{"id":"94e93429-cffb-4b17-a1eb-d26926ed1b40","arxiv_id":"2607.23748","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"For linearly quasirandom 3-graphs, tight Hamilton cycles are forced by vertex-degree above the curve f(d) when d>1/3, while the sharp codegree diagonal is κ≈0.3177, not 1/4.","lead":"The paper finds the exact density-plus-degree cutoffs that force a tight Hamilton cycle in weakly quasirandom 3-uniform hypergraphs, and shows the codegree cutoff is higher than experts conjectured. It settles two open problems and disproves one conjecture in extremal hypergraph theory.","discovery_kind":"extension","skeptic_critique":{"model":"moonshotai/kimi-k3","headline":"Internal argument holds up on close reading; the one genuinely load-bearing dependency is the quoted specialization of two very recent external preprints (Lemma 2.2 from Lang–Sanhueza-Matamala, Lemma 2.3 from Lang), through which both main theorems route with no fallback.","rationale":"The reader identified the external Hamilton-framework embedding theorem plus the inheritance lemmas as the weakest assumption, and my independent pass lands in the same place: it is the only point where a single failure would bring down both main theorems, and it is not independently re-derived in the paper. I additionally hand-checked the paper's novel internal content — the weighted monochromatic-clique theorem (Lemma 5.3) and its robust form (Theorem 5.4), the stability lemma 5.5 around κ=(1−κ)³, the two-color structure theorem 5.8, the cherry lemma 5.9, both dominant-component lemmas, and both sharpness constructions — and found the arguments correct at the level of detail given, including the constant hierarchies and the boundary roles of d≥1/3 and q>1/4. The sharpness constructions match the upper bounds exactly (f(d) curve and κ respectively), and the disclosed concurrent work of Shu does not subsume the vertex-degree curve or the codegree negative result. Because the concern is citation fidelity of recent preprints rather than an identified defect, and because the reader already priced this in with MODERATE confidence, I would not move the verdict: ACCEPT stands, with the explicit recommendation that the two external specializations be verified before the result is treated as fully settled.","tokens_in":33942,"tokens_out":16321,"duration_ms":1072706,"concrete_test":"Line-check the two citations against their sources. (i) In arXiv:2412.14891v2, verify that the k=t=3 specialization of Theorem 3.5 yields exactly Lemma 2.2: that [30]'s Hamilton-framework axioms instantiate to (F1)–(F4) of Definition 2.1 (spanning tight component, perfect fractional matching, closed walk of order 1 mod 3, pairwise consistency on (s+1)-vertex extensions), and that its hypothesis matches \"for every 6-set R, at least (1−1/s²)binom(n−6,s−6) sets S⊇R with H[S]∈P_s.\" (ii) In arXiv:2308.12281v3, confirm Lemma 8.1 gives Lemma 2.3 with the e^{−s^{1/6}} exceptional fraction under the hierarchy 1/r,d,μ′≫1/s≫μ≫1/n. If either fails to specialize as quoted, Thms 1.1 and 1.4 are unproven as written; if both check out, the central claims stand.","verdict_should_be":"UNCHANGED","load_bearing_attack":"I read the internal machinery closely and could not find an internal inconsistency. The new weighted-graph content checks out: Lemma 5.1's probabilistic disjointness trick is correct; Lemma 5.2's endpoint analysis uses d≥1/3 exactly where needed ((1−d)/(1+d) ≤ d/(1−d) ⟺ d≥1/3); the K4 case analysis in Lemma 5.3 (patterns 2+2, 2+1+1, 1+1+1+1, each giving d⁴<4⁻⁴ or d⁶<4⁻⁶) is correct, and the induction step needs m≥5 precisely to find an edge avoiding three vertices. Lemma 5.9's cherry double-count (coefficient of a_z(1−a_y) at most w_yz+(1−w_yz)=1) is correct and sharp at q=1/4. The constructions in §3 are clean: the color-class separation genuinely blocks tight Hamilton cycles, the McDiarmid/Chernoff concentration arithmetic (exp(−μ²n²/36) vs 8ⁿ set triples) closes, and the optimization max min{q,(1−q)³,b,(1−q)²(1−b)}=κ is verified by b=q=κ. The reduction in §4.3 has a consistent constant hierarchy (d0,α0 → μ′,s0 → s → μ → n0) and the e^{−s^{1/6}}+e^{−√s}<1/s² choice meets Lemma 2.2's hypothesis.\n\nThe genuinely load-bearing point is external and matches the reader's weakest assumption: Lemma 2.2 is quoted as \"the case k=t=3 of [30, Theorem 3.5]\" (arXiv:2412.14891v2, revised March 2026, apparently unpublished), and Lemma 2.3 as [29, Lemma 8.1] (arXiv:2308.12281v3, revised May 2026). Both main theorems collapse if either specialization is misquoted — e.g., if [30, Thm 3.5]'s framework definition differs from (F1)–(F4) in Definition 2.1 in some axiom, or if its local hypothesis is not exactly \"for every 6-set R, a (1−1/s²)-fraction of s-sets through R,\" or if the quantitative rate in [29, Lemma 8.1] is weaker than e^{−s^{1/6}}. There is no alternative route in the paper: absorption is explicitly inapplicable, so the framework black box is a single point of failure. This is a correctness-risk concern about citation fidelity, not about consensus or internal logic.","agreement_with_reader":"agree"},"referee_report":{"model":"moonshotai/kimi-k3","summary":"The paper studies tight Hamilton cycles in (n,d,μ)-dense 3-graphs under minimum-degree conditions. Theorem 1.1 shows that for every d∈(1/3,1] and α>f(d)=(1−√((4d−1)/3))/2, uniform density together with δ₁≥α\\binom{n−1}{2} forces a tight Hamilton cycle, and §3.1 gives matching constructions showing this boundary is sharp; the diagonal consequence (Corollary 1.2) answers Problem 8.3(i) of Araújo–Piga–Schacht and confirms Conjecture 8.1 of Han–Shu–Wang. For the codegree condition, Theorem 1.3 constructs (n,κ,μ)-dense 3-graphs with δ₂≥(κ−ε)n and no tight Hamilton cycle, where κ=(1−κ)³≈0.3177>1/4, thereby disproving Conjecture 8.2 of Han–Shu–Wang and answering Problem 8.3(ii) negatively; Theorem 1.4 shows (κ,κ) is the sharp diagonal threshold. The upper-bound proofs use a common reduction to the Hamilton-framework embedding theorem of Lang–Sanhueza-Matamala via inheritance lemmas, with the framework conditions supplied by two distinct dominant tight-component lemmas: in the vertex-degree case via a component coloring, Kruskal–Katona, multicolor regularity, and a new weighted monochromatic-clique theorem; in the codegree case via a stability analysis around κ=(1−κ)³ and a weighted two-color cherry-counting lemma.","tokens_in":34520,"tokens_out":3232,"duration_ms":71298,"significance":"If correct, this is a strong result that settles two open problems from Araújo–Piga–Schacht and Han–Shu–Wang, one positively and one negatively, and in the vertex-degree case determines the entire sharp boundary curve α=f(d) for d>1/3 rather than only the diagonal. The negative codegree answer is quantitatively interesting: the diagonal threshold κ=(1−κ)³≈0.3177 strictly exceeds 1/4, so density and codegree above 1/4 provably do not suffice, and the matching construction in §3.2 is new and clean. Beyond the headline theorems, the paper contributes reusable machinery: the weighted monochromatic-clique theorem (Lemma 5.3) and its robust form (Theorem 5.4), the two-color structural theorem (Theorem 5.8), the sharp cherry lemma (Lemma 5.9, sharp at q=1/4), and a dominant-component method that avoids absorption. The sharpness constructions are explicit and the concentration arithmetic is verified in full, so the thresholds are derived from matching extremal examples rather than assumed. The combination of new internal machinery with the recent Hamilton-framework embedding technology is a natural and apparently effective way to bypass the connecting-ends obstacle identified in earlier的工作","major_comments":[{"comment":"Both main theorems route through Lemma 2.2, stated as 'the case k=t=3 of [30, Theorem 3.5]' (Lang–Sanhueza-Matamala, arXiv:2412.14891v2, revised March 2026, apparently unpublished), together with the inheritance statements Lemma 2.3 ([29, Lemma 8.1], arXiv:2308.12281v3, revised May 2026) and Lemma 2.4 ([30, Lemma 4.3]). If the specialization is inaccurate in any framework axiom — e.g. if [30]'s framework conditions differ from (F1)-(F4) of Definition 2.1, or if the quantitative hypothesis of [30, Thm 3.5] is weaker/stronger than the 'at least (1-1/s^2) of the s-sets through every 6-set' formulation used here — the deductions in §4.3 have no fallback. Since both cited works are very recent, revised preprints, the manuscript should include a short appendix or an expanded remark that (i) derives the k=t=3 statement from [30, Thm 3.5] explicitly (identifying the parameters and confirming the","section":"§2.1, Lemma 2.2; §4.3"},{"comment":"Lemma 4.3 is load-bearing for (F4) in the codegree case (Lemma 4.5), but its proof in §7.2 ends with the truncated sentence 'contradicting.' with no object. From context the intended contradiction is with Lemma 5.9 (with q' in place of q, on the reduced red weight system and its complement), but the argument as printed also skips a step: Lemma 5.9 gives a positive proportion βr_j^2/2 of bad pairs, whereas the proof establishes the common-neighbor lower bounds only on 'all but o(r_j^2)' pairs, so one must verify that o(r_j^2) < βr_j^2/2 for large j — this is fine but should be written out, and the missing reference restored. As printed, the final step of the proof is not checkable.","section":"§7.2, proof of Lemma 4.3"}],"minor_comments":[{"comment":"The sentence 'We prove Lemmas 4.2 and 4.3. Throughout this section, κ is...' appears at the head of §5.2, but §5.2 contains the weighted-coloring auxiliary results (Lemmas 5.5-5.9); Lemmas 4.2 and 4.3 are proved in §7. The sentence presumably belongs at the start of §7.","section":"§5.2, opening paragraph"},{"comment":"The symbol q is used for the cluster size in §6, for the cherry-weight threshold in Lemma 5.9 and §7, and for the common-neighbor parameter in Lemma 2.12; m is used both for the number of clusters (§6) and as the induction variable in Lemma 5.3. Consider renaming to avoid confusion.","section":"notation throughout §§5-7"},{"comment":"In the statement of Lemma 4.3, the quantification 'there exist ε>0 and s0... Let R and B be... Then no such pair (R,B) exists' is slightly awkward; the conclusion is that the hypotheses are mutually inconsistent. A phrase such as 'no such pair exists' directly after the hypotheses would read better.","section":"§4.1, Lemma 4.3 statement"},{"comment":"Lemma 2.2 requires s≥6 and a statement for every 6-set R; a sentence explaining why r=6 (and not a larger anchor) suffices in the applications of Lemmas 2.3-2.4 in §4.3 would help the reader follow the constant hierarchy d0,α0 → μ',s0 → s → μ → n0.","section":"§4.3"},{"comment":"Please double-check reference data: [17] is listed as J. Combin. Theory Ser. B 177:1-30, 2026; [29] and [30] are cited only as preprints with 2026 revision dates. If any of these have since appeared, updated citations (and, for [30], a stable statement number for Theorem 3.5) should be given.","section":"References"},{"comment":"The diagonal formulation Corollary 1.2 is stated as an 'immediate consequence' of Theorem 1.1; one line noting that d,α>1/3 implies α>f(d) (since f(d)<1/3 for d>1/3) would make this literally immediate.","section":"§1.3, Corollary 1.2"}],"recommendation":"minor_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"This paper settles the two diagonal questions of Araújo–Piga–Schacht and Han–Shu–Wang for tight Hamilton cycles in linearly quasirandom 3-graphs, and goes further: it gives the full sharp vertex-degree curve α > f(d) for every d > 1/3, and shows the codegree diagonal is exactly (κ, κ) with κ ≈ 0.3177 > 1/4, so the 1/4 conjecture is false.\n\nWhat is new is clean. The vertex-degree obstruction (random 2-coloring of pairs plus two apex vertices) and the new codegree obstruction (random graph on A plus a small set B, with the space barrier |A| ≤ 2|B|) match the upper bounds. The two dominant-component lemmas are the real structural contribution; they are proved without absorption, via a weighted monochromatic-clique theorem (the K4 case analysis and induction check out, and the d ≥ 1/3 threshold is used exactly where the elementary inequality needs it) and a codegree stability argument around κ = (1 − κ)³ plus a cherry-counting lemma sharp at 1/4. The Hamilton-framework reduction is then routine once those lemmas are in hand.\n\nThe soft spot is real but narrow: both main theorems route through the Lang–Sanhueza-Matamala embedding theorem (and Lang’s inheritance lemma) with no fallback. The paper quotes the specialization carefully and the constant hierarchy is consistent, but if the external local hypothesis or the quantitative rate differs from what is written, the theorems collapse. That is a citation-fidelity risk on very recent preprints, not an internal contradiction. Concurrent work of Shu is disclosed and does not swallow the vertex-degree curve or the negative codegree result.\n\nThis is for people already working on Dirac-type or quasirandom hypergraph Hamiltonicity. The math and citation pattern look solid; the constructions and the weighted-graph lemmas are the parts I would actually reuse. I would send it to referees.","headline":"Sharp thresholds that settle two named problems, with solid internal math that routes through one recent external embedding black box.","tokens_in":36473,"tokens_out":499,"would_cite":true,"duration_ms":10121,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["05C65","05C45","05C35"],"pacs":[],"model":"grok-4.5","headline":"Above density 1/3, a sharp vertex-degree curve forces tight Hamilton cycles in uniformly dense 3-graphs; the codegree diagonal threshold is about 0.3177, not 1/4.","keywords":["tight Hamilton cycles","uniformly dense hypergraphs","minimum vertex degree","minimum codegree","linear quasirandomness","Hamilton frameworks","dominant tight components"],"falsifier":"Exhibit an infinite family of (n,d,μ)-dense 3-graphs with minimum vertex degree above f(d) (or codegree above κ) that still lack a tight Hamilton cycle, or show that the Lang–Sanhueza-Matamala embedding fails at the error scales used here.","tokens_in":35931,"feed_emoji":"🔗","tokens_out":986,"duration_ms":19191,"temperature":0.7,"pith_summary":"The paper asks when a 3-uniform hypergraph that is only mildly quasirandom—edges are spread roughly evenly—must contain a tight Hamilton cycle, provided its local degrees are not too small. For minimum vertex degree it gives the exact trade-off curve above density 1/3: density d together with normalized degree larger than f(d) is enough, and f(d) is tight. In particular the diagonal point (1/3,1/3) works, settling an open problem. For minimum codegree the diagonal threshold is the larger number κ≈0.3177 solving κ=(1−κ)³, so density and codegree both just above 1/4 do not force the cycle. Both proofs reduce Hamiltonicity to finding one dominant tight component that supplies a local Hamilton framework; the two degree regimes need different component lemmas.","feed_headline":"Tight Hamilton cycles need degree f(d), not just 1/3","feed_subtitle":"Uniform density above 1/3 plus a sharp vertex-degree curve forces the cycle; codegree needs ≈0.3177","key_machinery":"A Hamilton-framework reduction: almost every large induced subgraph must contain a spanning tight component that carries a perfect fractional matching, an aperiodic closed walk, and consistency across overlapping subgraphs; that component is located by two distinct dominant-component lemmas (weighted monochromatic-clique selection for vertex degree; stability around κ plus cherry inequalities for codegree).","core_discovery":"In (n,d,μ)-dense 3-graphs, density d>1/3 and minimum vertex degree strictly above f(d)=(1−√((4d−1)/3))/2 force a tight Hamilton cycle, and the bound is asymptotically sharp; on the codegree side the sharp diagonal is exactly (κ,κ) with κ the unique root of κ=(1−κ)³≈0.3177, so the previously conjectured 1/4 threshold is false.","pith_inferences":["The jump from 1/4 to κ on the codegree diagonal suggests that space barriers and component separation, not mere density, govern the obstruction once pairs rather than vertices are controlled.","Because the weighted-clique step needs monochromatic triangle weight >1/3, the method itself explains why density 1/3 is a natural wall for the vertex-degree argument.","Independent determination of the codegree curve above 1/3 (h2(d)=f(d)²) together with this diagonal result nearly completes the picture except in the sparse regime d≤1/3."],"forward_implications":["The diagonal vertex-degree problem above density 1/3 is settled: any α>1/3 works once d>1/3.","The codegree conjecture that density and codegree >1/4 suffice is false; the correct diagonal constant is κ≈0.3177.","The same dominant-component lemmas supply structural information usable for other spanning problems in linearly quasirandom 3-graphs.","Below density 1/3 the threshold functions h1(d) and h2(d) remain open and are now the natural next targets."],"fun_headline_variants":["Vertex degree above f(d) forces tight Hamilton cycles in dense 3-graphs","Codegree threshold for tight Hamilton cycles is κ≈0.3177, not 1/4","Density d>1/3 plus δ1>f(d) yields tight Hamilton cycles","Sharp diagonal codegree (κ,κ) for tight Hamilton cycles in dense 3-graphs","f(d) is the sharp vertex-degree threshold for tight Hamilton cycles"],"cache_read_input_tokens":128,"weakest_assumption_plain":"The argument treats an external embedding theorem as a black box: once almost every medium-sized subset carries a local Hamilton framework, the whole graph is automatically Hamiltonian.","fun_headline_variants_meta":{"raw":{"variants":["Vertex degree above f(d) forces tight Hamilton cycles in dense 3-graphs","Codegree threshold for tight Hamilton cycles is κ≈0.3177, not 1/4","Density d>1/3 plus δ1>f(d) yields tight Hamilton cycles","Sharp diagonal codegree (κ,κ) for tight Hamilton cycles in dense 3-graphs","f(d) is the sharp vertex-degree threshold for tight Hamilton cycles"]},"model":"grok-4.5","effort":"low","cost_usd":0.005492,"raw_usage":{"total_tokens":1623,"prompt_tokens":1007,"num_sources_used":0,"completion_tokens":100,"cost_in_usd_ticks":54924000,"prompt_tokens_details":{"text_tokens":1007,"audio_tokens":0,"image_tokens":0,"cached_tokens":128},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":516,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":1007,"tokens_out":100,"duration_ms":10190,"temperature":1.0,"reasoning_tokens":516,"cache_read_input_tokens":128,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-30T13:40:23.029607+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Exhibit an infinite family of (n,d,μ)-dense 3-graphs with minimum vertex degree above f(d) (or codegree above κ) that still lack a tight Hamilton cycle, or show that the Lang–Sanhueza-Matamala embedding fails at the error scales used here.","supporting_citations":[],"review_version":1}