{"id":"c5360a99-d7bf-40f6-95dd-bdb7fc1c5da1","arxiv_id":"2507.12354","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For large n, the balanced complete bipartite 3-graph is the unique extremal construction for the ℓ2-norm Turán problem of the Fano plane, confirming a conjecture of Balogh-Clemen-Lidický.","lead":"This paper resolves a conjecture about the Fano plane, a famous small hypergraph, in a norm-based Turán problem. It proves that for large hypergraphs without a Fano plane, the balanced complete bipartite 3-graph is the unique maximizer of the ℓ2-norm.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Proposition 2.3 is a deferred, load-bearing black box; the proof of Theorem 1.1 is conditional on it.","rationale":"The central result is Theorem 1.1, and its proof has three pillars: (a) the multigraph stability theorem 2.4, proved in Section 5; (b) the refined Ahlswede–Katona bound, Proposition 2.3, stated without proof; (c) the imported flag-algebra result Theorem 2.8. I checked the internal structure of the argument as far as possible: Lemma 3.5, Claim 3.7, and the applications of Theorem 2.4 are consistent, and the vertex-extendability argument in Section 4 does not have an evident gap. The calculations in Claims 3.8–3.10 are algebraically consistent: e.g., the substitution of α1 and α2 into the bound of Proposition 2.3 yields the stated numerical roots, and the margin in the final contradiction, while small, is positive. Thus the proof is not internally inconsistent. The single point at which the argument is not self-contained is Proposition 2.3. It is load-bearing: without it, Lemma 3.6 has no proof, and Theorem 1.1 immediately loses its main contradiction. The reader's conditional verdict is therefore appropriate. I would not recommend rejection, because the framework and all other components are credible and the missing proposition is plausibly true; but the paper should either supply the proof of Proposition 2.3 or be explicitly conditional on [HLZ]. The concrete test is to verify that proposition and the subsequent threshold algebra. This is an agreement with the reader's weakest-assumption analysis.","tokens_in":39384,"tokens_out":26128,"duration_ms":250995,"concrete_test":"Obtain or independently produce the full proof of Proposition 2.3, and verify the stated bound on N(S2,G)/n^3 for the exact parameter combinations used in Claims 3.8–3.10 (x=ρ≈0.3446 with α=α1≈0.225 for (i); x≈0.3466 with α=α2≈0.3466 for (ii); x≈0.3467 with α=2/5 for Claim 3.10). Then substitute the resulting expression into the inequalities '5/4 − ε ≤ ...' in Claims 3.8 and 3.9 and confirm that the thresholds ρ ≥ 61/177 and ρ ≥ 253/730 (respectively) still hold with all o(1) terms made explicit. If the bound weakens by more than ~10^{-5} relative to the stated form, the contradiction in Lemma 3.6 is lost.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Proposition 2.3 (Section 2.1) is the keystone of the proof of Theorem 1.1. It is a refinement of the Ahlswede–Katona bound, stated with the remark that its proof is 'rather tedious' and deferred to a separate paper [HLZ] 'in preparation'. The proposition is used in Lemma 3.6: Claim 3.8 applies Proposition 2.3(i) to force dmin ≥ 61n^2/177, Claim 3.9 applies Proposition 2.3(ii) to force dmin ≥ 253n^2/730, and Claim 3.10 applies Proposition 2.3(ii) again to force min{|G3|,|G4|,|G5|} ≥ 321n^2/926. These three lower bounds are then combined in the final line of Lemma 3.6 to contradict Lemma 3.5. The numerical margin of that contradiction is only 5154779/2872915 − 61/34 ≈ 1.5×10^{-4} in units of n^2. If the deferred proposition is false, or is true but with a slightly weaker bound on N(S2,G)/n^3, the inequalities in Claims 3.8–3.10 fail and the proof of Theorem 1.1 collapses. Theorem 2.8 (πℓ2(K3_5)=5/16, imported from [BCL22b]) is also used, but it is published and less suspect; the genuine gap is the missing proof of Proposition 2.3.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves Theorem 1.1, an Andrásfai–Erdős–Sós-type stability theorem for Fano-plane-free 3-graphs in the ℓ2-norm: for sufficiently large n, every F-free 3-graph on n vertices with minimum ℓ2-norm degree at least (5/4 − ε)n^3 is bipartite. As a consequence, the balanced complete bipartite 3-graph Bn is the unique extremal construction for the ℓ2-norm Turán number of the Fano plane, confirming a conjecture of Balogh–Clemen–Lidický. The proof combines a refined Ahlswede–Katona bound on star counts (Proposition 2.3), a new Andrásfai–Erdős–Sós-type stability theorem for K4-free 5-multigraphs (Theorem 2.4), the published flag-algebra value πℓ2(K3_5) = 5/16 (Theorem 2.8), and the vertex-extendability framework of [CL24].","tokens_in":39695,"tokens_out":19529,"duration_ms":202017,"significance":"If the result holds, it confirms Conjecture 3.1 of Balogh–Clemen–Lidický and provides the first exact ℓ2-norm Turán result for a nondegenerate hypergraph family with a uniqueness statement. The paper contains substantial original components: a full proof of Theorem 2.4, an application giving the exact value of ex5(n, K4) for n ≥ 632, and a self-contained proof of uniqueness among bipartite 3-graphs (Lemma 4.6). The case analysis in Section 5 is extensive and appears structured. However, the central claim is conditional on Proposition 2.3, whose proof is explicitly deferred to an in-preparation companion paper [HLZ]; the manuscript itself states that the proof is 'rather tedious' and presented elsewhere. The published flag-algebra input [BCL22b] is less concerning but is another external dependency.","major_comments":[{"comment":"Proposition 2.3 is stated with its proof deferred to the authors' in-preparation manuscript [HLZ]. This is not a peripheral tool: the proposition is applied in Lemma 3.6, Claims 3.8, 3.9, and 3.10, and the final contradiction in Lemma 3.6 has a numerical margin of only about 1.5×10^{-4} n^2 between 5154779/2872915 and 61/34. If Proposition 2.3 is false, or is true with a slightly weaker bound on N(S2, G)/n^3, the inequalities in Claims 3.8–3.10 fail and the proof of Theorem 1.1 collapses. A complete proof of Proposition 2.3, or a reference to a publicly available version of [HLZ], is necessary before the main theorem can be considered established.","section":"Section 2.1, Proposition 2.3"},{"comment":"As printed, the first term of the N(S2, G) bound in Proposition 2.3 is α^3 + ((2x−α^2)(2x+α^2)^{1/2})/2, but the 'Consequently' line and the described construction Ŝ(n, αn, β3 n) correspond instead to (α^3 + (2x−α^2)(2x+α^2)^{1/2})/2 before doubling. The two readings differ by α^3, which is not negligible at the scale of the constants used in Claims 3.8–3.10. Since Proposition 2.3 is unproved and load-bearing, the statement must be corrected and proved in a consistent form; as it stands, the inequality solved in Claims 3.8–3.10 is not clearly the one supplied by the proposition.","section":"Section 2.1, statement of Proposition 2.3"},{"comment":"In Claim 3.8, Proposition 2.3(i) is applied to Gi0 with α treated as the size parameter, but Claim 3.7 only guarantees an independent set of size at least α2 n, where α2 = 3α1/2. Since α2 ≥ α1, one can pass to a subset of size α1 n and then verify the degree condition, but this subset argument is not stated. The same issue occurs in Claim 3.9, where a subset of size α2 n (or any admissible α in [1/3, 2/5]) must be explicitly chosen. Without this clarification, the hypotheses of Proposition 2.3 are not satisfied by the sets produced in Claim 3.7 as written.","section":"Lemma 3.6, Claims 3.8–3.10"}],"minor_comments":[{"comment":"The notation δS2(G) is used before it is defined; please define it as min_{v∈V(G)} dS2,G(v) for consistency with the rest of the paper.","section":"Section 3.2, proof of Theorem 3.2"},{"comment":"The statement that a 'minor modification' of Lemma 4.6 yields the bound N(S2, H) ≤ N(S2, Bn) for all F-free H needs justification: Lemma 4.6 concerns ∥H∥2, and the relation ∥H∥2 = 2N(S2, H) + 3|H| involves the edge count, so the claimed transfer is not immediate. Please provide the details or mark the statement as conjectural.","section":"Section 6, Theorem 6.3"},{"comment":"There are several minor typographical issues, such as missing spaces in 'p = 2and r = 2' and 'Fix a vertexvn ∈ V'; these should be corrected in the final version.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is technically ambitious and, if Proposition 2.3 is supplied or replaced by a complete proof, the main theorem is likely to be correct. I would ask the editor to require the authors to provide a complete proof of Proposition 2.3, or a publicly available version of [HLZ], before acceptance. The ambiguity in the statement of Proposition 2.3 should also be resolved, as the proof relies on the exact form of the bound. The paper fits the journal's scope and the remainder of the proof is substantial and careful."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper settles a real conjecture — exact ℓ2-norm Turán number of the Fano plane is 5/16 with Bn unique for large n — and the proof architecture is credible. The AES-type multigraph stability theorem (Theorem 2.4) is new and comes with a full, detailed proof in Section 5; that part looks solid and is useful on its own. Lemma 3.5 is clean, and the application of the [CL24] framework is careful. The paper is also honest about where the missing piece lives.\n\nThe problem is Proposition 2.3. It is a refinement of Ahlswede–Katona, stated without proof and deferred to an in-preparation paper [HLZ] by the same authors. It is not a sidelight: Claims 3.8–3.10 all use it to force the degree lower bounds that produce the contradiction in Lemma 3.6. The stress-test note is right that the numerical margin is thin — about 1.5×10^{-4} in units of n^2. If the proposition is false, or true but slightly weaker, the proof collapses. That makes the main theorem conditional as submitted, not wrong in logic. The imported flag-algebra result (Theorem 2.8) is published, so less of a concern, though it is still a computational black box.\n\nMy bottom line: the reader's conditional verdict is fair. The paper deserves a serious referee, but the referee should push hard for either a proof of Proposition 2.3 or a restatement of Theorem 1.1 as conditional on [HLZ]. I would bring this to a reading group, mainly to discuss the proof strategy and the multigraph theorem, and I would cite it for Theorem 2.4 and the overall approach — with a caveat on the main result. If the missing lemma appears, this becomes a strong paper; until then, treat it as a well-structured claim with a load-bearing gap.","headline":"Strong and likely correct, but the main theorem is conditional on a deferred unpublished lemma that carries the proof's weight.","tokens_in":40224,"tokens_out":2899,"would_cite":true,"duration_ms":35202,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["05C65","05D05","05C35"],"pacs":[],"model":"deepseek-v4-flash","headline":"The balanced complete bipartite 3-graph is the unique extremal Fano-free 3-graph for the ℓ2-norm Turán problem, for all large n.","keywords":["Fano plane","ℓ2-norm Turán number","hypergraph Turán density","extremal stability","balanced complete bipartite 3-graph","star counting","multigraph Turán problem","vertex-extendability"],"falsifier":"Search for, for the value of ε claimed in Theorem 1.1, any Fano-free 3-graph on large n that is not bipartite but has δℓ2(H) ≥ (5/4 − ε)n³; finding one would disprove the main theorem. A cheaper check is to verify Proposition 2.3 numerically on graphs with edge density in [17/50, 7/20] and an independent set of the prescribed size and degrees: any counterexample to that bound would break the proof, even if the theorem itself remains true.","tokens_in":39165,"feed_emoji":"📐","tokens_out":11652,"duration_ms":121088,"temperature":0.7,"pith_summary":"This paper determines the ℓ2-norm Turán problem for the Fano plane, the unique linear 3-graph on seven vertices with seven edges. For all sufficiently large n, the balanced complete bipartite 3-graph Bn is the unique Fano-free 3-graph maximizing ||H||2, the maximum equals ||Bn||2, and the corresponding Turán density is 5/16; this confirms a conjecture posed in [BCL22a]. The route is a stability theorem of the classical extremal-graph type: every Fano-free 3-graph on n vertices with minimum ℓ2-norm degree at least (5/4 − ε)n³ must be bipartite. The proof combines a new structural theorem for K4-free 5-multigraphs, a refined upper bound on two-edge stars in graphs with large high-degree independent sets, and a vertex-extendability argument showing that near-extremal graphs that become bipartite after deleting a vertex were already bipartite. A reader should care because it shows a nonlinear norm Turán problem can have exactly the same unique extremal construction as the classical edge-count problem.","feed_headline":"Balanced 3-graph solves Fano plane's ℓ2 Turán problem","feed_subtitle":"High ℓ2-degree forces bipartiteness, confirming the conjectured unique extremal 3-graph.","key_machinery":"The argument is carried by rewriting ℓ2-norms as counts of two-edge stars. For a 3-graph H, ||H||2 = 2N(S2, H) + 3|H| and for each vertex v, d2,H(v) = 2dS2,H(v) + 3dH(v), where S2 is the 2-edge star whose center is a pair of vertices; this identity lets the minimum ℓ2-norm degree condition speak about how many S2 copies sit at each vertex. Around this identity the proof builds three tools: a stability theorem for K4-free 5-multigraphs (Theorem 2.4), which describes the possible structure of the five link graphs of a K5³ under a minimum-degree restriction; a refinement of the classical star-counting theorem of [AK78] (Proposition 2.3), which bounds two-edge stars in a graph whose edge density lies in [17/50, 7/20] and which has a large independent set of high-degree vertices; and the vertex-extendability framework of [CL24], which converts edge-stability plus vertex-extendability into the degree-stability needed for the final theorem.","core_discovery":"The central claim is that, for large n, the ℓ2-norm Turán number of the Fano plane F is exactly ||Bn||2, and Bn is the unique extremal construction; equivalently πℓ2(F) = 5/16. To prove this, the paper establishes the stronger statement that any F-free 3-graph on n vertices with δℓ2(H) ≥ (5/4 − ε)n³ must be bipartite. The proof first shows πℓ2(F) = 5/16: an extremal graph either avoids the complete 3-uniform hypergraph K5³ on five vertices, in which case the known value πℓ2(K5³) = 5/16 applies, or it contains a copy of K5³, in which case the five link graphs form a K4-free 5-multigraph whose structure is incompatible with a high minimum ℓ2-norm degree. Then edge-stability, vertex-extendability, and a uniqueness computation for bipartite 3-graphs upgrade the density bound to the exact unique extremal result.","pith_inferences":["The same mechanism should extend to every ℓp-norm with p ≥ 1: since ||H||p is a linear combination of p-star counts via Stirling numbers, an analogue of the refined star-counting bound would likely force Bn to remain the unique extremal construction for all p; the paper proves only p = 2.","The paper's own authors note that Proposition 2.3 is proved only for edge densities x in [17/50, 7/20] but believe the bound should hold for all x > 1/4; proving that extension would simplify the case analysis in Lemma 3.6 and might reduce numerical constants such as 58/17 and 61/176.","A small-n exhaustive or SAT-based search for Fano-free 3-graphs exceeding ||Bn||2 could test how early the asymptotic extremal result begins; if no counterexample appears up to moderate n, it would support the natural guess that the exact result holds for all n ≥ 4, not merely for large n."],"forward_implications":["For all sufficiently large n, exℓ2(n, F) = ||Bn||2, and Bn is the unique Fano-free 3-graph attaining it.","The ℓ2-norm Turán density of the Fano plane is exactly 5/16, matching the value conjectured in [BCL22a].","Every F-free 3-graph with minimum ℓ2-norm degree at least (5/4 − ε)n³ is bipartite, and near-extremal graphs are bipartite after removing o(n³) edges.","The exact value ex5(n, K4) = 2 binom(n,2) + 3 floor(n²/4) for n ≥ 632 follows from the new multigraph stability theorem.","The associated generalized Turán number N(S2, H) is uniquely maximized by Bn."],"supporting_citations":[{"why":"States the conjecture that Bn is the unique extremal F-free 3-graph for the ℓ2-norm Turán problem, which Theorem 1.1 confirms.","marker":"[BCL22a, Conjecture 3.1]"},{"why":"Supplies πℓ2(K5³) = 5/16 and the structural stability statement used to dispose of the K5³-free case.","marker":"[BCL22b, Theorem 1.6]"},{"why":"The classical star-counting theorem whose refinement, Proposition 2.3, is needed to obtain the degree lower bounds in Lemma 3.6.","marker":"[AK78]"},{"why":"Introduces the 5-multigraph Turán bound ex5(n, K4) and the construction behind nice and good partitions, used throughout and in Theorem 6.1.","marker":"[BR19]"},{"why":"The companion paper that is supposed to contain the proof of the deferred Proposition 2.3; the main proof cites it as a black box.","marker":"[HLZ]"},{"why":"Provides the general framework, Theorem 2.10, that turns ℓ2-edge-stability plus ℓ2-vertex-extendability into the degree-stability needed for Theorem 1.1.","marker":"[CL24]"},{"why":"Supplies the ℓp-norm notation, the expression for d2,H(v), and related star-counting lemmas used throughout.","marker":"[CIL+24]"},{"why":"The stability method template invoked to state and prove the Erdős–Simonovits-type stability for F-free 3-graphs.","marker":"[Sim68]"}],"fun_headline_variants":["ℓ2 Turán number of Fano plane: balanced bipartite is unique extremal","Exact ℓ2 Turán number for Fano plane found, extremal is bipartite","Fano-free 3-graphs with high ℓ2-degree must be bipartite","Unique extremal for ℓ2 Turán of Fano plane: balanced bipartite 3-graph"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is a refined bound on two-edge stars in graphs with a large independent set of high-degree vertices, stated as Proposition 2.3 with its proof deferred to a separate paper [HLZ]; the contradiction establishing the main theorem collapses if that bound is false or unavailable, and the proof also imports, without proof, the value of the ℓ2-norm Turán density of the complete 3-graph K5³.","fun_headline_variants_meta":{"raw":{"variants":["ℓ2 Turán number of Fano plane: balanced bipartite is unique extremal","Exact ℓ2 Turán number for Fano plane found, extremal is bipartite","Fano-free 3-graphs with high ℓ2-degree must be bipartite","Unique extremal for ℓ2 Turán of Fano plane: balanced bipartite 3-graph"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00051,"raw_usage":{"total_tokens":2593,"prompt_tokens":1168,"completion_tokens":1425,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":784,"completion_tokens_details":{"reasoning_tokens":1327}},"tokens_in":784,"tokens_out":1425,"duration_ms":12687,"temperature":1.0,"reasoning_tokens":1327,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T16:47:22.160925+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Search for, for the value of ε claimed in Theorem 1.1, any Fano-free 3-graph on large n that is not bipartite but has δℓ2(H) ≥ (5/4 − ε)n³; finding one would disprove the main theorem. A cheaper check is to verify Proposition 2.3 numerically on graphs with edge density in [17/50, 7/20] and an independent set of the prescribed size and degrees: any counterexample to that bound would break the proof, even if the theorem itself remains true.","supporting_citations":[],"review_version":1}