{"id":"1f5c4ddc-f42a-44ab-9588-6140e137750a","arxiv_id":"2501.13374","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The common vertices of the permutohedron and Loday's associahedron form exactly the maximal never-middle Condorcet domain of size 2^(n-1), namely the permutations avoiding 132 and 231.","lead":"This short paper proves that Loday's geometric realization of the associahedron shares exactly 2^(n-1) vertex points with the permutohedron. These points form a well-known maximal Condorcet domain from voting theory, linking two central polytopes to preference aggregation.","discovery_kind":"unification","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The theorem's set equality is false: the intersection of the two convex polytopes contains segments between common vertices, so it is infinite, not the finite 2^{n-1} domain; the proof only establishes equality of the common vertices.","rationale":"The paper's most significant claim is the identification of the set of common vertices of Perm_n and Asso_n with the maximal never-middle Condorcet domain. The inductive proof is essentially correct for that intended claim: it shows that any vertex of Asso_n that lies in Perm_n must have a root weight n, after which the facet property of Perm_n (which is standard and true for vertices) transfers the remaining coordinate vector into Perm_{n-1} ∩ Asso_{n-1}, allowing induction. The base case n=3 checks out. So the underlying combinatorial result is sound. The load-bearing flaw is the theorem's literal statement 'Perm_n ∩ Asso_n = D'. Since Perm_n and Asso_n are convex, their intersection is convex; whenever D contains two distinct points (n≥3), the segment between them is contained in the intersection. These segment points are not in the finite set D. Therefore the equality is false. This is not a stylistic issue: a reader relying on the theorem statement would be misled. The proof never attempts to characterize non-vertex intersection points, and indeed cannot, because the intersection is an infinite convex set. The reader's mentioned facet property is a genuine gap in the write-up but is secondary because it is standard and the proof can be read as applying to vertices once the theorem is restated appropriately. I recommend retaining the reader's conditional verdict: the paper should state the result precisely as an equality of the set of common vertices (or 'vertices of Asso_n in Perm_n') with D, and add a sentence justifying the facet property used in the induction. With these corrections, the result appears correct. Hence I agree with the conditional disposition but for a slightly different primary reason.","tokens_in":3895,"tokens_out":11307,"duration_ms":758685,"concrete_test":"For n=3, compute p = (1/2)((1,2,3)+(2,1,3)) = (1.5, 1.5, 3). Verify that p is a convex combination of vertices of Perm_3 (e.g., of (1,2,3) and (2,1,3)) and also a convex combination of vertices of Asso_3 (the same two points are vertices of Asso_3 as shown in Fig. 2), so p ∈ Perm_3 ∩ Asso_3. Show that p ∉ D_3,2 = {(1,2,3),(2,1,3),(3,1,2),(3,2,1)}. This directly falsifies the set equality in Theorem 2. For general n, take any two distinct elements of D and their midpoint, which will be in the intersection but not in the finite set D.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim, Theorem 2, asserts that Perm_n ∩ Asso_n equals the finite set D = ((...(1⋆2)⋆3)...⋆n). This is not merely imprecise; it is false as stated. Both Perm_n and Asso_n are convex polytopes, so their intersection is a convex set. For n=3, the common vertices include (1,2,3) and (2,1,3), both in D. The midpoint of these two points lies on the segment between them and hence, by convexity, belongs to Perm_3 ∩ Asso_3; however, it is not one of the four permutations in D. Thus the intersection contains points outside D, so the equality cannot hold for any n where D has at least two distinct elements (which it does for all n≥3). The proof itself only uses the condition M_n(t) ∈ Perm_n to show that a vertex of Asso_n lying in Perm_n must be a permutation in D; it never addresses non-vertex intersection points. A secondary issue is that the induction step implicitly uses the standard facet property of Perm_n (deleting a coordinate with value n from a point in Perm_n leaves a point in Perm_{n-1}), which should be stated; but the primary defect is that the theorem's statement overclaims by identifying the entire polytope intersection with a finite vertex set.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies Loday's polytopal realization of the associahedron Asso_n and its intersection with the permutohedron Perm_n. It states (Theorem 2) that Perm_n ∩ Asso_n equals the iterated ⋆-composition domain ((...(1⋆2)⋆3)...⋆n), a maximal never-middle Condorcet domain of size 2^{n-1}, equivalently the set of permutations avoiding the patterns 132 and 231. The proof is an induction on the root weight of the binary tree corresponding to a vertex of Asso_n.","tokens_in":4143,"tokens_out":3315,"duration_ms":29741,"significance":"The proposed connection between Loday's coordinates, pattern avoidance, and maximal Condorcet domains is attractive and potentially useful. The inductive argument is elementary and, once the statement is corrected to concern common vertices rather than the full polytope intersection, it gives a clean proof of a genuine combinatorial fact. The paper also explicitly identifies the domain as the known maximal never-middle domain, relying on the cited Karpov–Slinko result rather than reproving maximality, which is appropriate for a short note. However, the main theorem as stated is mathematically false, so the contribution needs a substantive correction before it can be accepted.","major_comments":[{"comment":"The statement 'Perm_n ∩ Asso_n = (...((1⋆2)⋆3)...⋆n)' is false as written. Since Perm_n and Asso_n are convex polytopes, their intersection is convex and hence infinite for n ≥ 3. For example, when n = 3, the points (1,2,3) and (2,1,3) are common vertices lying in D, so their midpoint (1.5,1.5,3) belongs to Perm_3 ∩ Asso_3 by convexity, but it is not a permutation and is not in D. The proof in §4 only establishes that the common vertex sets coincide, i.e. that the set of vertices of Asso_n that lie in Perm_n equals D (and conversely). The theorem and the abstract should be restated accordingly, for example as 'the set of common vertices of Perm_n and Asso_n equals D'.","section":"Theorem 2, §4"},{"comment":"The induction step implicitly uses the facet property of the permutohedron: if a point of Perm_n has a coordinate equal to n, then deleting that coordinate leaves a point of Perm_{n-1} on the remaining coordinates. This property is standard, but it is load-bearing: without it, the condition M_n(t) ∈ Perm_n does not by itself place the reduced coordinate vector into Perm_{n-1}. The same reduction for Asso_n follows directly from Loday's coordinate definition, but the permutohedron side should be stated and justified explicitly. This is not the same as the theorem's claim, so it does not create circularity, but it must be made visible in the proof.","section":"Proof of Theorem 2, §4"}],"minor_comments":[{"comment":"The text says that S_n is in a bijection with full binary trees with levels, but then defines a surjective map ψ : S_n → Y_n; the terminology should be adjusted to distinguish the level-labeled trees from the unlabeled ones in Y_n.","section":"§2"},{"comment":"The notation t−n is used before it is clearly defined; writing t_-n and explicitly stating t_-n ∈ Y_{n-1} would improve readability.","section":"§4, displayed trees"},{"comment":"The expression 'Mt−1(t−n)' should be 'M_{n-1}(t_-n)' to avoid confusion between the index and the argument.","section":"§4, proof"},{"comment":"The sentence 'Perm_n ∩ Asso_n consists of all permutations of S_n that avoid patterns 132 and 231' inherits the false equality from Theorem 2; after the theorem is corrected to 'common vertices', this note should be updated to match.","section":"Note 1"},{"comment":"The sentence 'For the tree t on Figure 1 we have M_3(t) = (1,4,1)' refers to a specific tree, but Figure 1 is not described precisely enough to identify which tree is meant; a small clarifying phrase would help.","section":"Example 1"}],"recommendation":"major_revision","confidential_remarks":"The central claim as stated is false, but the intended statement about common vertices is almost certainly correct and the proof is essentially complete modulo the unstated facet property. I recommend major revision rather than rejection: the author should correct Theorem 2 and the abstract to refer to the common vertices, and then the paper would be a sound short note. If the author insists on keeping the literal equality Perm_n ∩ Asso_n = D, the paper would have to be rejected, since that equality is not fixable by local edits."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this note has a genuinely new observation, but the main theorem is stated incorrectly. The equality Perm_n ∩ Asso_n = D cannot hold for n ≥ 3 because both polytopes are convex, so their intersection is infinite. The proof actually establishes what the author probably meant: the set of vertices common to the permutohedron and Loday's associahedron is exactly the maximal never-middle Condorcet domain D. That corrected claim appears sound, and it is the interesting part of the paper.\n\nWhat is new: the domain itself is already known (Karpov–Slinko), as is Loday's realization. The contribution is the identification of this domain with the common vertices of two standard polytopes. That geometric link is not in the cited literature, and the inductive proof is short and persuasive. It also gives a nice pattern-avoidance reformulation: the common vertices are exactly the permutations avoiding 132 and 231.\n\nSoft spots: first, the overstatement in Theorem 2 is not a harmless wording issue. A reader comparing the statement to the proof will notice that the proof only talks about vertices of Asso_n, and never about non-vertex points in the intersection. The abstract's phrase \"common points\" is similarly ambiguous. The fix is easy: replace \"intersection\" with \"set of common vertices\" in the theorem, and note explicitly that the polytopes share more points along edges or faces. Second, the induction step silently uses the fact that if a point of Perm_n has coordinate n, deleting that coordinate gives a point of Perm_{n-1}. That is a standard facet property, but it is load-bearing and should be stated. Once both fixes are in, the argument goes through.\n\nI would not desk-reject this. The idea is correct and worth refereeing, provided the author is asked to fix the statement. The corrected version would be a nice short paper in combinatorics and social choice.","headline":"A correctable overstatement hides a genuinely new polytopal realization of a known Condorcet domain.","tokens_in":4673,"tokens_out":2499,"would_cite":true,"duration_ms":22732,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["05A05","52B11"],"pacs":[],"model":"deepseek-v4-flash","headline":"The common vertices of the permutohedron and Loday's associahedron are exactly the $2^{n-1}$ permutations of the maximal never-middle Condorcet domain.","keywords":["Condorcet domains","associahedron","permutohedron","Tamari lattice","pattern avoidance","never-middle condition","Loday realisation","voting theory"],"falsifier":"For some $n\\ge 5$, compute the Loday coordinates of all $2^{n-1}$ permutations obtained by iterating the star operation; if any resulting coordinate vector is not a vertex of $\\mathrm{Perm}_n$ (i.e., not a permutation of $1,\\dots,n$), the equality fails. Equivalently, search the common vertices of $\\mathrm{Asso}_n$ and $\\mathrm{Perm}_n$ for a permutation containing the pattern 132 or 231.","tokens_in":3672,"feed_emoji":"🗳️","tokens_out":11853,"duration_ms":98549,"temperature":0.7,"pith_summary":"This paper establishes a bridge between voting theory and discrete geometry. It proves that, for every $n$, the vertices shared by the permutohedron $\\mathrm{Perm}_n$ and Loday's associahedron $\\mathrm{Asso}_n$ are exactly the $2^{n-1}$ permutations obtained from the iterated star operation $((1\\star 2)\\star 3)\\cdots\\star n$. That set is the maximal never-middle Condorcet domain, meaning that for every triple $i<j<k$, the middle alternative $j$ never appears between $i$ and $k$ in any ranking. Equivalently, these are precisely the permutations of $\\{1,\\dots,n\\}$ avoiding the patterns 132 and 231.","feed_headline":"Polytope overlap equals a maximal Condorcet domain","feed_subtitle":"The shared vertices of the permutohedron and associahedron are exactly a maximal voting-theory domain.","key_machinery":"The machinery is Loday's coordinate assignment $M_n$ for full binary trees. To each internal vertex of a tree with leaves $0,1,\\dots,n$, the vertex between leaves $i-1$ and $i$ carries the weight $a_i b_i$, where $a_i$ and $b_i$ are the numbers of leaf descendants of its left and right children; the point $M_n(t)$ has these weights as coordinates. The associahedron is the convex hull of these $M_n(t)$ points. The proof's load-bearing inequality is the root-weight bound $(i+1)(n-i)\\le n$, which is forced by membership in the permutohedron; it leaves only $i=0$ or $i=n-1$, reducing any common vertex to a common vertex of $\\mathrm{Perm}_{n-1}$ and $\\mathrm{Asso}_{n-1}$ and matching the recursive star construction.","core_discovery":"The paper's central claim is that $\\mathrm{Perm}_n\\cap\\mathrm{Asso}_n = ((1\\star 2)\\star 3)\\cdots\\star n$, where $\\star$ is the doubled-concatenation operation $D_1\\star D_2=\\{u_1u_2,u_2u_1: u_i\\in D_i\\}$. The set on the right has size $2^{n-1}$, is maximal among Condorcet domains, and satisfies the never-middle condition. In pattern terms, these are exactly the permutations that avoid 132 and 231. The proof works by induction: a tree vertex of Loday's associahedron has root weight $(i+1)(n-i)$, and membership in the permutohedron forces this weight to be at most $n$, which forces $i=0$ or $i=n-1$, so the largest alternative sits at one of the two ends and the remaining coordinates form a common vertex in dimension $n-1$.","pith_inferences":["Beyond the paper, the proof's root-weight bound suggests the same intersection pattern may persist for other Loday-type realisations built from product coordinates, since the bound depends only on the shape of the root subtree rather than on the internal tree structure.","Beyond the paper, the common-vertex count $2^{n-1}$ makes explicit a binary-choice encoding of the domain: each new largest alternative is placed left or right, yielding a natural bijection between the domain and all subsets of $\\{2,\\dots,n\\}$, which the paper does not spell out.","Beyond the paper, the recursive structure raises the question, not addressed here, of whether the $2^{n-1}$ common vertices form the vertex set of a common subpolytope of $\\mathrm{Perm}_n$ and $\\mathrm{Asso}_n$, and if so, which polytope that is."],"forward_implications":["For each $n$, exactly $2^{n-1}$ of the vertices of Loday's associahedron are also vertices of the permutohedron, and they are precisely the permutations that avoid 132 and 231.","The maximal never-middle Condorcet domain therefore has a geometric realisation: it appears as the common vertex set of two classical polytopes, so voting-theoretic questions about the domain can be studied through polytope combinatorics.","Membership in the domain has a simple recursive test: generate the $2^{n-1}$ rankings by repeatedly placing the next largest alternative at either the left or the right end, or check pattern avoidance.","Because the domain is maximal, no additional ranking can be added to this common-vertex set without creating a cycle in pairwise majority voting.","The induction identifies the common vertices recursively: in any shared vertex, the largest alternative sits in first or last position, and removing it leaves another shared vertex in one dimension lower."],"supporting_citations":[{"why":"Provides the coordinate map $M_n$ and the theorem that these points give a polytopal realisation of the Tamari lattice.","marker":"Loday [2004]"},{"why":"Introduces the star operation $D_1\\star D_2$ used to define the domain recursively.","marker":"Danilov and Koshevoy [2013]"},{"why":"Supplies the known result that the iterated star domain is a maximal never-middle Condorcet domain of size $2^{n-1}$.","marker":"[Karpov and Slinko, 2023]"},{"why":"Supplies the definition and facet facts of the permutohedron used in the induction step.","marker":"Ziegler [2012]"}],"fun_headline_variants":["Associahedron and permutohedron meet in a maximal Condorcet domain","Common vertices of two polytopes form a maximal Condorcet domain","A polytope intersection that is a maximal Condorcet domain","Loday's associahedron hides a maximal Condorcet domain","Maximal Condorcet domain from polytope overlap"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The induction step assumes without proof that if a point of $\\mathrm{Perm}_n$ has a coordinate equal to $n$, then deleting that coordinate leaves a point of $\\mathrm{Perm}_{n-1}$ on the remaining coordinates; this facet property of the permutohedron is standard but load-bearing.","fun_headline_variants_meta":{"raw":{"variants":["Associahedron and permutohedron meet in a maximal Condorcet domain","Common vertices of two polytopes form a maximal Condorcet domain","A polytope intersection that is a maximal Condorcet domain","Loday's associahedron hides a maximal Condorcet domain","Maximal Condorcet domain from polytope overlap"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001288,"raw_usage":{"total_tokens":5177,"prompt_tokens":776,"completion_tokens":4401,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":392,"completion_tokens_details":{"reasoning_tokens":4309}},"tokens_in":392,"tokens_out":4401,"duration_ms":25784,"temperature":1.0,"reasoning_tokens":4309,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T16:01:12.613163+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"For some $n\\ge 5$, compute the Loday coordinates of all $2^{n-1}$ permutations obtained by iterating the star operation; if any resulting coordinate vector is not a vertex of $\\mathrm{Perm}_n$ (i.e., not a permutation of $1,\\dots,n$), the equality fails. Equivalently, search the common vertices of $\\mathrm{Asso}_n$ and $\\mathrm{Perm}_n$ for a permutation containing the pattern 132 or 231.","supporting_citations":[{"cited_title":"Realization of the S tasheff polytope","cited_arxiv_id":null,"evidence_quote":"Provides the coordinate map $M_n$ and the theorem that these points give a polytopal realisation of the Tamari lattice."},{"cited_title":"Symmetric maximal condorcet domains","cited_arxiv_id":null,"evidence_quote":"Supplies the known result that the iterated star domain is a maximal never-middle Condorcet domain of size $2^{n-1}$."},{"cited_title":"Lectures on polytopes, volume 152","cited_arxiv_id":null,"evidence_quote":"Supplies the definition and facet facts of the permutohedron used in the induction step."}],"review_version":1}