{"id":"0a6017ae-7393-4683-801c-2da681d8d75b","arxiv_id":"2508.18660","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Every block-transitive automorphism group of a nontrivial t-(k^2,k,λ) design has a socle that is not a finite simple exceptional group of Lie type.","lead":"This paper proves that no finite simple exceptional group of Lie type can act as a block-transitive automorphism group of a nontrivial t-(k^2,k,λ) design. The result narrows the classification of block-transitive designs by removing an entire family of candidate symmetry groups.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Parabolic elimination in Lemma 3.1 rests on unpublished [11, Table 5]; a single wrong p-part entry would leave an entire case unexamined, so independent recomputation is required.","rationale":"The reader's weakest_assumption identified exactly the same point: the final paragraph of Lemma 3.1 is a blanket delegation to the unpublished [11, Table 5], and a single incorrect entry would leave an unexamined parabolic case. My reading of the proof confirms that this is the most load-bearing step. The overall reduction is coherent: Corollary 2.5 makes the point stabilizer large, Lemma 2.6 classifies large maximal subgroups, and Lemmas 3.4–3.12 reduce to parabolic, numerical, subfield, and listed non-parabolic cases. However, the parabolic and subfield eliminations are not reproducible from the manuscript as submitted. Lemma 3.2 also contains a divisibility sentence that is false as written: for G2(4)/A1(13), k+1=481 does divide v−1=230399, although the required gcd divisibility fails; this supports the need for an arithmetic audit rather than refuting the case. No internal inconsistency in the main reduction is apparent, so the theorem remains plausible. The verdict should therefore remain CONDITIONAL: accept only after the missing table and computations are supplied or independently verified.","tokens_in":18182,"tokens_out":8024,"duration_ms":76552,"concrete_test":"Reconstruct [11, Table 5] for every maximal parabolic pair (X, P_i) not handled explicitly in Lemma 3.1: compute v_i = |X(q)|/|P_i(q)| from the standard order formulas, then compute p^a = |v_i − 1|_p by exact integer arithmetic in Sage or Magma. For each pair, verify the inequality p^a < sqrt(v_i), which is the condition that makes k+1 ≤ p^a contradict k = sqrt(v). If any pair has p^a ≥ sqrt(v_i), the final paragraph of Lemma 3.1 does not eliminate it. Include the full table and the inequality check in the paper so the proof is self-contained.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The theorem is a universal negative over all exceptional Lie-type socles, so every maximal point stabilizer must be eliminated. The only argument covering almost all parabolic cases is the final paragraph of Lemma 3.1: it invokes [11, Table 5] for the values of |v−1|_p and then states that these values give upper bounds too small for v=k^2. Neither the table nor the inequalities are reproduced, and the cited preprint is not substituted by the published [2]. Since Lemma 3.1 also does not show the intermediate step that k+1 divides |v−1|_p in each row, a reader cannot check the decisive numerical premise. The concern is concrete rather than stylistic: in Lemma 3.2, the G2(4)/A1(13) row says k+1 | v−1 fails, but 481 divides 230399; what actually fails is the gcd condition k+1 | gcd(1092,230399). That discrepancy shows the numerical eliminations need independent auditing, even though it may not invalidate the case. The subfield cases in Lemma 3.3 similarly depend on unshown Magma computations. One wrong p-adic valuation in the parabolic table would leave a whole family of possible designs untreated, so the central claim is not yet checkable at its load-bearing step.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves Theorem 1.1: no block-transitive automorphism group of a nontrivial t-(k^2,k,λ) design can have socle a finite simple exceptional group of Lie type. The proof uses the known reduction that such a group is point-primitive and almost simple or affine, then combines the classification of large maximal subgroups of exceptional groups with subdegree and p-part divisibility arguments to eliminate, in turn, parabolic, subfield, and the remaining maximal-subgroup cases for 2B2(q), 2G2(q), 3D4(q), 2F4(q), G2(q), F4(q), E6±(q), E7(q), and E8(q).","tokens_in":18436,"tokens_out":5206,"duration_ms":44466,"significance":"The main theorem is a clean universal negative result: it closes the exceptional Lie-type family for block-transitive t-(k^2,k,λ) designs, complementing existing reductions and the sporadic and alternating cases. The high-level architecture is sensible: the reduction to point-primitivity, the use of Lemma 2.6 to bound the maximal stabilizers, and the subdegree arguments in Lemmas 3.4-3.12 are mostly explicit and checkable. That said, the proof leans on two kinds of large unshown computational input—unpublished Table 5 of [11] in Lemma 3.1 and Magma XGCD identities in Lemma 3.3—and contains at least one incorrect divisibility assertion in Lemma 3.2. Because the theorem is a universal negative over all exceptional groups, every maximal point stabilizer must actually be eliminated by a verifiable argument; as written, the decisive parabolic elimination is not independently checkable. I regard the central claim as very likely correct, but the manuscript needs substantial revision before it meets a standard acceptable for publication.","major_comments":[{"comment":"This paragraph eliminates all remaining parabolic cases by citing [11, Table 5] for the values of |v-1|_p and asserting that the resulting upper bounds are 'too small to satisfy v=k^2'. Neither the table nor the inequalities are reproduced, and the cited item is an arXiv preprint (1702.01257) rather than the published [2]. Since every parabolic maximal subgroup of every exceptional group except the few cases handled explicitly in Lemma 3.1 is disposed of only in this paragraph, a single erroneous entry in the table would leave an entire family of cases untreated. The manuscript must either reproduce the relevant rows of [11, Table 5] together with the derived upper bounds, or provide a self-contained derivation of the inequalities. In addition, the sentence 'Thus, k+1 ||v-1|_p' should be justified by showing that the unique p-power subdegree equals |v-1|_p in each row, not merely that k+1 divides that subdegree.","section":"Lemma 3.1, final paragraph"},{"comment":"The proof says that for X = G2(4), Gx∩X = A1(13), v = 230400, k = 480, and 'this does not hold for k+1 | v−1'. This statement is factually wrong: 481 divides 230399, because 481·479 = 230399. The case is eliminated correctly by Corollary 2.3 only if one observes that k+1 does not divide gcd(|Gx|,v−1); indeed gcd(1092,230399) = 13, so 481 ∤ gcd(1092,230399). As written, the justification for this row is false. This discrepancy shows that the numerical eliminations in Table 3 need independent checking; the authors should verify every row using the correct gcd condition and state the relevant gcd values.","section":"Lemma 3.2, G2(4) row"},{"comment":"The proof of the subfield cases depends on Magma XGCD computations that are not shown: e.g., the assertion that for (2B2(q), 2B2(q0)) there exist polynomials P and Q with P|Gx∩X| + Q(v−1) = 20, and the claim that 'using similar computations' yields the list in Table 4. Without the explicit polynomials, the inequality v < (|Gx|,v−1)^2, or the exhaustive search leading to the candidate q values, a reader cannot verify that all subfield subgroups have been eliminated. The table also lists 'possible q' entries such as '2 2' and 'i2 with i ≤ 5' that are cryptic; the manuscript should state the actual q values and, for each, the value of v that fails to be a square.","section":"Lemma 3.3"}],"minor_comments":[{"comment":"In the 2B2(q) paragraph, 'k = 2m−1' should read 'k+1 = 2^m' (and similarly 'k = 3m−1' in the 2G2(q) paragraph should read 'k+1 = 3^m'); the displayed equations then become (2^m−1)^2 = 2^{2(2e+1)}+1 and (3^m−1)^2 = 3^{3(2e+1)}+1, which are the intended contradictions.","section":"Lemma 3.1, 2B2(q) and 2G2(q) arguments"},{"comment":"The row 'G2(2) 12096 484375' is ambiguous: G2(2) is not simple, and the numerical data appear to belong to X = G2(5) with H∩X = G2(2). The table should label the row as (X, Gx∩X) = (G2(5), G2(2)) with the appropriate |Gx∩X| and v.","section":"Table 3, G2(2) row"},{"comment":"The displayed inequality '(q^{12}-1)(q^9-1) < q^2(q^4+1)^2(q^4-1)(q-1)' seems dimensionally inconsistent and is not explained; it should be derived from k+1 | q(q^4+1) and v = k^2. Please show the intermediate algebra.","section":"Lemma 3.1, E6(q), P1 case"},{"comment":"The data availability statement reads 'No date was used for the research described in this article.' This should be corrected to 'No data was used...'.","section":"Data availability"},{"comment":"Reference [11] appears to be an earlier arXiv version of the published [2]; the authors should clarify its relationship to [2] and, if the required table appears in [2], cite the published source instead, or explain why the preprint version is necessary.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The central theorem is plausible and the overall strategy is sound, but the current proof is not verifiable at its two most load-bearing steps: the parabolic elimination in Lemma 3.1 rests on an unpublished table, and the subfield elimination in Lemma 3.3 rests on undocumented Magma computations. In addition, the G2(4) row in Lemma 3.2 contains a definite false statement. These are fixable within the manuscript's scope, but the authors should be required to make the numerical evidence fully transparent before the paper is accepted."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe short version: the theorem is likely right, and the paper is a real step forward—it eliminates all exceptional Lie-type socles for block-transitive t-(k^2,k,λ) designs, closing the gap left after alternating and sporadic cases. The case analysis is structured sensibly: reduce to point-primitive, invoke the classification of large subgroups [2], then grind through parabolics, subfields, and the remaining maximal subgroups with subdegree and gcd arguments. Much of the explicit arithmetic in Lemmas 3.4–3.12 is checkable and looks correct. Credit where due: this is honest, substantial work.\n\nBut the proof has two load-bearing spots that are not substantiated in the manuscript. The final paragraph of Lemma 3.1 dismisses almost all parabolic cases by citing [11, Table 5] for the values of |v-1|_p and asserting the resulting bounds are too small. That table is not reproduced, and [11] is an unpublished preprint, not the published [2]. Since the theorem is a universal negative, one wrong p-part entry would leave an entire family of possible designs unexamined. That is the weakest point in the paper. Lemma 3.3 has a similar problem: the subfield cases are eliminated with Magma gcd computations, but the code and output are not shown, so the reader has to take the entries of Table 4 on faith.\n\nThere is also a concrete slip in Lemma 3.2. For the G2(4) with G_x∩X = A1(13) row, the paper says k+1 | v−1 fails. It does not: 481 divides 230399. What actually fails is the divisibility condition from Corollary 2.3, k+1 | (|G_x|, v−1), since gcd(1092, 230399)=13. The case probably still dies, but the sentence is wrong and it signals that the numerical eliminations need independent auditing.\n\nThe central argument holds up. These are not fatal flaws; they are missing evidence at exactly the points where the proof is least transparent. I would send this to a serious referee, not desk-reject it, and the referee should demand the parabolic table (or a published source), the Magma transcripts, and a corrected Lemma 3.2. Anyone working on block-transitive designs or on subgroup classification applications will want this result once it is verifiable.","headline":"A serious and probably correct classification result, but the decisive parabolic and subfield exclusions rest on an unpublished table and unshown Magma output, so the proof is not yet independently checkable.","tokens_in":18966,"tokens_out":3858,"would_cite":true,"duration_ms":35152,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["05B05","05B25","20B25"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves that no block-transitive t-(k²,k,λ) design can have a finite simple exceptional group of Lie type as the socle of its automorphism group.","keywords":["t-design","block-transitive","automorphism group","exceptional group of Lie type","point-primitive","large subgroups","t-(k^2,k,λ) design"],"falsifier":"Recompute, for each finite simple exceptional group of Lie type and each maximal parabolic subgroup $P_i$, the ratio $v=|X:P_i|$ and the power of the defining prime dividing $|v-1|$; the theorem requires $k+1$ to divide the relevant gcd for the almost simple point stabilizer and the implied bound on $k$ to fall below $\\sqrt{v}$, so a single parabolic pair satisfying all divisibility and square conditions would point directly to a candidate counterexample, and a wrong entry in the table used in Lemma 3.1 would invalidate that step.","tokens_in":17949,"feed_emoji":"📐","tokens_out":17107,"duration_ms":155379,"temperature":0.7,"pith_summary":"The paper proves a negative classification statement: if a nontrivial $t$-$({k}^{2},k,\\lambda)$ design has an almost simple block-transitive automorphism group, then the socle of that group cannot be a finite simple exceptional group of Lie type. This matters because earlier work had reduced block-transitive automorphism groups of such `square' designs to affine or almost simple groups, so the theorem closes off the entire exceptional family at once and leaves only alternating, classical, and sporadic socles to consider. The proof exploits the point-primitivity that block-transitivity forces: the point stabilizer must be a `large' maximal subgroup, and the classification of those subgroups for exceptional Lie-type groups gives a short candidate list. Each candidate is then checked against the divisibility constraint $k+1 \\mid (|G_x|, v-1)$ that follows from block-transitivity, and in every case the arithmetic forces $v=k^2$ to be impossible.","feed_headline":"Square designs rule out exceptional Lie-type group socles","feed_subtitle":"For block-transitive t-(k²,k,λ) designs, only alternating, classical, and sporadic socles remain possible.","key_machinery":"The load-bearing mechanism is the pair of inequalities that convert block-transitivity into arithmetic restrictions. Corollary 2.5 gives $|G|<|G_x|^3$, so the point stabilizer is a `large' subgroup (meaning $|G|\\le |H|^3$); Lemma 2.2 and Corollary 2.3 give $k+1 \\mid n$ for every nontrivial subdegree $n$, hence $k+1 \\mid (|G_x|,v-1)$. Tits' lemma (a proper subgroup whose index is prime to the defining characteristic lies in a parabolic subgroup) and a theorem on unique $p$-power subdegrees for parabolic actions turn the divisibility into explicit upper bounds on $k$. The classification of large maximal subgroups of almost simple groups with exceptional Lie-type socle is the central object: it supplies the finite candidate list for $G_x\\cap X$ that the arithmetic checks eliminate one by one.","core_discovery":"The paper's central claim is Theorem 1.1: for a nontrivial $t$-$({k}^{2},k,\\lambda)$ design $\\mathcal{D}$ with an almost simple block-transitive automorphism group $G$, the socle $X=\\mathrm{Soc}(G)$ cannot be a finite simple exceptional group of Lie type. Assuming the contrary, the argument uses the known reduction that $G$ is point-primitive, so the point stabilizer $G_x$ is maximal in $G$ and does not contain $X$. From $v=k^2=|G:G_x|$ and $k+1 \\mid (|G_x|,v-1)$ one obtains $|G|<|G_x|^3$, meaning $G_x$ is a large subgroup; the classification of large maximal subgroups of almost simple groups with exceptional socle then leaves only parabolic subgroups, subfield subgroups, and a short table of exceptional candidates. The proof removes the parabolic cases using subdegree divisibility and $p$-part bounds, removes subfield cases by gcd computations, handles small numerical cases directly, and then runs through the remaining maximal subgroups of ${}^2B_2(q)$, ${}^2G_2(q)$, ${}^3D_4(q)$, ${}^2F_4(q)$, $G_2(q)$, $F_4(q)$, $E_6(q)$, ${}^2E_6(q)$, $E_7(q)$, and $E_8(q)$, showing that in every case the forced divisibility makes $v=k^2$ impossible.","pith_inferences":["My inference: the same `large stabilizer plus $k+1$-divisibility' strategy should extend to classical socles, using the known large-subgroup classification; it may exclude or greatly shrink the classical candidate list.","My inference: the proof suggests a computational route to the same theorem, namely enumerating the classified large maximal subgroups, computing $v=|X:H|$ and the gcd constraints for admissible $q$, and checking that $v$ is never a square; this would make the unpublished parabolic table auditable.","My inference: because the divisibility inequality involves only $v=k^2$ and the point-stabilizer order, the method transfers to designs with $v=c k^2$ for a fixed small integer $c$ after replacing the square test by $v=c k^2$; the paper does not explore this."],"forward_implications":["Any almost simple block-transitive automorphism group of a $t$-$({k}^{2},k,\\lambda)$ design must have alternating, classical, or sporadic socle; the exceptional Lie-type families are fully excluded.","For flag-transitive $2$-$({k}^{2},k,\\lambda)$ designs with $\\lambda \\mid k$, the earlier no-exceptional-socle result is recovered as a special case of the block-transitive theorem.","Searches for such designs can skip stabilizers inside ${}^2B_2(q)$, ${}^2G_2(q)$, ${}^3D_4(q)$, ${}^2F_4(q)$, $G_2(q)$, $F_4(q)$, $E_6(q)$, ${}^2E_6(q)$, $E_7(q)$, and $E_8(q)$.","Future classification work on point-primitive almost simple groups for these designs is reduced to the classical, alternating, and sporadic families."],"supporting_citations":[{"why":"Supplies the reduction that a block-transitive automorphism group of a t-(k²,k,λ) design is point-primitive and affine or almost simple, and the subdegree divisibility k+1|n used throughout.","marker":"[10]"},{"why":"Classifies the large maximal subgroups of almost simple groups with exceptional Lie-type socle and gives the subdegree bounds in Table 2 that drive the stabilizer checks.","marker":"[2]"},{"why":"Provides the table of p-parts |v−1|_p for all remaining parabolic stabilizers; Lemma 3.1's exclusion of parabolic cases rests directly on these values.","marker":"[11]"},{"why":"Earlier result excluding exceptional socles for flag-transitive 2-(k²,k,λ) designs with λ|k, which the paper states its Theorem 1.1 generalizes.","marker":"[24]"},{"why":"Supplies explicit subdegree formulas for parabolic actions of E6(q) and classical overgroups used in the parabolic and G2(q) cases.","marker":"[25]"},{"why":"Tits' lemma, used to force the defining characteristic to divide v and to control the point stabilizer in Corollary 2.11.","marker":"[26]"},{"why":"Gives the unique p-power subdegree for maximal parabolic actions of most exceptional groups, used in Lemma 3.1 to convert subdegree divisibility into p-part bounds.","marker":"[19]"}],"fun_headline_variants":["Exceptional Lie-type socles impossible for block-transitive square designs","No exceptional Lie groups as socles for block-transitive square designs","Square block designs exclude exceptional Lie-type automorphism groups","Block-transitive t-(k²,k,λ) designs forbid exceptional Lie-type socles","Exceptional Lie-type socles ruled out for square design automorphisms"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that an unpublished numerical table used in Lemma 3.1 correctly lists, for every remaining parabolic stabilizer, the exact power of the defining prime dividing $|v-1|$, and that each listed value makes $k+1$ divide $|v-1|$ with that power too small to permit $v=k^2$; if any entry is wrong or missing, a parabolic case is left open.","fun_headline_variants_meta":{"raw":{"variants":["Exceptional Lie-type socles impossible for block-transitive square designs","No exceptional Lie groups as socles for block-transitive square designs","Square block designs exclude exceptional Lie-type automorphism groups","Block-transitive t-(k²,k,λ) designs forbid exceptional Lie-type socles","Exceptional Lie-type socles ruled out for square design automorphisms"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001108,"raw_usage":{"total_tokens":4595,"prompt_tokens":901,"completion_tokens":3694,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":517,"completion_tokens_details":{"reasoning_tokens":3600}},"tokens_in":517,"tokens_out":3694,"duration_ms":27016,"temperature":1.0,"reasoning_tokens":3600,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T16:55:55.914225+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute, for each finite simple exceptional group of Lie type and each maximal parabolic subgroup $P_i$, the ratio $v=|X:P_i|$ and the power of the defining prime dividing $|v-1|$; the theorem requires $k+1$ to divide the relevant gcd for the almost simple point stabilizer and the implied bound on $k$ to fall below $\\sqrt{v}$, so a single parabolic pair satisfying all divisibility and square conditions would point directly to a candidate counterexample, and a wrong entry in the table used in Lemma 3.1 would invalidate that step.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the reduction that a block-transitive automorphism group of a t-(k²,k,λ) design is point-primitive and affine or almost simple, and the subdegree divisibility k+1|n used throughout."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Classifies the large maximal subgroups of almost simple groups with exceptional Lie-type socle and gives the subdegree bounds in Table 2 that drive the stabilizer checks."},{"cited_title":"Montinaro and E","cited_arxiv_id":null,"evidence_quote":"Earlier result excluding exceptional socles for flag-transitive 2-(k²,k,λ) designs with λ|k, which the paper states its Theorem 1.1 generalizes."},{"cited_title":"Saxl, On finite linear spaces with almost simple flag-transitive automorphism groups","cited_arxiv_id":null,"evidence_quote":"Supplies explicit subdegree formulas for parabolic actions of E6(q) and classical overgroups used in the parabolic and G2(q) cases."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Tits' lemma, used to force the defining characteristic to divide v and to control the point stabilizer in Corollary 2.11."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the unique p-power subdegree for maximal parabolic actions of most exceptional groups, used in Lemma 3.1 to convert subdegree divisibility into p-part bounds."}],"review_version":2}