{"id":"a36ca423-817d-4507-9898-810474f77376","arxiv_id":"2508.20296","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Amenable groups of finite Assouad-Nagata dimension satisfy Shalom's property H_FD and therefore admit a virtual homomorphism onto Z, so they are neither simple nor torsion.","lead":"This paper proves that every finitely generated amenable group with finite Assouad-Nagata dimension satisfies Shalom's property H_FD, which forces a virtual homomorphism onto the integers. The result rules out simple and torsion examples in this class and is built on new diameter bounds for Følner couples.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The key step in Theorem 3.1 needs an unstated averaging argument to pass from the mean bound on the whole color to a single part with #B(A_alpha,n) <= (d+1)#A_alpha; the gap is repairable, so the conditional verdict stands.","rationale":"The reader's weakest-assumption analysis correctly identifies the inference in the paragraph after Lemma 3.2 as the point where the proof of Theorem 3.1 is compressed. The desired bound on an individual part is not a direct consequence of the lemma as stated; it needs the contrapositive using the finite decomposition into left-translation classes, which is present only inside the proof of Lemma 3.2 rather than in the application. This is a real gap in exposition but not a flaw in the underlying strategy: the missing argument is short and uses only ingredients already in the paper. The central claim of Theorem A is therefore plausible and likely correct, and the appropriate disposition remains conditional acceptance pending a repaired proof. Minor issues in Corollary 4.1 regarding total cautiousness do not affect the main theorem, since cautiousness on a subsequence suffices for H_FD. No fatal objection to the main theorem was found.","tokens_in":12157,"tokens_out":12306,"duration_ms":111902,"concrete_test":"Write out the missing contradiction argument in the paragraph after Lemma 3.2, using the equivalence-class decomposition from the proof of Lemma 3.2: assume every ratio #B(A_alpha,n)/#A_alpha exceeds d+1, group the alpha by left-translation type, apply the identities mu(union in a class of B_alpha) = (b_i/a_i) mu(union in the class of A_alpha), and sum over the finitely many classes to reach mu(union of all B_alpha) > (d+1) mu(union of all A_alpha) >= 1, contradicting mu(G) = 1. If this argument cannot be completed as stated, Theorem 3.1 fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"In the proof of Theorem 3.1, Lemma 3.2 is applied to the parts A_alpha of the chosen color and their n-neighborhoods B_alpha = B(A_alpha,n). As stated, the lemma requires a fixed lambda > 0 such that #B_alpha >= lambda #A_alpha for every alpha, and then bounds the mean of the union of the A_alpha by 1/lambda. In the application, no such lambda is known in advance, so the conclusion that some alpha satisfies #B_alpha <= (d+1)#A_alpha does not follow from the lemma statement alone. The conclusion can be derived from the proof of Lemma 3.2: if every alpha had #B_alpha > (d+1)#A_alpha, then in each of the finitely many left-translation classes C_i one would have mu(union over C_i of B_alpha) = (b_i/a_i) mu(union over C_i of A_alpha) > (d+1) mu(union over C_i of A_alpha); summing over the finitely many classes gives mu(union of all B_alpha) > (d+1) mu(union of all A_alpha) >= 1, contradicting mu(union of all B_alpha) <= 1. This weighted-averaging argument is not supplied in the text, so the proof as written is incomplete. The repair is short and does not require uniformly bounded cardinalities of the parts. A separate small issue: the text says that putting r_i = i in Lemma 4.5 of [12] gives total cautiousness, but that only yields a subsequence; total cautiousness is stronger than needed for H_FD, so this does not affect the main theorem.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves Theorem A: every finitely generated amenable group of finite Assouad-Nagata (AN) dimension satisfies Shalom's property H_FD. Since infinite amenable groups with H_FD admit a virtual homomorphism onto Z, the theorem implies that no infinite finitely generated amenable group of finite AN-dimension can be simple or torsion. The proof is based on new diameter estimates for Følner couples: Theorem 3.1 shows that a finitely generated amenable group of finite asymptotic dimension with control function f admits a sequence of Følner couples (F'_n,F_n) with C = d+1 and diam(F_n) ≤ f(2n)+2n. The paper also derives several smallness conditions for groups of finite AN-dimension, including diffusive random walks, large return probabilities, and total cautiousness, and it concludes with open questions about elementary amenability and dimensions of groups of intermediate growth.","tokens_in":12527,"tokens_out":4698,"duration_ms":40616,"significance":"If the proof is correct, this is a significant contribution to the study of Shalom's property H_FD and to the geometric theory of amenable groups. The theorem gives a clean geometric sufficient condition for H_FD, and the corollaries (no simple groups, no torsion groups among infinite amenable groups of finite AN-dimension) address natural questions that were previously open. The paper also strengthens Nowak's result on diameters of Følner sets to Følner couples, which is a useful technical advance. The exposition is clear, and the argument draws on published results of Erschler-Ozawa and Erschler-Zheng, which are appropriately cited. The main technical tool, Lemma 3.2, is an elegant mass-transport argument that is likely to be of independent interest.","major_comments":[{"comment":"The application of Lemma 3.2 to obtain a part A_alpha with #B(A_alpha,n) ≤ (d+1)#A_alpha is not justified as written. Lemma 3.2 assumes a fixed lambda > 0 such that #B_alpha ≥ lambda #A_alpha for every alpha, and its conclusion is an upper bound on the mean of the union of the A_alpha, not a bound on the ratio for an individual part. In the proof of Theorem 3.1, no such uniform lambda is known in advance, so the stated conclusion does not follow directly from the lemma. This gap is load-bearing because it is exactly the step that produces the Følner couple. The gap is repairable: if the inequality #B_alpha ≤ (d+1)#A_alpha failed for every alpha, then in each of the finitely many translation classes C_i one would have #B_alpha > (d+1)#A_alpha; using the equalities M_i^B = (b_i/a_i) M_i^A from the proof of Lemma 3.2, one obtains mu(union B_alpha) > (d+1) mu(union A_alpha) ≥ 1, contradicting mu(union B_alpha) ≤ 1. This weighted-averaging argument should be included in the proof.","section":"Section 3, proof of Theorem 3.1, after Lemma 3.2"}],"minor_comments":[{"comment":"The sentence 'Putting r_i = i we can conclude the total cautiousness' is not fully justified as written, because Lemma 4.5 of [12] as stated gives cautiousness along a subsequence. The authors note that the proof of that lemma yields total cautiousness when the l2-profile estimates hold for all r, but this should be spelled out more explicitly, for example by stating the uniform constant delta(c,C) in the displayed probability bound. Since cautiousness (without 'total') is already sufficient for H_FD via [11], this issue does not affect Theorem A, but the presentation should be corrected.","section":"Section 4, proof of Corollary 4.1(2)"},{"comment":"In the last displayed estimate of the proof, the notation 'mu( 8ğ i=1 A_i)' is a typo; the union should be over i=1 to m, matching the preceding sums. Please correct this.","section":"Section 3, proof of Lemma 3.2"},{"comment":"There are a few typographical errors, such as 'reperesentations' in Section 2.2 and 'Dranishikov' in Section 2.1 (should be 'Dranishnikov'). These do not affect the mathematics.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The paper is a strong contribution and the main theorem is likely correct, but the proof of Theorem 3.1 as written contains a genuine gap in the application of Lemma 3.2. The repair is short and uses only the internal equalities of Lemma 3.2, so I expect the authors can fix it without changing the framework. The central claim is defensible, and the issues in Section 4 are secondary because cautiousness, not total cautiousness, is what is needed for H_FD. I recommend major revision to require the missing argument to be supplied explicitly."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper proves a genuinely new structural fact: finitely generated amenable groups of finite Assouad-Nagata dimension satisfy Shalom's H_FD, so they virtually map onto Z and cannot be infinite simple or torsion. The route through controlled Følner couples is the real contribution. Theorem 3.1 extends Nowak's diameter bound from Følner sets to Følner couples, and the invariant-mean machinery is a new proof technique, not a repackaging of known arguments. If Theorem 3.1 holds, the corollaries follow from published Erschler-Ozawa and Erschler-Zheng results; those citations are not circular.\n\nThe soft spot is exactly where the reader put it. In the paragraph after Lemma 3.2, the proof applies the lemma to a chosen color with mean at least 1/(d+1) and concludes some part A_alpha has #B(A_alpha,n) ≤ (d+1)#A_alpha. The lemma as stated bounds the mean of the union under a uniform lambda assumption on every part; it does not license that individual conclusion. The stress-test repair is correct: if every part had #B_alpha > (d+1)#A_alpha, summing the right-invariant means over the finitely many translation classes would force mu(union B_alpha) > (d+1)mu(union A_alpha) ≥ 1, contradicting mu(union B_alpha) ≤ 1. That argument is short but must be written into the proof. Until then the proof of Theorem 3.1 is incomplete, though I expect the gap can be repaired.\n\nA smaller editorial issue: the text says taking r_i = i in Lemma 4.5 of [12] gives total cautiousness, but that lemma only produces a subsequence. Total cautiousness is stronger than needed for H_FD, so this does not affect Theorem A.\n\nOverall the paper is clearly written and the main claim is significant enough to justify a serious referee. The gap is localized and repairable. I would send it to peer review and ask for the averaging argument to be supplied. It deserves a reading group slot.","headline":"Følner-couple diameter theorem is new and significant, but one repairable gap in the application of Lemma 3.2 blocks Theorem 3.1 as written.","tokens_in":13064,"tokens_out":2598,"would_cite":true,"duration_ms":24295,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["05C12","20F65","20F67","20F69","20F18"],"pacs":[],"model":"deepseek-v4-flash","headline":"Finitely generated amenable groups of finite Assouad–Nagata dimension satisfy Shalom's property $H_{FD}$, so they virtually map onto $\\mathbb{Z}$ and cannot be simple or torsion.","keywords":["Assouad-Nagata dimension","asymptotic dimension","amenable groups","Følner couples","Shalom's property H_FD","cautious random walks","simple groups","torsion groups"],"falsifier":"Find an infinite finitely generated amenable group of finite AN-dimension that has no finite-index subgroup surjecting onto $\\mathbb{Z}$; by Theorem A none exists, so such a group would refute the paper's main claim. The simplest candidate class to check is torsion groups, which Corollary 1.3 explicitly excludes.","tokens_in":11948,"feed_emoji":"📐","tokens_out":12907,"duration_ms":97989,"temperature":0.7,"pith_summary":"This paper proves that every finitely generated amenable group with finite Assouad–Nagata dimension satisfies Shalom's property $H_{FD}$, a statement about which unitary representations have nonzero reduced first cohomology. A known consequence of $H_{FD}$ is that such a group has a finite-index subgroup that maps onto the integers. That in turn rules out two long-suspected possibilities: an infinite amenable group of finite AN-dimension cannot be simple, and it cannot be a torsion group. The route to the theorem is geometric rather than algebraic: the authors show that finite asymptotic dimension forces the existence of Følner couples inside balls of controlled diameter, and controlled Følner couples are a known sufficient condition for $H_{FD}$. The proof is the first step in a planned series relating diameters of Følner sets to dimension invariants of groups.","feed_headline":"Finite-dimension amenable groups virtually map onto Z","feed_subtitle":"New bounds on Følner couples prove Shalom's H_FD, ruling out simple and torsion groups.","key_machinery":"The load-bearing object is the Følner couple, a nested pair of finite sets $(F',F)$ with $F' \\subset F$, $\\#F \\le C\\#F'$, and distance from $F'$ to the complement of $F$ at least $n$, so that the inner set is deep inside the outer one. The new contribution is a construction of such couples in any finitely generated amenable group of finite asymptotic dimension: a right-invariant mean is used with a colored partition coming from the definition of asymptotic dimension, and Lemma 3.2 --- a mass-transport-style averaging argument over translation-equivalence classes --- shows some color class contains a part whose $n$-neighborhood has cardinality at most $(d+1)$ times the part itself. Taking that part as $F'_n$ and its $n$-neighborhood as $F_n$ yields a Følner couple of diameter $f(2n)+2n$. In the AN-dimension case $f$ is linear, which triggers a previously known chain: controlled Følner couples give small $l^2$-profile inside balls, small profile gives total cautiousness of simple random walks, and total cautiousness is a known sufficient condition for Shalom's $H_{FD}$.","core_discovery":"The central claim, Theorem A, is that a finitely generated amenable group of finite Assouad–Nagata dimension has Shalom's property $H_{FD}$: every unitary representation with non-zero reduced first cohomology admits a finite-dimensional subrepresentation. Since Shalom's theorem says an infinite finitely generated amenable group with $H_{FD}$ has a finite-index subgroup surjecting onto $\\mathbb{Z}$, the paper derives that such groups cannot be simple and cannot be torsion. The engine behind the theorem is a new estimate on Følner couples: for a finitely generated amenable group of finite asymptotic dimension $d$ with control function $f$, there exist nested finite sets $F'_n \\subset F_n$ with $\\#F_n \\le (d+1)\\#F'_n$ and distance from $F'_n$ to the complement of $F_n$ at least $n$, inside a ball of radius at most $f(2n)+2n$. When AN-dimension is finite the control function is linear, so the couples are \\emph{controlled} --- they live in balls of linear radius. The authors then connect controlled Følner couples to small $l^2$-profile, total cautiousness of simple random walks, and finally to $H_{FD}$ through previously established criteria.","pith_inferences":["The Følner-couple estimate of Theorem 3.1 needs only finite asymptotic dimension, not the stronger AN-dimension; whether sublinear control functions still yield useful cohomological consequences for amenable groups is a natural extension the paper does not pursue.","A concrete stress-test of the result would be to compute explicit Følner couples in a solvable group of finite AN-dimension, such as a solvable Baumslag–Solitar group, and compare the constant $d+1$ with the isoperimetric ratio; sharpness of this constant is not addressed.","If the paper's Question A has a positive answer, Theorem A would become one step toward a dimension-based proof that no amenable group of intermediate growth has finite AN-dimension, since such groups are not elementary amenable."],"forward_implications":["Every infinite finitely generated amenable group of finite AN-dimension has a finite-index subgroup surjecting onto $\\mathbb{Z}$, so it cannot be a counterexample to the virtual-Hopfian intuition for such groups.","There are no infinite simple groups among finitely generated amenable groups of finite AN-dimension.","There are no infinite torsion groups among finitely generated amenable groups of finite AN-dimension.","Any finitely generated amenable group of finite asymptotic dimension with control function $f$ admits Følner couples of diameter at most $f(2n)+2n$, extending Nowak's result from Følner sets to Følner couples.","The smallness conditions of Corollary 4.1 all hold: small $l^p$-profile inside balls, totally cautious simple random walks, return probability at least $\\exp(-C n^{1/3})$, and diffusive drift."],"supporting_citations":[{"why":"Defines $H_{FD}$ and proves that an infinite finitely generated amenable group with it virtually maps onto $\\mathbb{Z}$.","marker":"[24]"},{"why":"Shows cautiousness of simple random walks is a sufficient condition for Shalom's $H_{FD}$.","marker":"[11]"},{"why":"Bridges controlled Følner couples to small $l^2$-profile and total cautiousness, via Lemma 4.5.","marker":"[12]"},{"why":"Introduces controlled Følner pairs and the $l^p$-profile bound inside balls used in Corollary 4.1.","marker":"[29]"},{"why":"Introduced Følner couples and proved the return-probability lower bound from them.","marker":"[9]"},{"why":"Prior diameter bound for Følner sets in finite AN-dimension groups, which Theorem 3.1 strengthens.","marker":"[28]"}],"fun_headline_variants":["Finite AN-dimension amenability forbids simple and torsion groups","Shalom's H_FD holds for amenable groups of finite AN-dimension","Controlled Følner couples prove finite AN-dimension amenability","Amenable groups of finite AN-dimension virtually surject onto Z","No simple, no torsion: finite-dimension amenable groups"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof of Theorem 3.1 relies on the step that a color class whose parts have total mean at least $1/(d+1)$ must contain a single part $A$ with $\\#B(A,n) \\le (d+1)\\#A$; this follows from Lemma 3.2 only by a contrapositive argument that also uses the uniform finiteness of the parts, and the paper does not spell that argument out.","fun_headline_variants_meta":{"raw":{"variants":["Finite AN-dimension amenability forbids simple and torsion groups","Shalom's H_FD holds for amenable groups of finite AN-dimension","Controlled Følner couples prove finite AN-dimension amenability","Amenable groups of finite AN-dimension virtually surject onto Z","No simple, no torsion: finite-dimension amenable groups"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000455,"raw_usage":{"total_tokens":2287,"prompt_tokens":951,"completion_tokens":1336,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":567,"completion_tokens_details":{"reasoning_tokens":1244}},"tokens_in":567,"tokens_out":1336,"duration_ms":11728,"temperature":1.0,"reasoning_tokens":1244,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T16:47:38.153491+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Find an infinite finitely generated amenable group of finite AN-dimension that has no finite-index subgroup surjecting onto $\\mathbb{Z}$; by Theorem A none exists, so such a group would refute the paper's main claim. The simplest candidate class to check is torsion groups, which Corollary 1.3 explicitly excludes.","supporting_citations":[{"cited_title":"Erschler and N","cited_arxiv_id":null,"evidence_quote":"Shows cautiousness of simple random walks is a sufficient condition for Shalom's $H_{FD}$."},{"cited_title":"Erschler and T","cited_arxiv_id":null,"evidence_quote":"Bridges controlled Følner couples to small $l^2$-profile and total cautiousness, via Lemma 4.5."},{"cited_title":"Tessera, Asymptotic isoperimetry on groups and uniform embeddings into Banach spaces, Comment","cited_arxiv_id":null,"evidence_quote":"Introduces controlled Følner pairs and the $l^p$-profile bound inside balls used in Corollary 4.1."},{"cited_title":"Coulhon, A","cited_arxiv_id":null,"evidence_quote":"Introduced Følner couples and proved the return-probability lower bound from them."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Prior diameter bound for Følner sets in finite AN-dimension groups, which Theorem 3.1 strengthens."}],"review_version":1}