $A_n,B_n,C_n,D_n$, the paper gives complete polyhedral descriptions of $C_A$---bases of the lineality space and generators of the rays---in the models with special frozen variables and in the no-frozen case. It also proves the deformation-theory conjecture that nontrivial first-order embedded deformations of $\\operatorname{Spec}(K[z]/I^\\times_A)$ cannot have their negative multidegrees in the semigroup coming from primitive exchange relations.","feed_headline":"Compatibility-degree vectors fill every finite-type Gröbner cone","feed_subtitle":"Every finite-type cluster algebra gets a circular term order from compatibility degrees; types A-D have explicit cones.","key_machinery":"The load-bearing machinery is the compatibility-degree vector $\\omega_v$ together with the dual description of the Gröbner cone: by Lemma 2.2.4, $C_A$ is the dual cone of $C_{\\mathrm{prim}}$, the cone generated by degrees $\\deg(xx') - \\deg(y_1)$ of primitive exchange relations $xx' = y_1 + y_2$. To place $\\omega_v$ in $C_A$ it suffices to check that $\\omega_v$ has nonnegative dot product with every such degree, and Proposition 4.1.2 proves the stronger balancing statement that left and right sides give the same maximum. In the classical types the geometry of exchange quadrilaterals turns this algebraic check into crossing counts: a cluster variable $v$ corresponds to a diagonal or pair of diagonals, and $\\omega_v \\cdot \\deg(y_i)$ equals the number of crossings between that diagonal and the sides of the quadrilateral encoding the monomial $y_i$ (up to a constant). For the exceptional types the same inequality is checked computationally through the Coxeter-element construction, where cluster variables are weights $c^k\\omega_i$ and exchange relations are produced from the two-element set $\\{\\lambda+\\mu, \\lambda \\uplus_c \\mu\\}$ associated to exchangeable weights.","core_discovery":"The paper's central claim is Theorem 4.1.1: for any cluster algebra $A$ of finite cluster type, with cluster variables $V$ and frozen variables $W$, and for any $v \\in V$, the vector $\\omega_v = [(v||y)]_{y \\in V \\cup W}$ belongs to the Gröbner cone $C_A$ of $I_A$ with respect to $I^\\times_A$. The proof of this membership is reduced to Proposition 4.1.2, which asserts that for every primitive exchange relation $xx' = y_1 + y_2$ and every $v \\neq x,x'$, the weight $\\omega_v$ pairs equally with the left-hand monomial and with the maximum of the two right-hand monomials. In the classical types this equality is proved by translating exchange relations into quadrilaterals in regular polygons, where dot products become crossing counts and the relevant inequality becomes a small geometric lemma about line segments crossing a quadrilateral. In the exceptional types the same equality is verified by a computer calculation, using the Coxeter-element model to compute compatibility degrees and exchange monomials. The paper then upgrades the membership statement into explicit structure theorems: the sum of all $\\omega_v$ lies in the interior of $C_A$, and in types $A_n,B_n,C_n,D_n$ the cone's lineality space and rays are described by explicit vectors built from diagonals of the polygon model.","pith_inferences":["The realization that all $\\omega_v$ lie in $C_A$ suggests that in the exceptional types the rays may also be expressible as small alternating sums of compatibility-degree vectors; extending the computer check to search for such expressions would give structural formulas matching the classical ray lists.","Because setting frozen variables to $1$ recovers the no-frozen case, the explicit polyhedral descriptions here are compatible with coefficient specialization; this suggests that other choices of frozen variables should produce Gröbner cones obtained by linear projections of the same cones, connecting these ray lists to tropical cluster varieties.","The circular term order produced by summing all $\\omega_v$ is uniform but likely far from minimal; a testable extension is to find whether a small subset of cluster variables already gives an interior point of $C_A$, which would yield cheaper term orders for explicit computation.","The computer verification for exceptional types is a black box in the paper; an independent, non-computational proof of Proposition 4.1.2 for one exceptional type would likely reveal combinatorial structures analogous to the exchange quadrilaterals used in the classical types."],"forward_implications":["Every finite-type cluster algebra has at least one explicit circular term order: the weight $\\omega = \\sum_v \\omega_v$ lies in the interior of $C_A$, so the initial ideal of $I_A$ is exactly $I^\\times_A$ under a term order constructed directly from compatibility degrees.","For any fixed cluster variable $v$, the weight $\\omega_v$ is itself circular, so one can build many distinct degenerations of $I_A$ to $I^\\times_A$ by varying the chosen $v$.","In types $A_n,B_n,C_n,D_n$ the Gröbner cone is completely determined: the paper's lists give bases of the lineality space and generators of the rays, both with the special frozen variables and with no frozen variables.","The deformation conjecture, stated as Conjecture 6.2.9 in the earlier deformation-theory paper, is proven: if a first-order embedded deformation of $\\operatorname{Spec}(K[z]/I^\\times_A)$ is nontrivial, then the negative of its multidegree is not a nonnegative combination of primitive exchange degrees.","The type $A_n$ result recovers the known Gröbner-cone description for the Plücker coordinate ring of the Grassmannian $G(2,n+3)$, placing it as a special case of the cluster-algebra framework."],"supporting_citations":[{"why":"Supplies the classification of finite cluster type via Cartan matrices and the identification of cluster variables with almost positive roots.","marker":"[FZ03a]"},{"why":"Defines compatibility degree and gives the polygon-based models for classical types used throughout Sections 3-6.","marker":"[FZ03b]"},{"why":"Provides the Coxeter-element construction, the $c$-compatibility degree formula, and Theorem 2.3.4 for exchange relations, used for the exceptional-type computation.","marker":"[YZ08]"},{"why":"Corollary 4.2 supplies the description of the two exchange monomials for exchangeable weights, completing the reconstruction of exchange relations in exceptional types.","marker":"[Ste13]"},{"why":"Establishes that $I^\\times_A$ is an initial ideal, gives the dual-cone characterization in Lemma 2.2.4, and states the conjecture proven as Corollary 4.4.1.","marker":"[INCT21]"},{"why":"Provides the Gröbner-cone description for the Grassmannian $G(2,n)$ that the paper recovers as its type $A_n$ special case.","marker":"[SS04]"}],"fun_headline_variants":["Compatibility vectors determine all finite-type Gröbner cones","Circular term orders exist for every finite-type cluster algebra","Explicit Gröbner cones for cluster algebras of types A-D","Conjecture on cluster algebra term orders proved via compatibility"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the computer verification for the exceptional types $E_6,E_7,E_8,F_4,G_2$ is correct; the code is only referenced as an accompanying file, not reproduced in the paper, so if the computation or the reconstruction of exchange relations from $g$-vectors contains an error, Theorem 4.1.1 would fail for those types.","fun_headline_variants_meta":{"raw":{"variants":["Compatibility vectors determine all finite-type Gröbner cones","Circular term orders exist for every finite-type cluster algebra","Explicit Gröbner cones for cluster algebras of types A-D","Conjecture on cluster algebra term orders proved via compatibility"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001631,"raw_usage":{"total_tokens":6516,"prompt_tokens":1005,"completion_tokens":5511,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":621,"completion_tokens_details":{"reasoning_tokens":5442}},"tokens_in":621,"tokens_out":5511,"duration_ms":37145,"temperature":1.0,"reasoning_tokens":5442,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T20:50:05.546724+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Independently recompute, for type $E_8$, the complete list of cluster variables, all compatibility-degree vectors $\\omega_v$, and all primitive exchange relations from the Coxeter-element model, then test inequality (4.1.3) for every cluster variable and every primitive relation. Finding one $v$ and one primitive relation where the inequality fails would disprove Theorem 4.1.1; a clean independent pass would convert the asserted computational verification into reproducible evidence.","supporting_citations":[],"review_version":1}