{"id":"beb513a0-e724-4563-8486-0f0e0e5a5b7d","arxiv_id":"2608.04520","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"This paper proves that, among irreducible affine Weyl groups of rank at least two, every antipodal triple of ideal chambers is generic exactly for the types C2, G2, and B3.","lead":"For triples of opposite chambers on the boundary of an affine building, this paper compares two notions of genericity and classifies when every such triple is automatically generic. It shows this happens only in Weyl types C2, G2, and B3, and constructs a locally constant barycenter map for the generic triples.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'only' direction of the classification is unproven: §6.2's construction of non-generic triples for all non-exceptional types uses an unverified bisector-projection assumption.","rationale":"The reader's weakest_assumption targets the angle assertion in Prop. 6.2, which underpins the positive direction. On closer reading, that assertion is correct: for a fixed line L with opposite endpoints ξ1 and ξ2, the rays from any two points on L toward ξ1 and ξ2 have opposite directions, so equality of the two Euclidean angles forces α = 90°. The proof's reliance on retractions preserving angles and parallelism is standard in building theory. Therefore I do not treat Prop. 6.2 as the main obstruction. The main obstruction is the converse/classification claim. The paper's own Step 3 in §6.2 concedes that C2 and C3 need not be opposite and adds a bisector-projection assumption with no proof. Since the abstract claims automatic ideal-genericity 'only' in the three types, this unproved existence assertion is load-bearing. The algebraic checks of Props. 6.5–6.11 are solid, and spot checks of the Weyl vectors and root-subspace containments confirm them. The conditional nature of §6.2 and §6.3 is honestly flagged by the authors, which supports a CONDITIONAL verdict but not ACCEPT as a complete classification. If the projection assumption turns out to hold in all non-exceptional cases, the concern would be resolved; the B4 check proposed above is a concrete way to test it.","tokens_in":25396,"tokens_out":25258,"duration_ms":215785,"concrete_test":"Work in the standard Coxeter complex of type B4 with F = R(e1 − e2 − e3 + e4), the one-dimensional root-flat given by Prop. 6.7. Decompose X(η+, η−) ≅ R × X_I, choose an ideal-generic triple (C1, C2, C3) in X∞_I, and construct the lifted chambers C1, C2 and the bisector γ12 of Step 1. Compute the ideal endpoints of pI(γ12) in X∞_I and compare them with the barycenters of C1 and C2. If pI(γ12) is not the bisector of (C1, C2), then the extra assumption in §6.2 fails in this case, and the 'only' direction of the abstract remains unsupported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The algebraic half of Theorem 6.13 (Props. 6.5–6.11) is checkable and appears correct, and the angle argument in Prop. 6.2, though under-explained, is valid once one notes that the rays toward the two opposite barycenters on a fixed bisector point in opposite directions, forcing α = 90°. The load-bearing gap is in the converse direction, which the abstract needs for 'only in types B2 = C2, G2, B3'. Section 6.2 sets up a construction of a non-affine-generic triple from a root-flat perpendicular to a bisector, but Step 3 explicitly requires that the bisector γ12 projects via pI to a bisector of X_I; this is neither proved nor checked for the families in Props. 6.6–6.11, and Remark 6.12 only cites eA2. The projection of a bisector perpendicular to the Euclidean factor need not be a bisector, because the projection of the ideal endpoint (the barycenter of a chamber in the spherical join S^{m-1} * X∞_I) is generally not the barycenter of the projected chamber. Also Step 2's assertion that A12 ∩ A13 is a convex cone parallel to Q1 is unjustified: in X_I the intersection of two apartments for an ideal-generic triple has nonempty bounded boundary, so the product with R^{n-|I|} is a cylinder over a bounded polyhedron, not merely a cone. Thus the existence of bad triples in all non-exceptional types is not established.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper defines two competing notions of genericity for triples of pairwise opposite ideal chambers in the boundary of a locally finite affine building: ideal-genericity, requiring the three apartments at infinity to have empty common intersection, and affine-genericity, requiring each intersection of two affine apartments to have a bounded boundary component. The main results are: Proposition 4.1, which shows that failure of affine-genericity propagates through the whole triple and yields a canonical root-flat; Proposition 5.2, which constructs a locally constant (hence continuous) barycenter map on affine-generic triples; Propositions 6.2–6.3, which reduce the automatic-genericity question to a root-system condition; and a case-by-case computation, Theorem 6.13, asserting that among irreducible finite Weyl groups of rank at least two the automatic-genericity condition holds exactly for types B2=C2, G2 and B3. The paper also proposes explicit constructions of non-generic triples in all other types and derives algebraic consequences for stabilizers of triples in Bruhat–Tits buildings.","tokens_in":25636,"tokens_out":9245,"duration_ms":81770,"significance":"The paper's ideas are attractive and the fully proved parts are useful. The root-system computations in Propositions 6.5–6.11 are explicit, checkable, and appear correct; I spot-checked the arithmetic in B2, B3, A_n, B_4 and E_6. The local constancy of the barycenter map in Proposition 5.2 is a genuinely interesting structural statement, and the algebraic consequences in Section 7 would be a natural application of the geometric results. However, the converse direction of Theorem 6.13 is not currently established: the construction of non-generic triples for all non-exceptional types rests on two unproved assumptions in Section 6.2. The theorem may well be true, but the manuscript as written does not yet prove it.","major_comments":[{"comment":"The construction of a non-affine-generic triple for every non-exceptional type depends on the condition, stated in Step 3, that the bisector γ12 in A12 projects under p_I to a bisector in A12. This condition is not proved, and Remark 6.12 only asserts it for eA2. The projection of a bisector is not automatically a bisector, because the ideal endpoint of a projected ray is generally not the barycenter of the projected chamber. Without a proof of the projection condition for the families in Propositions 6.6–6.11, the existence of bad triples in types A_n (n≥2), B_n (n≥4), C_n (n≥3), D_n (n≥3), F_4 and E_n is not established, and the 'only' direction of Theorem 6.13 remains open.","section":"6.2, Step 3 (pp. 25–26)"},{"comment":"The assertion that Q'_1 = A12 ∩ A13 is 'a convex cone parallel to Q1' is unjustified. In the splitting X(η+,η−) ≅ R^{n-|I|} × X_I, the apartments A12 and A13 are products of R^{n-|I|} with apartments of X_I; for an ideal-generic triple in X_I, the intersection of two apartments generally has nonempty bounded boundary, so Q'_1 is a cylinder over a bounded polyhedron rather than a cone. Since the construction of the geodesic ray whose endpoint defines C3 starts from this asserted cone geometry, this step needs a corrected proof or a different argument.","section":"6.2, Step 2 (p. 25)"},{"comment":"The key angular claim that the angle α at x1 'can coincide with the corresponding angle at x2 ... if and only if α = 90°' is stated without proof. This is the step that converts the failure of ideal-genericity into the existence of a root-flat perpendicular to a bisector, so it is load-bearing for the automatic-genericity half of Theorem 6.13. The claim is true — the rays r1 and r2 point in opposite directions along the same parallel class, while the two bisector rays are parallel with the same orientation — but the text should supply the short argument, because the retraction argument alone does not make the equality of the two angles explicit.","section":"Prop. 6.2 (p. 13)"}],"minor_comments":[{"comment":"The manuscript uses 'root-flat orthogonal/perpendicular to a bisector' for flats of arbitrary dimension without a formal definition; please state precisely that every direction of the flat is orthogonal to the bisector direction.","section":"Definitions 2.1 and 6.1"},{"comment":"The existence and properties of the unipotent element u are assumptions, not theorems; please state them as explicit hypotheses in the subsection's conclusion so that the conditional nature of the construction is preserved in any later citation.","section":"6.3 (p. 26)"},{"comment":"Defining the k-rank of the setwise stabilizer Stab_G{C1,C2,C3} as the k-rank of P1∩P2∩P3 is nonstandard; please state this convention at the start of Section 7 and add a remark explaining why the possible permutations of the three chambers do not change the rank conclusion.","section":"7, Remark 7.1"},{"comment":"Since Step 3 of Section 6.2 assumes that γ12 projects to a bisector, the eA2 example deserves an explicit verification rather than a bare assertion.","section":"Remark 6.12"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's the short version: the paper builds a genuinely useful framework for thinking about genericity of antipodal ideal-chamber triples in affine buildings, and the algebraic classification of finite Weyl groups it contains is solid. The problem is the claimed converse—that the only types where every triple is automatically generic are B2=C2, G2, B3—is not proven. The construction in Section 6.2 that is supposed to produce non-generic triples in all other types has a load-bearing gap. I would still send this to a serious referee, but not with a recommendation to accept as is.\n\nWhat is genuinely new: the distinction between ideal-genericity and affine-genericity, Proposition 4.1 showing that non-genericity propagates and yields a canonical root-flat, and Proposition 5.2 giving a locally constant barycenter map on affine-generic triples. The algebraic part is also a real contribution. Propositions 6.5 through 6.11 carry out an explicit, checkable analysis of which irreducible finite Weyl groups satisfy the root-space condition of Proposition 6.3. I spot-checked several of the arithmetic claims and they work. The later section connecting the rank of the setwise stabilizer of a triple to the dimension of the canonical root-flat (Theorem 7.7) is a nice payoff and is cleanly argued given the geometric setup.\n\nThe soft spots are real but concentrated in one place. Section 6.2, Step 2 asserts that the intersection Q'_1 = A12 ∩ A13 is a convex cone parallel to Q1. That is not justified. In the transverse building, the analogous intersection for an ideal-generic triple typically has nonempty bounded boundary, so the product with the Euclidean factor is a cylinder over a bounded polyhedron rather than a cone. This affects the construction of the third chamber. Step 3 then explicitly requires that the bisector γ12 projects to a bisector in X_I; the paper only indicates this happens for eA2 in Remark 6.12, and that projection property is exactly what is needed to ensure C2 and C3 are opposite. Without these steps, the counterexamples for all non-exceptional types do not go through, and the 'only' direction of Theorem 6.13 is unproven. Section 6.3 is even more explicitly conditional, relying on assumptions about unipotent subgroups that are not known to hold generally. The authors do acknowledge some of these limitations, which I appreciate, but the abstract and theorem statement overclaim.\n\nOne concern the reader flagged—the angle assertion in Proposition 6.2—looks repairable. The stress-test note is right that the rays toward the two opposite barycenters on a fixed bisector point in opposite directions, so the angle argument can be made rigorous. I would not treat that as a serious flaw.\n\nWho is this for: people working on boundaries of affine buildings, group actions on buildings, or C*-simplicity and its relatives. The framework and the positive results in Sections 4 and 5 will be used, regardless of what happens to the converse.\n\nRecommendation: engage with it, but send to a referee with explicit instructions to chase the projection assumption in Section 6.2. If that step cannot be proved, the correct conclusion is that the classification has only a proven 'if' direction, and the paper should say so. With that change, it is still a solid contribution.","headline":"Useful framework and checkable Weyl-group computations, but the converse half of the main theorem rests on a missing projection argument—worth refereeing, not worth believing yet.","tokens_in":26241,"tokens_out":2961,"would_cite":true,"duration_ms":28024,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["20E42","51E24","20F55"],"pacs":[],"model":"deepseek-v4-flash","headline":"In affine buildings of types C2, G2 and B3, every antipodal triple of ideal chambers is automatically generic, and all other irreducible types admit explicit non-generic configurations.","keywords":["affine buildings","spherical buildings at infinity","antipodal triples","ideal chambers","affine-genericity","root-flats","bisectors","Weyl groups"],"falsifier":"Search a type eB3 affine building, for instance the Bruhat–Tits building of split SO7 over Q_p, for an antipodal triple of ideal chambers whose three apartment boundaries have non-empty intersection; finding such a triple would directly refute Theorem 6.13. Equivalently, exhaust the finite W-orbits of proper root subspaces for the Weyl vector (5/2, 3/2, 1/2) in type B3: if this vector lies in any such orbit, the algebraic criterion of Proposition 6.3 fails for B3.","tokens_in":25041,"feed_emoji":"📐","tokens_out":8476,"duration_ms":76316,"temperature":0.7,"pith_summary":"This paper studies triples of pairwise opposite chambers in the spherical boundary at infinity of a locally finite affine building. It distinguishes ideal-genericity, meaning the three apartment boundaries have empty common intersection, from affine-genericity, a condition on bounded boundary components of pairwise intersections of affine apartments, and shows the affine notion supports a continuous barycenter map. The main classification result is that, among irreducible finite Weyl groups of rank at least two, the geometric criterion of Proposition 6.3 holds exactly for types G2, B2=C2, and B3. Consequently, in affine buildings of those types every antipodal triple of ideal chambers is automatically ideal-generic and affine-generic. In all remaining irreducible types the paper constructs explicit non-generic configurations, and in the Bruhat–Tits setting it proves that the k-rank of the stabilizer of an antipodal triple equals the dimension of its canonical root-flat.","feed_headline":"Automatic genericity for antipodal triples: only G2, C2, B3","feed_subtitle":"A root-system criterion classifies when every opposite triple of chambers at infinity is generic, yielding a continuous barycenter map.","key_machinery":"The mechanism is the equivalence of Proposition 6.3: the affine Coxeter complex contains a root-flat perpendicular to a bisector exactly when the Weyl vector $\\rho$, the half-sum of the positive roots, lies in the span of a proper subset of roots. A root-flat is an unbounded intersection of affine walls, and a bisector is a bi-infinite geodesic whose endpoints are the barycenters of two opposite ideal chambers. Since bisectors have directions $w\\rho$, perpendicularity of $\\bigcap_{\\alpha\\in T}H_\\alpha$ to a bisector is equivalent to $w\\rho\\in\\mathrm{Span}(T)$, and Lemma 6.4 reduces the check to $W$-orbits of the standard parabolic hyperplanes. That reduction is what turns the classification over all irreducible finite Weyl groups into a finite, algorithmic computation.","core_discovery":"The paper's central claim is that two genericity notions for triples of pairwise opposite chambers at infinity are governed by a single root-system dichotomy. Proposition 4.1 shows that failure of affine-genericity propagates across all three pairwise apartment intersections and produces a canonical root-flat, defined up to parallelism, lying in all three apartment boundaries. Proposition 5.2 shows that the barycenter of a finite convex hull built from the bounded boundary components is locally constant, hence continuous. Theorem 6.13 then classifies the situation: among irreducible finite Weyl groups of rank at least two, the sufficient condition 'no root-flat is perpendicular to a bisector' holds exactly for types G2, B2=C2 and B3, so in affine buildings of affine types G2, C2 and B3 every antipodal triple of ideal chambers is ideal-generic and affine-generic. In all remaining types the paper gives algorithmic constructions of non-generic triples, and for Bruhat–Tits buildings it proves that the k-rank of the setwise stabilizer of any antipodal triple equals the dimension of the canonical root-flat attached to the triple.","pith_inferences":["If the unproved angle assertion in Proposition 6.2 is supplied with a proof, the classification should transfer to any affine building with the same finite Weyl group, not only the classical Bruhat–Tits examples, because the criterion is purely combinatorial in the Coxeter complex.","The rank equality in Theorem 7.7 suggests a testable converse: in types G2, C2 and B3, stabilizers of antipodal triples are always compact, whereas in other types one should be able to exhibit stabilizers whose free abelian quotient has rank equal to the root-flat dimension.","The non-generic triples constructed in Section 6.2 are natural test cases for boundary dynamics of p-adic groups, since their stabilizers act with positive k-rank and a lattice of translations along the canonical root-flat.","A further extension the authors do not draw: the same root-system criterion could be used to classify automatic genericity for k-tuples of pairwise opposite chambers, not just triples, by asking whether all bisectors and root-flats satisfy the same perpendicularity obstruction."],"forward_implications":["In affine buildings of types eG2, eC2 and eB3, every antipodal triple of ideal chambers is ideal-generic, so any three pairwise opposite chambers at infinity determine three apartments whose boundaries meet transversely.","Affine-genericity in these types gives a well-defined, locally constant, and therefore continuous barycenter map on the space of antipodal triples, producing a geometric invariant attached to each triple.","For Bruhat–Tits buildings of type eC2, eG2 or eB3, the stabilizer of every antipodal triple of ideal chambers is compact, because the canonical root-flat has dimension zero.","In every other irreducible affine type of rank at least two, non-generic configurations exist, and the paper's construction yields explicit triples that are not affine-generic or even not ideal-generic.","For any antipodal triple in a Bruhat–Tits building, the k-rank of its setwise stabilizer equals the dimension of the canonical root-flat, with the stabilizer splitting as a compact group extended by a free abelian lattice of that rank."],"supporting_citations":[{"why":"Supplies the root-system realizations, numbering, and tables used to compute Weyl vectors and check proper root-subspace containments in Section 6.","marker":"[Bou02]"},{"why":"Provides the building-theoretic foundations, including convex hulls, gates, and chamber subcomplexes, used in the barycenter construction and in the gating argument of Section 6.3.","marker":"[AB08]"},{"why":"Gives the splitting of the sub-building $X(\\eta^+,\\eta^-)$ as a Euclidean factor times a transverse affine building, used throughout Sections 3, 4, 6, and 7.","marker":"[Rou11]"},{"why":"The authors' preceding work that motivates antipodal triples and supplies the existence of ideal-generic triples in the transverse building used in the non-generic constructions.","marker":"[CLB26]"},{"why":"Provides the circumcenter of a bounded convex hull in a non-positively curved space, used to define and prove local constancy of the barycenter map.","marker":"[BH99]"}],"fun_headline_variants":["Genericity for antipodal triples: only G2, C2, B3","Root systems decide when all opposite triples are generic","In affine buildings, only types G2, C2, B3 guarantee genericity","Only three affine Weyl types give universal antipodal genericity"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is the unproved angle assertion in Proposition 6.2: in the rotation argument, the Euclidean angle at one point between a ray in the shared root-flat and a bisector can coincide with the corresponding angle at another point only if that angle is 90 degrees; if this assertion is false, Theorem 6.13 would not establish automatic genericity for types C2, G2, and B3.","fun_headline_variants_meta":{"raw":{"variants":["Genericity for antipodal triples: only G2, C2, B3","Root systems decide when all opposite triples are generic","In affine buildings, only types G2, C2, B3 guarantee genericity","Only three affine Weyl types give universal antipodal genericity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000565,"raw_usage":{"total_tokens":2715,"prompt_tokens":1018,"completion_tokens":1697,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":634,"completion_tokens_details":{"reasoning_tokens":1618}},"tokens_in":634,"tokens_out":1697,"duration_ms":11190,"temperature":1.0,"reasoning_tokens":1618,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T14:38:14.951943+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Search a type eB3 affine building, for instance the Bruhat–Tits building of split SO7 over Q_p, for an antipodal triple of ideal chambers whose three apartment boundaries have non-empty intersection; finding such a triple would directly refute Theorem 6.13. Equivalently, exhaust the finite W-orbits of proper root subspaces for the Weyl vector (5/2, 3/2, 1/2) in type B3: if this vector lies in any such orbit, the algebraic criterion of Proposition 6.3 fails for B3.","supporting_citations":[],"review_version":2}