{"id":"eb90038b-a0ab-49a1-8de1-5dec85c7bf1b","arxiv_id":"2504.15636","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Periagroups that are infinite and not virtually direct products contain contracting elements in their standard Cayley graphs, implying acylindrical hyperbolicity and transcendental conjugacy growth series.","lead":"This paper proves that, apart from groups that split as products of two infinite factors, a large family of groups called periagroups (which includes Coxeter groups and graph products) always contains so-called contracting elements in their standard Cayley graphs. It uses this to show these groups are acylindrically hyperbolic and that their conjugacy growth series are transcendental.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Graph-product case of Theorem 1.1 is underproved: Theorem 5.5 gives only a sufficient element-level condition, so the claimed existence in cases like F2×Z2 does not follow; this also makes Corollary 1.3 false.","rationale":"The reader correctly flagged the contradiction in Corollary 1.3 and recommended a repair, but treated it as a corollary-level overclaim. My stress-test identifies the same phenomenon as a gap in the proof of the central theorem itself: the graph-product case of Theorem 1.1 is justified by Theorem 5.5, whose hypothesis is not necessary for the existence of a contracting element. The concrete example F2×Z2 shows the proof mechanism misses a whole family of cases (complete graphs with finite factors). This does not make me doubt the truth of Theorem 1.1; the result is consistent with known facts (F2×Z2 is virtually free and has contracting elements). Rather, it means the proof of a central case is incomplete and the stated Corollary 1.3 is false. A careful revision should replace the appeal to Theorem 5.5 with a recursive/direct-product argument or with Corollary 5.11. Since the main claim is plausible and the gaps are local and repairable, the reader's CONDITIONAL verdict remains appropriate; I do not recommend a change to REJECT or UNVERDICTED.","tokens_in":40073,"tokens_out":19125,"duration_ms":181147,"concrete_test":"Apply the proof of Theorem 6.30 case (ii) to Γ = complete graph on {u,v}, G_u=F2, G_v=Z2, with S = S_F2 ∪ {z}. Enumerate all essential supports (subsets of {u,v}); all are complete, so Theorem 5.5's hypothesis never holds. Then check explicitly that (a,1) with a∈F2 a hyperbolic element is contracting in Cay(F2×Z2, S). If it is contracting, the proof as written has a gap and Corollary 1.3 is contradicted. The fix should invoke Corollary 5.11's second case or a direct-product reduction.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Theorem 1.1 asserts that a graph product ΓG has a contracting element iff it is infinite and not virtually a product of two infinite groups. In the proof, the graph-product case is dismissed with 'If Ψ is empty, then our periagroup coincides with a graph product and Theorem 5.5 applies' (Section 6.4). But Theorem 5.5 is a sufficient condition for an individual element: an element with essential support neither complete nor contained in a large join is contracting. It is not an existence criterion. Consider Γ complete on two vertices with G_u=F2 and G_v=Z2. Then ΓG = F2×Z2 is infinite and not virtually a product of two infinite groups, so Theorem 1.1 promises a contracting element. However, every possible essential support is {u}, {v}, or {u,v}, all complete subgraphs, so the hypothesis of Theorem 5.5 fails for every element. Thus the proof of Theorem 6.30(ii) does not produce the promised contracting element; it needs an additional argument, e.g. the second case of Corollary 5.11 (a vertex whose link is complete and finite). This gap is not cosmetic: it is exactly what makes Corollary 1.3 false, since Corollary 1.3 excludes the complete-graph case outright even though F2×Z2 does contain a contracting element (any hyperbolic element of the F2 factor). The central theorem may still be true, but the written proof of a main case does not cover the finite-factor situation.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a general criterion for an isometry of a paraclique graph endowed with a coherent system of local metrics to be contracting (Theorem 4.3), then applies it to periagroups. For a finitely generated periagroup Π(Γ,λ,G) it claims Theorem 1.1: Π has a contracting element in its standard Cayley graph if and only if Π is infinite and not virtually a direct product of two infinite groups. Consequences are drawn for Coxeter groups, graph products, and Dyer groups, including acylindrical hyperbolicity classifications and transcendence of conjugacy growth series via [GY22, Cor. 1.8]. The proof passes through a mediangle-geometric description of periagroups and a semidirect decomposition attached to a GP-Cox decomposition.","tokens_in":40289,"tokens_out":9067,"duration_ms":82422,"significance":"If correct, the results are substantial: they unify and extend known contracting-element and acylindrical-hyperbolicity results for Coxeter groups and graph products, and they yield transcendental conjugacy growth series for the standard generating sets. The paraclique-graph framework and the coherent-system-of-metrics technique are interesting tools in their own right, and Claim 6.35, which recovers finite-generating-set word metrics from local clique metrics, is a clean and useful observation. The main caveat is that the graph-product case of the main theorem is underproved and one of the announced corollaries is false as stated.","major_comments":[{"comment":"The graph-product case is dismissed with the sentence 'If Ψ is empty, then our periagroup coincides with a graph product and Theorem 5.5 applies.' This is not a valid existence argument. Theorem 5.5 is a sufficient condition on an individual element: an element whose essential support is neither complete nor contained in a large join is contracting. It does not assert that such an element always exists. For example, take Γ=K2, G_u=F2 and G_v=Z2; then every essential support is either {u}, {v}, or {u,v}, all complete, so the hypothesis of Theorem 5.5 fails for every element, even though F2×Z2 is infinite, is not virtually a product of two infinite groups, and contains contracting elements. The proof of Theorem 6.30(ii) therefore does not establish the existence direction of Theorem 1.1 for complete graphs. This can be repaired by a separate direct-product argument or by invoking the second case of Corollary 5.11, but the written proof is incomplete.","section":"Section 6.4, proof of Theorem 6.30(ii)"},{"comment":"As stated, this corollary is false. For Γ=K2 with G_u=F2 and G_v=Z2, the right-hand side fails because Γ is complete, yet Cay(F2×Z2, S_u∪S_v) contains a contracting element, for instance a hyperbolic element of the F2 factor. The statement is also stronger than what Theorem 5.5 proves, since Theorem 5.5 is only a sufficient condition. The corollary needs to be reformulated so that complete graphs with at most one infinite factor are handled, in agreement with Theorem 1.1 and Corollary 5.11.","section":"Corollary 1.3"}],"minor_comments":[{"comment":"The phrase 'provided by By [CF10]' should read 'provided by [CF10]'.","section":"Section 5.1, proof of Theorem 5.1"},{"comment":"The word 'reprensented' should be 'represented'.","section":"Section 5.2, before Theorem 5.5"},{"comment":"The word 'aslo' should be 'also'.","section":"Example 6.9"},{"comment":"The word 'exaclty' should be 'exactly'.","section":"Section 6.2, proof of Proposition 6.18"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Mark, read the paper through. The main new thing is the paraclique graph framework and the local-metric criterion (Theorem 4.3), which gives a uniform way to transfer contracting isometries from the big Cayley graph (with all factor elements) to the standard finite-generating-set Cayley graph. That is a genuinely useful idea, and it pays off: Theorem 1.1 for periagroups, and the corollaries for Coxeter and Dyer groups, look right in their main thrust. The Coxeter case is proved carefully, with the Davis complex argument. The conjugacy growth applications are straightforward once you have the contracting elements, but they're cleanly packaged.\n\nNow the soft spot, and it is not tiny. Corollary 1.3 states an iff for graph products: contracting element exists iff Γ is not complete and does not split as a large join. That is false as stated. Take Γ complete on two vertices, vertex groups F2 and Z2. The graph product is F2×Z2, which is infinite and not virtually a product of two infinite groups, and it clearly has contracting elements (any hyperbolic element of F2). The theorem's own criterion would promise one, but Corollary 1.3 excludes all complete graphs. The stress-test note is right: the proof of the graph-product case (Theorem 5.5) is a sufficient condition for an element, not an existence criterion. In the complete-graph finite-factor situation, every essential support is complete, so the hypothesis of Theorem 5.5 fails. The gap is real, but I think it is repairable: the missing cases are exactly the ones where the graph product has a direct factor that is either finite or has its own contracting element, and those can be handled by the same product arguments used elsewhere in the paper. Still, as written, the statement of Corollary 1.3 needs correction and the proof of Theorem 6.30(ii) needs an additional case.\n\nOther concern: the paper leans heavily on structural results from the second author's earlier work (mediangle geometry of periagroups, the semidirect decomposition with the crossing graph). These are quoted, not proved, so the paper is not self-contained. That is normal for this kind of paper, but it matters because the step from mediangle geometry to finite-generating-set metrics is load-bearing. If any of those structural theorems have hidden assumptions that fail for periagroups with mixed labels, the whole thing would need rework. I didn't find a concrete problem, only a dependency worth flagging.\n\nBottom line: the main theorem is likely true and the framework is a real advance. The paper deserves serious peer review, but it needs revision before publication: fix the graph-product corollary, add the missing case in the proof, and ideally state explicitly which structural results are imported from where. I'd send it to a good journal and ask for those changes. If you work on contracting elements or acylindrical hyperbolicity, the paraclique machinery is worth having on your radar.","headline":"Genuinely new machinery for contracting elements in Cayley graphs, but the graph-product iff is overclaimed and the proof misses a case with finite direct factors.","tokens_in":40911,"tokens_out":2557,"would_cite":true,"duration_ms":23778,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["20F65","20F67","20F55"],"pacs":[],"model":"deepseek-v4-flash","headline":"Theorem 1.1: a finitely generated periagroup has a contracting element in its standard Cayley graph if and only if it is infinite and not virtually a product of two infinite groups.","keywords":["contracting elements","periagroups","Coxeter groups","graph products","Dyer groups","conjugacy growth","acylindrical hyperbolicity","mediangle graphs"],"falsifier":"Compute the conjugacy growth series of the right-angled Coxeter group whose defining graph is a 5-cycle, with respect to its standard generating set. This group is infinite, irreducible, and not virtually a product of two infinite groups, so the theorem predicts transcendental conjugacy growth; if the series turned out to be algebraic (in particular rational), then either Theorem 1.1 or its consequence Corollary 1.6 would be false.","tokens_in":39770,"feed_emoji":"🧮","tokens_out":8349,"duration_ms":70642,"temperature":0.7,"pith_summary":"This paper proves that a finitely generated periagroup—a family of groups interpolating between Coxeter groups, graph products, and Dyer groups—admits a contracting element in its standard Cayley graph exactly when it is infinite and does not virtually decompose as a product of two infinite groups. Contracting elements are isometries whose orbits behave like geodesics in negative curvature, even when the ambient group is not hyperbolic. The paper then applies a known bridge: an element contracting in a Cayley graph forces the conjugacy growth series to be transcendental. Hence the theorem yields transcendental conjugacy growth series for all such periagroups, and in particular for Coxeter groups, graph products, and Dyer groups, with respect to their standard generating sets.","feed_headline":"Contracting elements exist in every unsplit infinite periagroup","feed_subtitle":"One criterion yields transcendental conjugacy growth for Coxeter groups, graph products, and Dyer groups.","key_machinery":"The load-bearing mechanism is the paraclique graph, a clique-gated graph in which parallelism of cliques is transitive (so hyperplanes make sense), together with a coherent system of metrics: each clique is given its own metric, compatible with projections between parallel cliques, and these local metrics unfold into one global metric that recovers the Cayley metric from a finite generating set. The criterion Theorem 4.3 reduces 'element is contracting' to a checkable geometric condition—admit an axis and skewer two hyperplanes whose intervening transverse hyperplanes have bounded total thickness. For periagroups, the GP-Cox decomposition $\\Pi = \\Omega_J \\rtimes C(\\Psi)$ splits the group into a graph product part and a Coxeter part, and the crossing graph $\\Omega$ encodes which hyperplanes can be found with finite stabiliser intersection.","core_discovery":"The central claim, Theorem 1.1, is that for a finitely generated periagroup $\\Pi := \\Pi(\\Gamma,\\lambda,\\mathcal{G})$, once each factor $G$ is given a finite generating set $S_G$, the Cayley graph $\\mathrm{Cay}(\\Pi, \\bigcup S_G)$ contains a contracting element if and only if $\\Pi$ is infinite and not virtually a product of two infinite groups. The proof's engine is a new criterion (Theorem 4.3): in a paraclique graph equipped with a coherent, group-invariant system of metrics, an isometry that has an axis and skewers a pair of well-separated hyperplanes is contracting in the associated global metric. The paper verifies this criterion for periagroups by passing through the GP-Cox decomposition $\\Pi = \\Omega_J \\rtimes C(\\Psi)$ and using the structure of the crossing graph $\\Omega$. Corollary 1.6 records the consequence the paper is ultimately after: the conjugacy growth series of these groups, with respect to standard generating sets, are transcendental.","pith_inferences":["Beyond the paper, the same criterion should apply to any group acting on a mediangle or paraclique graph with a coherent metric system, for example hypercellular graphs or small-cancellation polygonal complexes, once an axis and a well-separated skewered pair are found.","The paper's hypothesis in Corollary 1.7 that one factor dominates the growth rates is likely removable; if so, every non-virtually-abelian periagroup would have transcendental conjugacy growth regardless of how growth rates compare across factors.","The GP-Cox decomposition suggests a testable dichotomy: for periagroups, failure of acylindrical hyperbolicity is controlled by infinite centralisers of the graph-product part inside the Coxeter part, so analogous obstructions may appear in other semidirect products with a graph-product normal subgroup."],"forward_implications":["A finitely generated Coxeter group has a contracting element in its standard Cayley graph exactly when it is a product of irreducible Coxeter groups all but one finite, and the remaining factor is non-affine or infinite dihedral (Corollary 1.2).","A graph product has a contracting element exactly when its defining graph is not complete and does not split as a large join (Corollary 1.3).","A Dyer group has a contracting element exactly when it is infinite and not virtually a product of two infinite groups (Corollary 1.4).","Every periagroup covered by Theorem 1.1 is acylindrically hyperbolic (Section 7), giving uniform acylindrical hyperbolicity criteria for the three families.","The conjugacy growth series of these groups with respect to standard generating sets are transcendental (Corollary 1.6), and Corollary 1.7 extends this to direct products when one factor has strictly larger growth rate."],"supporting_citations":[{"why":"Supplies the mediangle-graph structure of Cay(Π,∪G) and the semidirect decomposition Π = Ω_J ⋊ C(Ψ) on which the whole periagroup section rests.","marker":"[Gen22a]"},{"why":"Provides the theorem used at the end: a contracting element in a Cayley graph implies the conjugacy growth series is transcendental.","marker":"[GY22]"},{"why":"Characterises rank-one isometries in Davis complexes, which the paper converts into contracting elements in Coxeter standard Cayley graphs.","marker":"[CF10]"},{"why":"Defines quasi-median graphs, quasi-cubulations, and the basic lemmas on cliques and geodesics used for graph products and the quasi-median closure.","marker":"[Gen17]"},{"why":"Supplies the lemmas on right hyperplanes (gated carriers, rotative stabilisers) used in the periagroup argument.","marker":"[Gen24]"},{"why":"Gives the implication from contracting isometries to acylindrical hyperbolicity.","marker":"[BBF15]"},{"why":"Provides the analytic fact that sequences with asymptotics α^n/n have transcendental generating functions.","marker":"[Fla87]"},{"why":"Shows virtually abelian groups have rational conjugacy growth series, used to isolate the transcendental part in direct products.","marker":"[Eve19]"}],"fun_headline_variants":["Contracting elements found in all unsplit infinite periagroups","Transcendental conjugacy growth proven for periagroups","Acylindrical hyperbolicity and transcendental growth from one criterion","Coxeter groups and graph products: transcendental conjugacy growth"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the structural theorems imported from the second author's earlier work hold: the Cayley graph of a periagroup generated by all non-trivial factor elements is mediangle, and the GP-Cox decomposition yields the semidirect product Π = Ω_J ⋊ C(Ψ); if either fails, the passage from mediangle geometry to finite-generating-set Cayley metrics and the search for skewering hyperplanes collapses.","fun_headline_variants_meta":{"raw":{"variants":["Contracting elements found in all unsplit infinite periagroups","Transcendental conjugacy growth proven for periagroups","Acylindrical hyperbolicity and transcendental growth from one criterion","Coxeter groups and graph products: transcendental conjugacy growth"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001142,"raw_usage":{"total_tokens":4679,"prompt_tokens":827,"completion_tokens":3852,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":443,"completion_tokens_details":{"reasoning_tokens":3780}},"tokens_in":443,"tokens_out":3852,"duration_ms":26819,"temperature":1.0,"reasoning_tokens":3780,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T11:22:17.950359+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the conjugacy growth series of the right-angled Coxeter group whose defining graph is a 5-cycle, with respect to its standard generating set. This group is infinite, irreducible, and not virtually a product of two infinite groups, so the theorem predicts transcendental conjugacy growth; if the series turned out to be algebraic (in particular rational), then either Theorem 1.1 or its consequence Corollary 1.6 would be false.","supporting_citations":[],"review_version":1}