{"id":"261501be-7894-4e59-9392-b3c7fb971aa5","arxiv_id":"2507.00431","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A knot K bounds a locally flat Z_d-disc representing a class x of divisibility d in a simply-connected 4-manifold N iff its Arf invariant matches a congruence and b_2(N) dominates all Levine-Tristram signature bounds.","lead":"This paper gives an algebraic test, using only the Arf invariant, the knot signature, and the manifold's intersection form, for when a knot in the 3-sphere boundary bounds a locally flat disc inside a simply-connected 4-manifold with cyclic disc complement. The test also gives the minimal number of S^2 x S^2 stabilizations needed, and it yields topological slice discs in cases where smooth discs are known not to exist.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The even-d case of Theorem 1.1 rests on an unproved relative Bredon theorem (Theorem B.1); Proposition 6.5 is unverified for even d unless that proof is supplied.","rationale":"The reader's weakest-assumption diagnosis is exactly the one I would make: the unproved relative Bredon theorem, Theorem B.1, is the most load-bearing gap in the paper. The paper is otherwise well organized and gives real evidence: the stable results in Theorem 1.10, the freeness and splitting arguments in Sections 4 and 5, the concrete computational examples in Section 1.3, and the authors' transparent Remark 6.6 about the unverified stronger Lee-Wilczyński assertion. None of that, however, supplies the missing proof of the relative Bredon statement, and Theorem 1.1 as stated covers even d. Since the gap is explicit and addressable but not a demonstrated falsehood, the appropriate posture remains a conditional acceptance pending a complete proof of Theorem B.1 and a clean derivation of Lemma 6.4. I therefore recommend no change to the reader's conditional verdict.","tokens_in":34402,"tokens_out":11063,"duration_ms":133703,"concrete_test":"Prove Theorem B.1 in the needed relative setting: X=Σ_d(D), A=∂Σ_d(D), T the order-two deck transformation, and F=eD. Concretely, check whether the relative statement follows from Bredon's absolute theorem applied to the double DX=X∪_A X with the glued involution; verify that the doubled class has nonzero cup-square and that its restriction to the fixed set of DX restricts to k^*(a) on F. If the reduction goes through, the gap is formal and can be closed by a short argument. If not, construct a small 4-dimensional pair (or a d=2 example with the explicit crossing-change disc for the left-handed trefoil in CP^2°) and test Lemma 6.4 directly by computing Q(y,Ty) and Q(y,z) mod 2 on a basis of H_2(Σ_2(D);Z_2); a failure would show Theorem 1.1 currently lacks a proof for even d.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central equivalence for even d depends on the evenness condition of Proposition 3.11(3), proved as Proposition 6.5. For d even, Proposition 6.5 invokes Lemma 6.4, whose proof uses a relative version of Edmonds' Proposition 5.1, stated as Proposition B.2. Proposition B.2 is derived from Theorem B.1, a 'relative variant' of Bredon's theorem whose proof is explicitly omitted: the paper says the adjustment is left to the reader. This is a load-bearing gap, not a cosmetic one. The absolute Bredon-Edmonds statement does not formally imply the relative statement without checking the pair (X,A), the restriction to F∩A, and the relative cup-product evaluation. In the needed setting, X=Σ_d(D), A=∂Σ_d(D), T is the order-two deck transformation, and F=eD; if the relative Bredon theorem fails here, the congruence Q_Σ(y,Ty)=Q_Σ(y,z) mod 2 in Lemma 6.4 can fail, so Proposition 6.5 has no proof and the destabilization step in Proposition 3.11 is unjustified for even d. Remark 6.6 confirms the authors could not verify Lee-Wilczyński's stronger assertion, so the weaker statement used here is not independently established. The odd-d case is unaffected, and the other main ingredients (freeness in Proposition 4.10, splitting in Proposition 5.1, and the extension of Gilmer's inequality in Proposition 3.9) appear sound modulo quoted results.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies when a knot K in the boundary S^3 of a compact, oriented, simply-connected 4-manifold N bounds a locally flat simple slice disc representing a given nonzero class x in H_2(N, ∂N) of divisibility d. Theorem 1.1 states that, under the hypothesis H_1(Σ_d(K)) = 0, such a disc exists exactly when two computable conditions hold: an Arf/Kirby-Siebenmann/signature congruence when x is characteristic, and an inequality comparing b_2(N) with the maximum over j of |σ(N) - (2j(d-j)/d^2)x·x + σ_K(e^{2πij/d})|. Theorem 1.10 characterizes stable representability, and Corollary 1.13 computes the stabilising number when d is a prime power. The proof follows the Lee-Wilczynski strategy: stable embedding via ambient surgery, an algebraic splitting theorem for pointed hermitian forms over Z[Z_d], and then verification of three conditions (freeness, splitting over Z, and evenness). The evenness condition for even d relies on a claimed relative version of a theorem of Bredon and Edmonds, stated in Appendix B.","tokens_in":34654,"tokens_out":6618,"duration_ms":75249,"significance":"If the proof is completed, the result is significant: it gives a parameter-free, computable characterization of when a knot bounds a simple topological slice disc in a prescribed relative homology class, extending the closed-manifold work of Lee and Wilczynski and providing new examples where topological and smooth sliceness diverge (Examples 1.15 and 1.17). The paper is carefully structured and contains several strong elements: the stable surgery argument is detailed, Proposition 3.9 gives a proof of the needed extension of Gilmer's inequality, and Remark 6.6 honestly records that a stronger assertion of Lee-Wilczynski could not be confirmed. However, the even-d case of the main theorem rests on an explicitly unproved relative Bredon theorem, so the central claim is not yet fully established as stated.","major_comments":[{"comment":"The relative Bredon theorem is stated without proof: the text says the adjustment of Bredon's proof to the relative case is 'left to the reader.' This is load-bearing for even d. Proposition 6.5 invokes Lemma 6.4, whose proof uses Proposition B.2, which is derived from Theorem B.1. The absolute Bredon-Edmonds statement does not formally imply the relative statement, since one must check the pair (X,A), the fixed-point restriction to F∩A, and the evaluation of relative cup products; in the needed setting X=Σ_d(D), A=∂Σ_d(D), and F=eD. Without a proof of Theorem B.1, the congruence Q_Σ(y,Ty)=Q_Σ(y,z) mod 2 in Lemma 6.4 is unverified, and consequently the evenness condition and the destabilization step in Proposition 3.11 are not established for even d. Remark 6.6 confirms that the authors could not verify Lee-Wilczynski's stronger assertion, and the weaker statement used here is not independently justified. The odd-d case is unaffected, but Theorem 1.1 as stated covers all d; this gap must be repaired.","section":"Appendix B, Theorem B.1; §6, Proposition 6.5"},{"comment":"The projectivity of H_2(Σ_d(D)) is quoted from [LW90, p. 399], and Remark 4.4 concedes that the argument in [LW90] relies on several unreferenced facts from group cohomology. Since the freeness condition in Proposition 4.10 is one of the three hypotheses needed to apply the splitting argument in Proposition 3.11, the paper should either supply a complete proof of projectivity for the branched covers of discs used here or give a precise, verifiable reference for this relative/disc case. A closed-manifold statement with only a sketch in a remark is not fully satisfactory for a load-bearing step of the main theorem.","section":"§4.2, Proposition 4.3"}],"minor_comments":[{"comment":"There are several typos and spacing issues, such as 'simply-connected4-manifold' in Section 2 and 'continuously' missing spaces elsewhere; these should be corrected in the final version.","section":"Throughout"},{"comment":"The bullet 'The knot K is sliced by a simple disc in N representing x' would read more naturally as 'K bounds a simple slice disc in N representing x'; the current phrasing is grammatically awkward.","section":"Theorem 1.1 statement"},{"comment":"The proof of Claim 1 says the argument for finding u is 'identical to the argument in the closed case from [LW90, page 393]' but gives no details; since this claim is used in the ordinary case of the ambient surgery criteria, a few sentences reproducing the argument would improve readability.","section":"§2.4, Claim 1"},{"comment":"The remark explains why the authors prefer their proof of stable freeness over the Lee-Wilczynski/Wilczynski argument; this is helpful, but the final sentence could be clarified to state precisely which exactness properties are being invoked.","section":"Remark 4.7"},{"comment":"The notation k_*(c) and [F] in the congruence should be explicitly identified: k is the inclusion of the fixed-point set and [F] is the fundamental class of the pair (F,∂F); a short sentence would avoid ambiguity.","section":"Appendix B, Proposition B.2"}],"recommendation":"major_revision","confidential_remarks":"The main unresolved issue is the unproved relative Bredon theorem in Appendix B. If the authors can supply a complete proof, or point to a precise reference that covers the relative case, the paper is likely acceptable. The reliance on [LW90] for projectivity should also be tightened. The paper is otherwise substantial, well-motivated, and within scope for a topology journal."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague — I've read the Conway-Orson-Pencovitch paper. The main theorem is almost certainly right, but the even-d case is not fully proved as written: the proof of the evenness condition rests on a relative Bredon theorem (Appendix B) whose proof the authors explicitly leave to the reader. That is a real, load-bearing gap, not a cosmetic one. The odd-d case and the stable criterion (Theorem 1.10) are in much better shape.\n\nWhat's new: the Z_d classification for general d, the stable necessary-and-sufficient criterion, and the stabilization number formula. The d=1 primitive case recovers KPRT24, which the authors say; the real content is the general case. The conditions are genuinely computable (Arf, signature, b_2), and the examples — trefoil satellites, K3, the CP^2 computations — are well chosen. The proof architecture is clear: stable embedding first, then the Lee-Wilczynski splitting machine, then the three verification sections. I credit the honesty of Remark 6.6, which says they could not confirm Lee-Wilczynski's stronger assertion. The trouble is that the weaker statement they need is also unproved, because it depends on that appendix.\n\nSoft spots, in order of severity. (1) Theorem B.1 is the one that matters. Proposition 6.5 for even d goes through Lemma 6.4, which explicitly invokes the relative Bredon-Edmonds statement. Without a proof of B.1, the destabilization step in Proposition 3.11 is unjustified for even d. There is no contradiction in the mathematics; there is a missing argument. (2) The freeness condition in Section 4 imports projectivity from LW90, and Lemma 4.6 contains a sketch rather than a full proof. It looks plausible, but it is another quotation. (3) The extension of Gilmer's inequality to non-prime-power d in Proposition 3.9 is a sketch; the sign correction they note is consistent with the literature, but the prime-power removal is not fully written out.\n\nOverall: the architecture is sound, there are no fitted parameters or invented entities, and the paper is an honest, substantial contribution. The even-d gap is addressable. I would send this to a serious referee rather than desk-reject, with the expectation that either the appendix gets completed or Theorem 1.1 is stated for odd d and prime-power d, with the even case marked conditional. If you read it, start with Appendix B and the proof of Lemma 6.4; that is where the load-bearing claim lives.","headline":"Even-d case of Theorem 1.1 hinges on an explicitly unproved relative Bredon theorem; odd-d case and stable results look solid.","tokens_in":35270,"tokens_out":2334,"would_cite":true,"duration_ms":26375,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["57K10","57R40","57R65"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves that, when the d-fold branched cover of the knot has trivial first homology, a knot is sliced by a simple disc in a simply-connected 4-manifold representing a given homology class if and only if two computable numerical…","keywords":["slice discs","locally flat embeddings","4-manifolds","Levine-Tristram signature","branched covers","Arf invariant","stabilising number","simply-connected 4-manifolds"],"falsifier":"The claim would be refuted by a concrete pair (N,x,K) with H_1(Sigma_d(K))=0 satisfying the two numerical conditions of Theorem 1.1 for which K nevertheless admits no locally flat simple slice disc in N representing x. A more targeted check: find a 2-torsion class in $H^{2}$(Sigma_d(D)) for an even d where the relative Bredon equality of Proposition B.2 fails, since Proposition 6.5's evenness argument depends on it.","tokens_in":34113,"feed_emoji":"🪢","tokens_out":6567,"duration_ms":62575,"temperature":0.7,"pith_summary":"This paper asks when a knot in the boundary 3-sphere can be filled by a locally flat disc inside a given simply-connected 4-manifold, with the disc's complement having finite cyclic fundamental group. The main theorem answers this exactly, under one assumption: the d-fold branched cover of the knot has first homology zero, where d is the divisibility of the homology class the disc is meant to represent. Sympathetically read, the paper establishes that in this setting, the existence of such a 'simple' slice disc is equivalent to two numerical conditions built from the Arf invariant, the Levine-Tristram signature of the knot, and the signature and Euler data of the 4-manifold. The conditions are computable, so the result turns a geometric existence question into arithmetic.","feed_headline":"Two formulas decide when a knot slices a 4-manifold","feed_subtitle":"Arf invariant, Levine-Tristram signature, and manifold signatures give a complete slice-disc criterion.","key_machinery":"The central object is the pointed hermitian form $(H_2(\\Sigma_d(D)), \\lambda, z)$ over the group ring $\\mathbb{Z}[\\mathbb{Z}_d]$ associated to a disc $D$ in a stabilized manifold; $\\lambda$ is the equivariant intersection form and $z$ is the class of the branch set in the $d$-fold branched cover. The argument runs through Lee-Wilczy\\'nski's splitting theorem, which says that under freeness, signature, and evenness conditions this form splits off hyperbolic summands, allowing surgery to destabilize from $N \\# k(S^2\\times S^2)$ back to $N$. The evenness condition is supplied by the relative Bredon-Edmonds result (Appendix B), and the signature condition is recast by a Rohlin-Viro formula (Lemma 3.8) comparing $j$-signatures of the branched cover with the Levine-Tristram signature of $K$.","core_discovery":"Theorem 1.1 asserts that for a compact, oriented, simply-connected 4-manifold $N$ with boundary $S^3$, a nonzero class $x \\in H_2(N,\\partial N)$ of divisibility $d$, and a knot $K$ with $H_1(\\Sigma_d(K)) = 0$, $K$ is sliced by a simple disc in $N$ representing $x$ if and only if (1) when $x$ is characteristic, $\\mathrm{Arf}(K) + \\mathrm{ks}(N) + \\tfrac{1}{8}(\\sigma(N) - x\\cdot x) \\equiv 0 \\bmod 2$, and (2) $b_2(N) \\geq \\max_{0\\leq j<d} |\\sigma(N) - \\frac{2j(d-j)}{d^2} x\\cdot x + \\sigma_K(e^{2\\pi i j/d})|$. Here $\\Sigma_d(K)$ is the $d$-fold branched cover of the knot, $\\mathrm{ks}(N)$ and $\\sigma(N)$ are the Kirby-Siebenmann invariant and signature of $N$, and $\\sigma_K$ is the Levine-Tristram signature of $K$. The paper also proves a stable version (Theorem 1.10): without the homology condition the same Arf condition is necessary and sufficient for sliceness after connected sum with enough copies of $S^2\\times S^2$, and when $d$ is a prime power the minimal number of stabilizations is given by half the excess of the signature maximum over $b_2(N)$.","pith_inferences":["One could try to remove the hypothesis $H_1(\\Sigma_d(K))=0$: the paper itself notes (Remark 1.8) it is not necessary, and a sharper theorem would presumably replace it by a condition on the linking form of the branched cover.","The relative Bredon gap suggests a natural test: prove or disprove Theorem B.1; if it is false, the even-$d$ case of the main theorem would need a different evenness argument, possibly using equivariant transversality.","The signature inequality in condition (2) resembles Gilmer's inequality but with the knot signature entered with the opposite sign; the paper attributes this to a sign convention and notes it matters, e.g., for the left-handed trefoil. A reader might check the convention against Viro's branched-cover formula in a simple example."],"forward_implications":["When $d=1$ (primitive $x$), condition (2) is automatic and Theorem 1.1 says every knot is sliced by a simple disc in $N$ representing $x$ unless $x$ is characteristic, in which case sliceness is equivalent to $\\mathrm{Arf}(K)+\\mathrm{ks}(N)+\\tfrac{1}{8}(\\sigma(N)-x\\cdot x) \\equiv 0 \\bmod 2$.","For $2$-divisible classes with $|\\det(K)|=1$, sliceness in $(\\mathbb{CP}^2)^\\circ$ is equivalent to $\\sigma(K)\\in\\{0,2\\}$ and in $(\\overline{\\mathbb{CP}}^2)^\\circ$ to $\\sigma(K)\\in\\{-2,0\\}$.","In punctured spin manifolds such as $K3^\\circ$, every knot bounds a locally flat simple disc in every primitive class, in contrast with the smooth category.","When $d$ is a prime power and the stabilising numbers are finite, the simple $(x,N)$-stabilising number equals $\\tfrac{1}{2}(\\max_{0\\leq j<d} |\\sigma(N) - \\frac{2j(d-j)}{d^2}x\\cdot x + \\sigma_K(e^{2\\pi i j/d})| - b_2(N))$, and this equals the ordinary stabilising number."],"supporting_citations":[{"why":"Supplies the stable-embedding and ambient-surgery machinery, including the Freedman-Kirby quadratic form and the framing lemmas, that prove Theorem 1.10.","marker":"[L W90]"},{"why":"Provides the algebraic splitting theorem used to lift hyperbolic summands from the Z-valued form to the group-ring form.","marker":"[L W93]"},{"why":"States the pointed hermitian splitting theorem (Theorem 3.5) that is the core destabilization tool.","marker":"[L W97]"},{"why":"Gives the sphere embedding theorem and the classification input that realize the algebraic splitting by a homeomorphism and prove the d=1 case.","marker":"[Fre82]"},{"why":"Supplies the topological 4-manifold foundations, including surgery and Kirby-Siebenmann additivity used in Proposition 3.11.","marker":"[FQ90]"},{"why":"Gives the signature inequality (Proposition 3.9) that is the necessity of condition (2) and the lower bound for stabilising numbers.","marker":"[Gil81]"},{"why":"Used to identify the Arf invariant of the Freedman-Kirby quadratic form with Arf(K)+ks(N)+(1/8)(sigma(N)-x*x).","marker":"[Klu20]"},{"why":"Source of Proposition B.2, the relative Bredon-Edmonds result on which the evenness condition for even d rests.","marker":"[Edm89]"},{"why":"Gives the absolute Bredon fixed-point theorem whose relative version (Theorem B.1) is stated but left unproved in Appendix B.","marker":"[Bre72]"}],"fun_headline_variants":["Knot slicing conditions tied to Arf and signatures","Criterion for slicing knots in 4-manifolds","Arf and Levine-Tristram signatures decide knot slicing","Stable slicing condition via Arf invariant alone","Exact slice-disc criterion for 4-manifolds with S^3 boundary"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"For even d the proof relies on an unproved relative version of Bredon's fixed-point theorem (Theorem B.1), stated in the appendix as 'left to the reader'; if that relative statement fails for the branched covers used here, the destabilization step and Theorem 1.1 for even d would not follow.","fun_headline_variants_meta":{"raw":{"variants":["Knot slicing conditions tied to Arf and signatures","Criterion for slicing knots in 4-manifolds","Arf and Levine-Tristram signatures decide knot slicing","Stable slicing condition via Arf invariant alone","Exact slice-disc criterion for 4-manifolds with S^3 boundary"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000494,"raw_usage":{"total_tokens":2422,"prompt_tokens":936,"completion_tokens":1486,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":552,"completion_tokens_details":{"reasoning_tokens":1403}},"tokens_in":552,"tokens_out":1486,"duration_ms":10474,"temperature":1.0,"reasoning_tokens":1403,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T21:17:11.234795+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"The claim would be refuted by a concrete pair (N,x,K) with H_1(Sigma_d(K))=0 satisfying the two numerical conditions of Theorem 1.1 for which K nevertheless admits no locally flat simple slice disc in N representing x. A more targeted check: find a 2-torsion class in $H^{2}$(Sigma_d(D)) for an even d where the relative Bredon equality of Proposition B.2 fails, since Proposition 6.5's evenness argument depends on it.","supporting_citations":[],"review_version":1}