{"id":"35cd4127-9e7d-45df-8d97-0cd0543b9a4f","arxiv_id":"2608.12783","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The paper proposes a delta-invariant encoding of element orders, class sizes, and character degrees that unifies characterization results for finite simple and related groups.","lead":"This note introduces a uniform delta-invariant formalism for the arithmetic data that finite group theorists use to recognize simple groups: element orders, conjugacy class sizes, and character degrees. It reformulates known theorems and conjectures in this language and states several new open problems.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Section 2's displayed identity is false for cs* and cd*, so the proof of Theorem 2.5 as written is invalid; the intended reduction is repairable with weighted sums, which downgrades the concern from fatal to a correctable proof gap.","rationale":"The reader's weakest assumption identifies the same false identity in Section 2, and I agree that it is a genuine mathematical error. However, the error is localized: the tuple δ^{inv*}(G) includes both the divisors d_i and the multiplicities, so the order can be recovered by the correct weighted sums d_i δ_i for class sizes and d_i^2 δ_i for character degrees. Thus the central claim of Theorem 2.5 remains salvageable for cs* and cd* despite the printed proof being invalid. The eo* case is unaffected, since Σ_i δ^{eo*}_i = |G| is correct. The additional reliance on the self-cited preprint [14] for the cs* branch supports a conditional verdict rather than outright rejection. My recommendation therefore does not move the reader's CONDITIONAL verdict: the paper needs a corrected identity and ideally independent confirmation of [14], but the framework is not shown to be false.","tokens_in":6989,"tokens_out":18380,"duration_ms":189522,"concrete_test":"Recompute the disputed sums for G = A5. The conjugacy-class multiplicities are {1:1, 12:2, 15:1, 20:1}, so Σ_i δ^{cs*}_i = 5, while |A5| = 60; the character-degree multiplicities are {1:1, 3:2, 4:1, 5:1}, so Σ_i (δ^{cd*}_i)^2 = 5, not 60. Then recompute |G| via the corrected formulas Σ_i d_i δ^{cs*}_i and Σ_i d_i^2 δ^{cd*}_i; for A5 (and, say, PSL(2,7) and A6) these should both give |G|. If the corrected formulas recover |G| generally, the printed identity is a repairable typo; if any simple group fails the corrected formulas, Theorem 2.5 is false as stated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central step in Section 2 is the claim that n(G) = Σ_i δ^{cs*}_i(G) = Σ_i (δ^{cd*}_i(G))^2, which is used to assert that the tuple δ^{inv*}(G) determines (n(G), δ^{inv}(G)). This is false as written. By the definitions in the same section, δ^{cs*}_i is the number of conjugacy classes of size d_i, so Σ_i δ^{cs*}_i = k(G), the number of conjugacy classes, not |G|. Likewise δ^{cd*}_i is the number of irreducible characters of degree d_i, and Σ_i (δ^{cd*}_i)^2 is a sum of squared multiplicities, not |G|. The identities that actually hold are the class equation Σ_i d_i δ^{cs*}_i = |G| and the character orthogonality relation Σ_i d_i^2 δ^{cd*}_i = |G|. Because the tuples carry the labels d_i, the conclusion that δ^{inv*} determines n is nevertheless recoverable for cs* and cd* if these corrected formulas are used. As printed, however, the proof of Theorem 2.5 is invalid, and for δ^{cs*} this transfer is the stated route to the theorem. The concern is limited by the fact that a one-line correction restores the reduction; it is a proof defect, not a demonstrated counterexample. Additionally, the cs* case depends on Theorem 2.3, which in turn relies on the self-cited preprint [14]; hence the cs* branch of Theorem 2.5 should be regarded as conditional on unpublished work until [14] is independently verified.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a uniform formalism for arithmetic invariants of finite groups: for each invariant family inv in {eo, cs, cd}, the tuple δinv(G) records, for each divisor of |G|, whether that divisor occurs in the corresponding set, while the starred tuple δinv*(G) records the multiplicity of each divisor in the corresponding multiset. The paper's central assertions are that, for simple groups and for symmetric groups, the starred tuples δinv*(G) form full systems of invariants (Theorems 2.5 and 3.4), obtained by reducing starred data to the order n(G) together with the unstarred tuple. It also restates Shi's conjecture in this language, reviews the status of Thompson's and Huppert's conjectures, discusses related groups, and poses several open problems, including a conjecture about almost quasisimple groups.","tokens_in":7345,"tokens_out":6994,"duration_ms":68317,"significance":"If the main claims are correct, the paper provides a genuinely useful unifying perspective: a single framework that packages element-order, conjugacy-class-size, and character-degree invariants with and without multiplicities, and that makes precise the sense in which multiplicity data is stronger than set data. The identification of Shi's conjecture as the fullness of (n(G), δeo(G)) and the intended transfer principle from δinv*(G) to (n(G), δinv(G)) are valuable organizational ideas. The survey of known results and open problems is also useful. The paper's strengths are the clarity of the proposed notation and the fact that the main reduction, once corrected as described below, is elementary and directly checkable. The present proof of Theorem 2.5, however, contains an incorrect identity, so the central claim is currently not established as written.","major_comments":[{"comment":"The displayed identity n(G) = Σ_i δcs*_i(G) = Σ_i (δcd*_i(G))^2 is false. By the definitions in the same section, Σ_i δcs*_i(G) is the number of conjugacy classes of G, not the order |G|, and Σ_i (δcd*_i(G))^2 is a sum of squared multiplicities of character degrees, not generally equal to |G|. The correct identities are n(G) = Σ_i d_i δcs*_i(G) and n(G) = Σ_i d_i^2 δcd*_i(G), where d_i runs over the divisors of n(G). Because the tuples are indexed by the labelled divisors d_i, the conclusion that δinv*(G) determines (n(G), δinv(G)) is still recoverable for inv* = cs* and cd* if these weighted formulas are used. As printed, however, the proof of Theorem 2.5 is invalid and must be rewritten; the identity for eo* is correct.","section":"Section 2, displayed identity after 'Since'"},{"comment":"The cs* results depend on the author's own preprint [14], including [14, Lemma 2.4], but the needed statement and proof are not reproduced in this note. Consequently, Theorem 2.3 and the cs* assertion of Theorem 2.5 are conditional on [14] being correct and publicly available. The paper should either state and prove the necessary lemma from [14] or explicitly flag these results as conditional on that preprint.","section":"Theorems 2.2, 2.3 and the cs* branch of Theorem 2.5"},{"comment":"The sentence 'it follows from [28,30,31] that δcd*(G) is [full]' is not accompanied by a precise statement of what those papers prove. Since Huppert's conjecture is open for generic classical groups, and since the passage from 'determined by character degrees' to 'fullness of δcd*' is not automatic without a statement about multiplicities, the reader cannot verify the cd* branch of Theorem 2.5 from the text. Please state the exact cited theorems and explain how they imply fullness of the starred tuple in the class of all finite groups.","section":"Theorem 2.5, cd* branch"}],"minor_comments":[{"comment":"The tuple δinv*(G) is defined as a vector of length τ(n), so its length depends on the group. To make it a well-defined invariant on the class of all finite groups, the paper should either define it as a function on all positive divisors (with value 0 outside D(n)) or explicitly fix a common index set; otherwise equality of tuples of different lengths is not formally defined.","section":"Definition of δinv*(G)"},{"comment":"In Problem 3.9(i), 'sovbale' should be 'solvable'.","section":"Problem 3.9"},{"comment":"The phrase 'Thompsons's conjecture' should be 'Thompson's conjecture', and the title of [14] misspells 'conjugacy' as 'cojugacy'.","section":"References [10], [11], and [14]"},{"comment":"The example with the two maximal subgroups of M23 is asserted via [33, Section 4.3] but is not made self-contained; stating the common multiset of element orders would strengthen the illustrative point.","section":"Remark after Conjecture 3.8"}],"recommendation":"major_revision","confidential_remarks":"The paper is best read as a survey/position note whose main new tool is the δ-formalism. The central proof defect in Section 2 is easily repairable with weighted sums, and I do not see a counterexample to the claimed theorems. However, the heavy reliance on the author's own unpublished preprint [14] for a load-bearing branch should be checked with the editor, especially whether [14] is publicly available and whether its results are independently verified. If the authors repair the identity and clarify the external dependencies, the note could be suitable for publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a useful packaging of known results into a single notation, and the open problems are worth having on record. But the key displayed identity in Section 2 is wrong, so Theorem 2.5 is not proved as written. The gap is repairable; this is a correction, not a death sentence.\n\nWhat's new: the δ-invariant formalism, Conjecture 3.8, and a set of open problems (2.4, 3.1, 3.3, 3.6, 3.7, 3.9). The reformulation of Shi's conjecture is faithful, and the survey of Thompson and Huppert results is broad and looks fair. The citations are mostly standard; the author's own preprint [14] is used for Theorems 2.2 and 2.3, which makes those parts conditional on unpublished work, but self-citation is defensible here because the results are relevant.\n\nSoft spot: the text says n(G) = Σ δ^{cs*}_i = Σ (δ^{cd*}_i)^2. The first sum is the number of conjugacy classes, not the order; the second is a sum of squared multiplicities, not the sum of squares of degrees. The correct identities are n = Σ d_i δ^{cs*}_i and n = Σ d_i^2 δ^{cd*}_i. With those, and with the tuple understood as labeled by the divisors, the reduction from the starred tuple to (n, δ) works. Without labels, a bare tuple of cs* or cd* multiplicities does not determine n. So the conclusion is likely right, but the proof as printed is invalid.\n\nThe eo* branch is fine, because Σ δ^{eo*}_i = n genuinely. The cd* branch can also bypass the identity entirely, since [28,30,31] already prove δ^{cd*} is full for simple groups. It's mainly the cs* branch that depends on both the bad identity and the unpublished [14].\n\nWho this is for: anyone working on arithmetic characterizations of simple groups who wants a uniform language and a list of what is open. It deserves a serious referee, but the referee should require a corrected Section 2 before acceptance. I wouldn't cite it until that's done.","headline":"Useful packaging of known results into one notation, but the central identity in Section 2 is false and Theorem 2.5 is unproven as written; a one-line fix should repair it.","tokens_in":7868,"tokens_out":6240,"would_cite":false,"duration_ms":60753,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["20D06","20D60","20C15","20E45"],"pacs":[],"model":"deepseek-v4-flash","headline":"Multiplicity-counted tuples of element orders, class sizes, and character degrees are claimed to form full invariants for finite simple groups and symmetric groups.","keywords":["finite simple groups","element orders","conjugacy class sizes","character degrees","full system of invariants","symmetric groups","quasisimple groups","Shi conjecture"],"falsifier":"Direct computation settles the claimed bridge: for the symmetric group $S_3$, $\\delta^{\\mathrm{cs}^*}$ has one class of each size $1$, $2$, and $3$, so $\\sum_i \\delta^{\\mathrm{cs}^*}_i = 3$ while $|S_3| = 6$; for the cyclic group $C_2$, $\\delta^{\\mathrm{cd}^*}$ has two entries equal to $1$, so $\\sum_i (\\delta^{\\mathrm{cd}^*}_i)^2 = 4$ while $|C_2| = 2$. Since $C_2$ is a simple group, the displayed identity in Section 2 cannot be used to derive Theorem 2.5, and the fullness assertion would need a different proof to stand.","tokens_in":6800,"feed_emoji":"🔢","tokens_out":11417,"duration_ms":102806,"temperature":0.7,"pith_summary":"This note proposes a uniform arithmetic language for the three classical datasets attached to a finite group: element orders, conjugacy class sizes, and irreducible character degrees, each considered with or without multiplicities and indexed by the divisors of the group order. In that language the paper claims that the multiplicity-counted tuple alone is a full system of invariants for finite simple groups, and also for symmetric groups; it observes that Shi's conjecture is exactly the statement that the order together with the element-order indicator tuple is full for simple groups. The framework is meant to make the Shi, Thompson, and Huppert recognition conjectures comparable instances of one question, and to turn open problems about related groups into concrete questions about which tuples separate isomorphism classes.","feed_headline":"One multiplicity tuple may identify every finite simple group","feed_subtitle":"The same divisor-indexed vectors cover element orders, class sizes, and character degrees.","key_machinery":"The central object is the delta-invariant tuple. For $\\mathrm{inv} \\in \\{\\mathrm{eo}, \\mathrm{cs}, \\mathrm{cd}\\}$, list the divisors $d_i$ of $n(G)$ in increasing order and put $\\delta^{\\mathrm{inv}}_i(G) = 1$ if $d_i$ occurs in the corresponding set and $0$ otherwise; for $\\mathrm{inv}^*$ replace this by the multiplicity of $d_i$ in the multiset. The paper's intended bridge is the assertion that $n(G)$ is recovered from the multiplicity tuples by $n(G) = \\sum_i \\delta^{\\mathrm{eo}^*}_i(G) = \\sum_i \\delta^{\\mathrm{cs}^*}_i(G) = \\sum_i (\\delta^{\\mathrm{cd}^*}_i(G))^2$, so that $\\delta^{\\mathrm{inv}^*}(G)$ determines $(n(G), \\delta^{\\mathrm{inv}}(G))$; the fullness results then reduce to known recognition theorems expressed in this vector language.","core_discovery":"On the paper's own terms, the central claim is that for $\\mathrm{inv}^* \\in \\{\\mathrm{eo}^*, \\mathrm{cs}^*, \\mathrm{cd}^*\\}$, the divisor-indexed multiplicity tuple $\\delta^{\\mathrm{inv}^*}(G)$ is a full system of invariants within the class of finite simple groups (Theorem 2.5) and within the symmetric groups (Theorem 3.4). The same formalism recasts Shi's conjecture as fullness of the pair $(n(G), \\delta^{\\mathrm{eo}}(G))$, and the paper reports that $(n(G), \\delta^{\\mathrm{cs}}(G))$ is full for simple and for alternating and symmetric groups, while $(n(G), \\delta^{\\mathrm{cd}}(G))$ remains open for simple groups. The note also surveys what is known for almost simple, quasisimple, and almost quasisimple groups, including fullness of $\\delta^{\\mathrm{cd}^*}$ for quasisimple groups and counterexamples showing that element-order data alone cannot separate some groups related to $A_6$.","pith_inferences":["If the intended bridge can be repaired by replacing the sums with $n = \\sum_i d_i \\delta^{\\mathrm{cs}^*}_i(G)$ and $n = \\sum_i d_i^2 \\delta^{\\mathrm{cd}^*}_i(G)$, the same framework would survive; checking which of Theorems 2.3, 2.5, and 3.4 remain derivable is a direct next step.","One could test the separating power of these tuples computationally over small groups: for each order up to some bound, ask whether any two non-isomorphic groups share the same $\\delta^{\\mathrm{inv}^*}$ tuple, which would give a low-cost empirical check of the spirit of Theorem 2.5 before a repaired proof appears.","The paper's distinction between the minimal input $(n, \\delta^{\\mathrm{inv}})$ and the maximal input $\\delta^{\\mathrm{inv}^*}$ suggests a natural interpolation problem: for which groups does a partial multiplicity tuple, say only the first few divisors, already separate isomorphism classes?"],"forward_implications":["For every finite simple group $L$, any finite group $G$ with the same element-order multiplicity tuple as $L$ would have to be isomorphic to $L$, and the analogous statement would hold for conjugacy-class-size and character-degree multiplicity tuples.","For every symmetric group $S_m$, the same uniqueness would hold under each of the three multiplicity tuples, extending the known characterization by character degrees alone.","Shi's conjecture would be exactly the statement that order plus the element-order indicator tuple separates simple groups, giving a common formulation under which Thompson's and Huppert's conjectures can be compared.","For quasisimple groups, fullness of the character-degree multiplicity tuple would mean that the complex group algebra determines the group, answering a question that originates in representation theory.","The open status of $(n, \\delta^{\\mathrm{cd}})$ for simple groups would be located as the one missing piece in the uniform framework, with Huppert's conjecture sufficient but not necessary for a positive answer."],"supporting_citations":[{"why":"Proves Shi's conjecture for all simple groups, which the paper restates as fullness of $(n, \\delta^{\\mathrm{eo}})$ in Theorem 2.1.","marker":"[34]"},{"why":"Source for Theorem 2.2, the fullness of $(n, \\delta^{\\mathrm{cs}})$ for alternating and symmetric groups, used for Theorem 2.3.","marker":"[14]"},{"why":"Establishes Thompson's conjecture for all simple groups except alternating groups, yielding fullness of $(n, \\delta^{\\mathrm{cs}})$ for those groups.","marker":"[11]"},{"why":"Covers the alternating-group case of Thompson's conjecture, modulo the binary Goldbach conjecture.","marker":"[10]"},{"why":"Shows alternating and sporadic simple groups are determined by their character degrees, part of the $\\delta^{\\mathrm{cd}^*}$ fullness used in Theorem 2.5.","marker":"[28]"},{"why":"Shows simple exceptional groups of Lie type are determined by character degrees, another component of $\\delta^{\\mathrm{cd}^*}$ fullness.","marker":"[30]"},{"why":"Shows simple classical groups of Lie type are determined by character degrees, completing the $\\delta^{\\mathrm{cd}^*}$ simple-group case.","marker":"[31]"},{"why":"Shows symmetric groups are determined by their character degrees, supporting Theorem 3.4 for $\\delta^{\\mathrm{cd}^*}$.","marker":"[29]"},{"why":"Shows $\\delta^{\\mathrm{cd}^*}$ is full for all quasisimple groups, one of the paper's principal related-group results.","marker":"[4]"}],"fun_headline_variants":["A single divisor-indexed tuple identifies every finite simple group","One tuple of multiplicities is a complete invariant for simple groups","The same arithmetic tuple separates all finite simple groups","Finite simple groups: one divisor-indexed tuple identifies each","A uniform tuple of invariants pinpoints every finite simple group"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise, displayed in Section 2, is that the multiplicity tuples determine the order of $G$ through the identities $n(G) = \\sum_i \\delta^{\\mathrm{cs}^*}_i(G) = \\sum_i (\\delta^{\\mathrm{cd}^*}_i(G))^2$; as written those identities are false, because the first sum counts conjugacy classes rather than elements and the second sums the squares of class counts of character degrees rather than the degrees themselves.","fun_headline_variants_meta":{"raw":{"variants":["A single divisor-indexed tuple identifies every finite simple group","One tuple of multiplicities is a complete invariant for simple groups","The same arithmetic tuple separates all finite simple groups","Finite simple groups: one divisor-indexed tuple identifies each","A uniform tuple of invariants pinpoints every finite simple group"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000794,"raw_usage":{"total_tokens":3417,"prompt_tokens":784,"completion_tokens":2633,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":400,"completion_tokens_details":{"reasoning_tokens":2551}},"tokens_in":400,"tokens_out":2633,"duration_ms":18100,"temperature":1.0,"reasoning_tokens":2551,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T23:23:12.122333+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Direct computation settles the claimed bridge: for the symmetric group $S_3$, $\\delta^{\\mathrm{cs}^*}$ has one class of each size $1$, $2$, and $3$, so $\\sum_i \\delta^{\\mathrm{cs}^*}_i = 3$ while $|S_3| = 6$; for the cyclic group $C_2$, $\\delta^{\\mathrm{cd}^*}$ has two entries equal to $1$, so $\\sum_i (\\delta^{\\mathrm{cd}^*}_i)^2 = 4$ while $|C_2| = 2$. Since $C_2$ is a simple group, the displayed identity in Section 2 cannot be used to derive Theorem 2.5, and the fullness assertion would need a different proof to stand.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Proves Shi's conjecture for all simple groups, which the paper restates as fullness of $(n, \\delta^{\\mathrm{eo}})$ in Theorem 2.1."},{"cited_title":"Characterization of the alternating and symmetric groups by the order and conjugacy class sizes","cited_arxiv_id":"2606.29866","evidence_quote":"Source for Theorem 2.2, the fullness of $(n, \\delta^{\\mathrm{cs}})$ for alternating and symmetric groups, used for Theorem 2.3."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes Thompson's conjecture for all simple groups except alternating groups, yielding fullness of $(n, \\delta^{\\mathrm{cs}})$ for those groups."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Covers the alternating-group case of Thompson's conjecture, modulo the binary Goldbach conjecture."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows alternating and sporadic simple groups are determined by their character degrees, part of the $\\delta^{\\mathrm{cd}^*}$ fullness used in Theorem 2.5."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows simple exceptional groups of Lie type are determined by character degrees, another component of $\\delta^{\\mathrm{cd}^*}$ fullness."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows simple classical groups of Lie type are determined by character degrees, completing the $\\delta^{\\mathrm{cd}^*}$ simple-group case."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows symmetric groups are determined by their character degrees, supporting Theorem 3.4 for $\\delta^{\\mathrm{cd}^*}$."},{"cited_title":"Bessenrodt, H","cited_arxiv_id":null,"evidence_quote":"Shows $\\delta^{\\mathrm{cd}^*}$ is full for all quasisimple groups, one of the paper's principal related-group results."}],"review_version":1}