{"id":"15785f01-538f-4314-90b6-b50e9ec632f3","arxiv_id":"2505.22851","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The number of separating circles depends only on the two group sizes, and the cluster algebra of the higher-order Voronoi decomposition is independent of the dot configuration.","lead":"This paper counts how many circles can separate a finite set of dots into two groups, showing the count depends only on the two group sizes and not on the dot positions. It also proves that the associated higher-order Voronoi decompositions change by simple local moves, so the associated cluster algebra is independent of the configuration.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Lemma 20's local case analysis is the load-bearing gap: it asserts, without an exhaustive derivation, that crossing one cocircularity wall changes the kth-order Voronoi decomposition only by the Figure 2 moves.","rationale":"The reader's weakest-assumption identification is exactly the one I would make: the informal local model in Lemma 20 is the single load-bearing step for Theorem C. The counting arguments for Theorems A and B are clean and appear correct, and the surrounding topological setup in Section 6 is standard. The proof of Lemma 20, however, rests on an unstated case analysis whose conclusion is supported by figures rather than by a formal derivation. My read does not move the verdict: the paper should remain conditional until the local analysis is either completed rigorously or verified computationally. I would not reject the paper, because the gap is localized and plausibly repairable; I would not accept as-is, because the main structural theorem is not yet fully proved. Thus UNCHANGED is the right verdict from this stress-test pass.","tokens_in":13629,"tokens_out":10204,"duration_ms":114025,"concrete_test":"Implement an exact combinatorial verifier for small n (say n up to 8 and all 1 <= k <= n-1): enumerate every oriented matroid type of a semigeneral configuration with exactly one cocircular quadruple, sample one configuration on each side of the wall with exact rational or algebraic coordinates, compute the full kth-order Voronoi decomposition combinatorially for t<0 and t>0, and check that the abstract bicolored graphs differ only inside two small disks around the two centers of the common circle and exactly by one or two of the Figure 2 moves. This tests the exhaustiveness of the two cases in Figures 7 and 8, including whether any stationary vertex can appear inside the moving neighborhood.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Theorem C depends entirely on Lemma 20, which is the only step converting an arbitrary semigeneral deformation into a sequence of Figure 2 moves. The gap is the assertion, in the two bullet cases, that 'the restriction to the neighborhood U will be equivalent to one of the pictures in Figure 7' (and analogously Figure 8). This equivalence is stated, not derived. In particular, the proof does not explicitly rule out three possibilities: (i) additional vertices of the kth-order Voronoi decomposition lying inside U, produced by incident circles that are not among the four triples of {d1,d2,d3,d4} but whose centers happen to lie close to a; such stationary vertices would alter the local graph in a way not depicted in Figures 7 and 8; (ii) a third combinatorial pattern for the four moving centers, beyond the two signs d1 left or right of C1, because the bullet cases implicitly assume that the signs of d1,d2,d3,d4 relative to the other three are strictly alternating, which is plausible but not proved; and (iii) a change outside U and its antipode, since the argument that 'the corresponding vertices ... do not change' assumes that no previously existing edge or vertex passes through a or sigma(a). If any of these failure modes occurs, a single wall crossing is not necessarily one or two Figure 2 moves, and the connectivity statement of Theorem C would not follow. This is a genuine soft spot because the rest of the paper's structural conclusions, including Proposition 24 and the cluster-algebra independence, are downstream of this lemma. Warning 23 itself shows that the relationship between semigeneral families and Postnikov moves is delicate, so the informal appeal to pictures is not merely cosmetic.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies circles that separate a finite set of n points ('dots') in general position on the sphere or plane. It proves two counting results: Theorem A gives the number of incident circles (through three dots) that separate the remaining dots into subsets of sizes k and l, and Theorem B gives the number of equivalence classes of avoidant circles (through no dots) that induce a partition of sizes k and l; both formulas are configuration-independent. The proofs use a double-counting induction for incident circles (Proposition 7) and, for avoidant circles, the kth-order Voronoi decomposition of the sphere, whose vertices and regions are counted via a convex-hull argument and Euler's formula (Theorems 13). The paper also studies continuous deformations of configurations, claiming in Lemma 20 that crossing a single cocircularity wall changes the kth-order Voronoi decomposition by one of three local moves (or an antipodal pair), and in Theorem C that any two decompositions are related by a sequence of such moves. This implies, via known cluster-algebra results, that the associated cluster algebra depends only on k and n (Proposition 24). The paper is clearly written and well motivated, and it includes helpful remarks and a discussion of the relation between spherical and planar Voronoi decompositions.","tokens_in":13898,"tokens_out":6595,"duration_ms":71255,"significance":"If the results hold, the paper gives clean, explicit formulas for two natural counting problems and, more importantly, establishes a structural connection between spherical higher-order Voronoi decompositions and Postnikov's local moves for bicolored graphs. This has attractive consequences, including the configuration-independence of the associated cluster algebra and explicit identifications with the Markov and X7 cluster algebras. The counting proofs are elegant and self-contained: Proposition 7 is a sound double-counting induction, Proposition 6 is a clean convex-hull argument, and Theorem 13 correctly combines vertex counts with the Euler characteristic. The paper is also commendably honest about where its arguments are informal, such as Warning 23 on the limitations of the move realization. However, the main structural theorem (Theorem C) rests entirely on Lemma 20, whose proof relies on an unproved local model. Because Theorem C and Proposition 24 are the paper's most significant advertised contributions, this gap is load-bearing and needs to be addressed.","major_comments":[{"comment":"The proof asserts that, in the two bullet cases, 'the restriction to the neighborhood U will be equivalent to one of the pictures in Figures 7 and 8.' This assertion is not derived. The proof does not explicitly rule out the existence of additional vertices of the kth-order Voronoi decomposition inside U that are not among the four a_i(t) (e.g., from incident circles that do not correspond to the four triples of {d1,d2,d3,d4} but whose centers might approach a); it does not justify that the three cases |D−|=k−3,k−2,k−1 are the only possible ones; and it does not verify that the incidence pattern of edges and regions near a matches the figures. Since Lemma 20 is the only step that connects a semigeneral wall-crossing to the moves in Figure 2, Theorem C and Proposition 24 inherit this gap. The authors should either provide a rigorous local model, for instance by writing down the positions of the four centers relative to a and deriving the Voronoi regions explicitly, or give a formal continuity argument that no other vertices/edges can enter U and that the local graph is exactly one of the depicted ones.","section":"Section 7, Lemma 20"},{"comment":"The statement 'For any triple of dots besides the four triples in {d1,d2,d3,d4}, the corresponding incident circle does not cross a dot, and therefore has the same set of dots on either side for all t. As a consequence, the corresponding vertices in the kth order Voronoi decomposition and their adjacencies to other vertices do not change as t varies' is not fully justified. The constancy of the sets of dots on either side is true by continuity (since no other quadruple is cocircular at t=0), but the vertices themselves are centers of circles that move with t. The proof must also ensure that, for sufficiently small epsilon, none of these moving vertices collides with another vertex, lies on an edge, or crosses the boundary of U. This can likely be achieved by choosing epsilon small enough, since all such coincidences would require a cocircularity not present at t=0, but the argument is omitted. The authors should spell out this stability argument explicitly.","section":"Section 7, Lemma 20, final paragraph"},{"comment":"The two bullet cases assume that once the sign of d1(t) relative to C1(t) is fixed, the signs of d2(t), d3(t), d4(t) relative to C2(t), C3(t), C4(t) are forced (alternating). This is a standard property of four points on an oriented circle, but it is not proved in the paper. Since the orientation of C is chosen 'so that it passes through the dots in that order,' the authors should include a short proof or a reference establishing that the four points appear in the stated alternating order on the circles C_i(t). Without this, the exhaustiveness of the case split in Figures 7 and 8 is not fully transparent.","section":"Section 7, Lemma 20, bullet cases"}],"minor_comments":[{"comment":"The notation '0<k<n>3' is confusing; it should be written as 'n>3 and 0<k<n' (or simply '3<n and 0<k<n') in Theorem C, Definition 10, Proposition 11, and elsewhere.","section":"Throughout"},{"comment":"In the paragraph beginning 'Next, let Ci(t) denote the orientation-reversal of C_i(t)', the text lists 'σ(a1(p)), σ(a1(p)), σ(a1(p)), σ(a1(p))' with a repeated index; it should read 'σ(a1(t)), σ(a2(t)), σ(a3(t)), σ(a4(t))'.","section":"Section 7, Lemma 20, proof"},{"comment":"The sentence 'By deforming D(t) in a tubular neighborhood of W_I, we may assume that the set of intersections is discrete' is vague; a brief explanation (e.g., a transversality argument) would make the proof more transparent.","section":"Proposition 19, proof"},{"comment":"The proof of Lemma 18 is clear, but the phrase 'this consists of disjoint open intervals, which are submanifolds of dimension 1' could be simplified to 'this is a one-dimensional submanifold' for readability.","section":"Section 6, Lemma 18"}],"recommendation":"major_revision","confidential_remarks":"The paper is a good fit for a combinatorics journal. The counting results (Theorems A and B) are solid, and the exposition is generally clear. However, the advertised structural result (Theorem C and Proposition 24) depends on the unproved local model in Lemma 20. I believe this gap is fixable within the manuscript's scope: one can provide a rigorous case analysis or a continuity argument. If the authors can do so, the paper would be a strong contribution. If not, the paper would still be publishable on the strength of Theorems A and B alone, but the cluster-algebra invariance claim would need to be softened. I recommend major revision with a request for a fully rigorous proof of Lemma 20."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The headline is simple: this paper re-proves two known counting results (Theorems A and B) in a clean, self-contained way, and contributes a genuinely new structural theorem (Theorem C) about how kth-order spherical Voronoi decompositions change under deformations of the dot set. The proofs of the counting theorems are solid. The double-counting induction in Proposition 7 is transparent, and the derivation of the stratum counts in Theorem 13 from vertex counts plus Euler characteristic is correct. The paper is honest about the prior work [CAdlHPH22] and positions its own contribution clearly: not the formulas, but the proof and the structural consequence for cluster algebras.\n\nThe soft spot is Lemma 20, and it is the load-bearing wall for everything after it. The lemma asserts that a single semigeneral wall crossing only changes the Voronoi decomposition in two small neighborhoods (around the two centers of the common circle) and that the change is exactly one or two of the Figure 2 moves. The proof does not actually derive this. It asserts that the local picture is 'equivalent to one of the pictures' in Figures 7 and 8, without a case-by-case argument. The stress-test note identifies plausible failure modes: other incident circles could create extra vertices near the center; the sign-pattern of the four moving dots relative to the circle might not be exhaustive; and changes outside the neighborhood are dismissed with 'do not change' without proof. Given that Warning 23 itself shows the semigeneral family / Postnikov move relationship is delicate, this informal appeal to figures is a genuine gap, not a cosmetic annoyance.\n\nIf Lemma 20 is fixed, Theorem C and the cluster-algebra independence (Proposition 24) follow cleanly. The rest of the paper is careful, and the appendix gives a useful connection to planar Voronoi diagrams. The future-direction section is speculative but clearly labeled.\n\nVerdict: this deserves serious peer review. The counting part is publishable as a note, and the structural theorem is interesting enough to warrant a rigorous lemma. I'd ask the authors for a complete proof of Lemma 20 — either a full case analysis or a more careful geometric argument establishing the locality and the exhaustive list of local patterns. With that, it's a solid paper. Without it, Theorem C remains a conjecture supported by pictures.","headline":"A clean spherical-Voronoi proof of known counting formulas plus a genuinely new local-moves theorem, but the key lemma is asserted from figures rather than proved.","tokens_in":14495,"tokens_out":3059,"would_cite":true,"duration_ms":32765,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["52B05","68U05","05E15","13F60"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves that separating-circle counts for point configurations are universal, and that the associated cluster algebra depends only on the number of dots and the order of the Voronoi decomposition.","keywords":["spherical Voronoi diagram","higher order Voronoi diagram","plabic graph","cluster algebra","incident circles","avoidant circles","general position","configuration space"],"falsifier":"Take a configuration of six dots, $k=2$, and a path through configuration space that crosses exactly one cocircularity wall. Compute the bicolored Voronoi graphs immediately before and after the crossing; if the two graphs differ by a move not appearing in Figure 2, or if the change occurs outside the two neighborhoods around the centers of the common circle, then Lemma 20 and Theorem C are false.","tokens_in":13412,"feed_emoji":"⭕","tokens_out":10058,"duration_ms":96677,"temperature":0.7,"pith_summary":"Given $n$ dots in general position in a plane or sphere, the paper proves that the number of circles through three dots that split the remaining dots into parts of sizes $k$ and $\\ell$ does not depend on where the dots are, and gives the formula $2(k+1)(\\ell+1)$ (with $(k+1)^2$ when $k=\\ell$). The same universality holds for partitions achievable by a circle that passes through no dots, with the formula $2k\\ell-k-\\ell+2$ (or $k^2-k+1$ when $k=\\ell$). The proof works by counting cells in the $k$th order Voronoi decomposition of the sphere, and shows that as the dots move, this decomposition changes only through a finite list of local moves. A consequence is that the cluster algebra associated to the decomposition depends only on $k$ and $n$, not on the particular configuration of dots. These universal counts matter because they convert a configuration-dependent geometric question into a fixed combinatorial answer.","feed_headline":"Same number of dots, same separating-circle count","feed_subtitle":"For any n dots, circle-split counts are fixed; Voronoi moves make the cluster algebra depend only on k,n.","key_machinery":"The $k$th order Voronoi decomposition of the sphere: a partition of the sphere into regions, edges, and vertices according to which $k$-element subset of the dots is closest to each point. Its vertices are left centers of oriented incident circles and are colored white or black according to whether the circle has $k-2$ or $k-1$ dots on its left side; the strata form a 3-regular bicolored graph. Euler characteristic then converts a count of vertices into a count of regions, and the bijection between regions and equivalence classes of oriented avoidant circles yields the circle-partition formula. The local-move mechanism is Lemma 20, which asserts that crossing a single cocircularity changes the graph only by the moves shown in Figure 2.","core_discovery":"The paper's central claim is that, for $0<k<n$ and $n>3$, the $k$th order Voronoi decomposition of the sphere determined by $n$ dots is, up to a small explicit list of local moves, independent of the dots' positions. Concretely, the decomposition is a 3-regular bicolored graph whose numbers of white vertices, black vertices, edges, and regions are fixed by $k$ and $n$; counting incident circles is the same as counting vertices by color, and counting avoidant circles is the same as counting regions. When four dots become cocircular during a deformation, the decomposition changes only in two small neighborhoods, and the change is one of three local moves (or an antipodal pair), so any two configurations are connected by a sequence of such moves. Since each move induces an isomorphism of the associated cluster algebra, the isomorphism class of that cluster algebra depends only on $k$ and $n$.","pith_inferences":["If the locality principle behind Lemma 20 holds in other settings, the same argument could yield configuration-independent cluster algebras for Voronoi-type decompositions on other surfaces or in other metric spaces.","A concrete testable extension is to explore, for small $n$ and $k$, whether the isomorphism class of the cluster algebra distinguishes configurations that are not connected by a sequence of allowed moves; the paper's Warning 23 indicates such obstructions can exist.","The paper's suggestion to derive planar counting formulas from spherical ones reverses the direction of earlier proofs, which could make future enumeration arguments shorter by working entirely on the sphere."],"forward_implications":["For any $n$ dots in general position in the plane or sphere, the number of incident circles separating the dots into parts of sizes $k$ and $\\ell$ is $2(k+1)(\\ell+1)$ when $k\\neq\\ell$, and $(k+1)^2$ when $k=\\ell$.","The number of partitions of $n$ dots into non-empty parts of sizes $k$ and $\\ell$ that can be separated by an avoidant circle is $2k\\ell-k-\\ell+2$ when $k\\neq\\ell$, and $k^2-k+1$ when $k=\\ell$.","The isomorphism class of the cluster algebra associated to the $k$th order Voronoi decomposition depends only on $k$ and $n$, not on the configuration of dots.","For four dots and $k=2$, the resulting cluster algebra is the Markov cluster algebra; for six dots and $k=3$, it is the $X_7$ cluster algebra.","The local moves preserve the numbers of white vertices, black vertices, edges, and regions, providing a second proof that these stratum counts depend only on $k$ and $n$."],"supporting_citations":[{"why":"Introduces the three local moves on bicolored graphs and the equivalence relation that Theorem C adapts to Voronoi decompositions.","marker":"[Pos06]"},{"why":"Proves Theorems A and B in full generality and supplies the gluing theorem connecting spherical and planar Voronoi decompositions.","marker":"[CAdlHPH22]"},{"why":"Provides the result that each local move induces an isomorphism of cluster algebras, the step that turns Theorem C into Proposition 24.","marker":"[FWZ25]"},{"why":"Establishes the k=1 spherical–planar gluing result that underlies the appendix's comparison.","marker":"[Bro79]"},{"why":"Adds explicit maps and the color-swapping gluing statement needed for the full planar–spherical correspondence.","marker":"[NLC02]"},{"why":"Records the planar Voronoi symmetry that earlier proofs relied on and that the present argument rederives from spherical counts.","marker":"[Lin03]"},{"why":"Treats the k=ℓ case of the incident-circle count, giving the special-case benchmark that the general formula extends.","marker":"[Ard04]"}],"fun_headline_variants":["Circle-split counts fixed for any point set","For n dots, separating-circle counts are independent of positions","Voronoi moves fix circle counts and cluster algebra","Cluster algebra depends only on dot count and order","Circle counts and cluster algebra invariant under dot configs"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument depends on Lemma 20's local picture: when four dots pass through cocircularity, the Voronoi decomposition is assumed to change only in two small neighborhoods, and the change is assumed to be exactly one of the moves in Figures 7 and 8; if that locality or the exhaustiveness of those pictures fails, Theorem C and the cluster algebra invariance collapse.","fun_headline_variants_meta":{"raw":{"variants":["Circle-split counts fixed for any point set","For n dots, separating-circle counts are independent of positions","Voronoi moves fix circle counts and cluster algebra","Cluster algebra depends only on dot count and order","Circle counts and cluster algebra invariant under dot configs"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000207,"raw_usage":{"total_tokens":1360,"prompt_tokens":867,"completion_tokens":493,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":483,"completion_tokens_details":{"reasoning_tokens":418}},"tokens_in":483,"tokens_out":493,"duration_ms":5377,"temperature":1.0,"reasoning_tokens":418,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T12:58:59.066276+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a configuration of six dots, $k=2$, and a path through configuration space that crosses exactly one cocircularity wall. Compute the bicolored Voronoi graphs immediately before and after the crossing; if the two graphs differ by a move not appearing in Figure 2, or if the change occurs outside the two neighborhoods around the centers of the common circle, then Lemma 20 and Theorem C are false.","supporting_citations":[],"review_version":1}