{"id":"dceb4f25-6a21-4e50-bead-09617245ac0e","arxiv_id":"2411.12677","paper_version":2,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"For generic metrics, stationary varifolds with only strongly isolated singularities either are smooth or have a more complicated singularity; in codimension one, only smooth or non-strongly-isolated objects persist.","lead":"For a generic small perturbation of the round metric on a four-sphere, the infinitely many Hsiang minimal hyperspheres cannot persist as continuous families below an explicit area threshold, and stationary varifolds with simple isolated cone singularities are forced to be smooth.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Countable covering in arbitrary codimension (Thm 4.23/App. E) is the load-bearing unverified step; if the adaptation of Edelen's cone decomposition fails, the generic nonnegativity of the index collapses.","rationale":"The paper's central claim rests on two largely independent pillars. The exact index formula (Theorem 3.2) is derived explicitly from Lockhart-McOwen theory and appears internally consistent, with a nontrivial check in the Clifford football example. The genuinely load-bearing point is the generic nonnegativity of the Fredholm index (Theorem 4.1), whose proof requires the local Sard-Smale theorem (Theorem 4.22) and, crucially, the countable covering of the space of metric-MSI pairs by canonical pseudo-neighborhoods (Theorem 4.23). The reader identified exactly this covering/parametrization as the weakest assumption, and my review agrees. Appendix E contains the only major step that simply asserts a verbatim adaptation of external machinery: Theorem E.9 is said to follow verbatim from [18, Theorem 7.1] with only Case 1 arising, and Theorem E.15 is 'essentially the same' as [29, Theorem 9.6]. In higher codimension the absent order structure and the need to handle unstable cones are acknowledged but not fully worked out in the text. A failure of the finite branching or countability in this adaptation would invalidate the Baire-category passage from local to global genericity, leaving no proof of Theorem 4.1. This is a verification gap rather than a demonstrated contradiction, so the appropriate verdict is conditional acceptance pending a complete, higher-codimension proof of Theorem E.9/E.15.","tokens_in":76249,"tokens_out":37541,"duration_ms":377299,"concrete_test":"Independently verify Theorem E.9 for a concrete higher-codimension case with an unstable link, e.g. a sequence of 2-dimensional MSI in R^5 converging to the cone over RP^2 ⊂ S^4. Check that the induction in [18, Theorem 7.1] requires no maximum principle and that the tree of strong-cone/smooth regions has uniformly bounded branching, with only Case 1 arising. If a new case appears or the tree branching is unbounded, the countable cover in Theorem 4.23 is incomplete and the proof of Theorem 4.1 collapses.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central generic conclusion Theorem 4.1 depends on two pillars: the index formula (Thm 3.2), which appears sound, and the Baire-generic nonnegativity of the augmented Jacobi index, built on the local Sard-Smale theorem (Thm 4.22) and its global assembly via the countable covering Theorem 4.23. The reader's weakest-assumption diagnosis is correct: Theorem 4.23 is proved in Appendix E by adapting Edelen's hypersurface decomposition to arbitrary codimension and to unstable, infinite-index cones. This adaptation is not a verbatim extension: in higher codimension there is no scalar maximum principle or order structure, and the paper itself notes that new techniques are required for the combinatorics of cascades (Section 1.4, Appendix E). Yet the key local decomposition Theorem E.9 is dismissed as following 'verbatim' from [18, Theorem 7.1], with only the assertion that 'only Case 1' arises, and Theorem E.15 is 'essentially the same' as [29, Theorem 9.6]. If the finite branching of the decomposition tree, the compactness of smooth models, or the countability of the pseudo-neighborhood cover fails in higher codimension, the Baire-category argument producing a generic metric has no basis. This is a load-bearing technical premise distinct from the index-counting formula itself, and a failure here would not merely weaken quantitative bounds but would invalidate the generic regularity theorem.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves a generic regularity theorem for stationary integral n-varifolds with only strongly isolated singularities in N-dimensional closed Riemannian manifolds, with no restriction on codimension. The main result, Theorem 1.1, asserts that for a Baire-generic metric, every such varifold is either smooth, or has a non-strongly-isolated singularity, or has only strongly isolated singularities whose links have Morse index exactly N; in codimension one the third alternative is impossible. The proof has two pillars: an exact Fredholm index formula for the Jacobi operator on augmented weighted Sobolev spaces (Theorem 3.2), and a Baire-generic nonnegativity statement for that index (Theorem 4.1), obtained by adapting White's Sard-Smale approach through a countable covering of the space of metric-MSI pairs by canonical pseudo-neighborhoods (Theorem 4.23). The paper also derives applications to persistence questions for the Clifford football and Hsiang's hyperspheres in nearly round four-spheres.","tokens_in":76522,"tokens_out":5974,"duration_ms":66726,"significance":"If the technical framework is complete, this is a substantial advance: it removes the codimension-one restriction from generic regularity arguments for singular minimal varieties and gives a genuinely new index-theoretic mechanism, the formula in Theorem 3.2 relating the Fredholm index of the augmented Jacobi operator to the effective Morse indices of the links. The geometric consequences are striking and clearly explained: for nearly round metrics on S^4, the Clifford football and all but finitely many Hsiang hyperspheres cannot persist, and generic finiteness holds under a sharp area threshold. The presentation of Section 3 is clean and the use of Lockhart-McOwen theory is well motivated. The main weakness is that the higher-codimension adaptation of Edelen's cone-decomposition and covering machinery, on which Theorem 4.1 depends, is asserted rather than proved in full in Appendix E.","major_comments":[{"comment":"The proof of the local cone decomposition is not actually supplied. The text says only that the proof follows verbatim from [18, Theorem 7.1] and that 'only Case 1' arises in the induction. In arbitrary codimension there is no scalar ordering of normal graphs and no maximum-principle argument of the type used in Edelen's hypersurface setting, so it is not evident that the cascade combinatorics, the finite branching of the decomposition tree, and the compactness of the smooth models survive without modification. This is not a cosmetic issue: Theorem E.9 feeds directly into Theorem E.15 and Proposition E.16, hence into the countable covering Theorem 4.23, which is the basis for the Baire-category argument proving Theorem 4.1. A failure of this decomposition would invalidate the generic regularity theorem, not merely weaken a quantitative bound. The authors should give a complete proof of the local decomposition in arbitrary codimension, or state and prove a precise modified theorem with all additional hypotheses explicitly verified.","section":"Appendix E, Theorem E.9"},{"comment":"Theorem E.15 is asserted to be 'essentially the same' as [29, Theorem 9.6], and Proposition E.16 is said to follow the arguments in [29, Subsection 9.2], but [29] concerns the hypersurface case, whereas the present paper explicitly warns in Section 1.4 that several changes and adaptations are needed in arbitrary codimension. In particular, the countability of the β-close tree representations and the sequential compactness of the intermediate neighborhoods L_0^{k,α}, which are exactly the steps needed to pass from the countable cover by intermediate neighborhoods to the countable cover by the canonical neighborhoods of Definition 4.19, are not written out. Since this covering is the load-bearing premise for the final Baire-category conclusion in the proof of Theorem 4.1, this is a second gap that must be filled before the main theorem can be considered fully established.","section":"Appendix E, Theorem E.15 and Proposition E.16"}],"minor_comments":[{"comment":"In the paragraph on Hsiang's hyperspheres, 'refereed to as Clifford football' should be 'referred to as Clifford football'.","section":"§1.1"},{"comment":"The sentence 'Let δ0 ∈ (1/4) be the dimensional constant determined in Lemma D.1' appears to contain a typo; it should presumably be δ0 ∈ (0, 1/4).","section":"Remark 2.18"},{"comment":"The notation for smooth models is inconsistent: Definition E.3 defines '(Λ, σ, γ)-smooth models', while Theorem E.15 refers to '(Λ, σ, β)-smooth models'; the parameter names should be aligned throughout Appendix E.","section":"Definition E.3 / Theorem E.15"},{"comment":"Theorem 4.1 is stated for stationary integral n-varifolds, but the parametrization space M_n^{k,α}(M) in Definition 4.10 is restricted to connected Σ; the proof should state explicitly that a disconnected MSI is treated componentwise, so that the countable covering argument applies to each connected component.","section":"Theorem 4.1 / Definition 4.10"}],"recommendation":"major_revision","confidential_remarks":"The paper is a strong candidate once the Appendix E gap is addressed. I would not reject it: the index formula and the geometric applications are substantial and clearly presented. However, the countable covering theorem is the keystone of the global generic result, and the current text does not provide a complete proof of its adaptation to arbitrary codimension. The authors should be asked to expand Appendix E substantially, either by giving full proofs of the decomposition and compactness statements or by stating precisely which external theorems are being used and why they apply verbatim to the present setting."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The main thing you should know: the index-counting formula (Theorem 3.2) is solid and new, and it does real work. The paper deserves a serious referee, not a desk reject. The part I would want verified line-by-line is Appendix E, and the authors are honest that it is an adaptation, not a quote.\n\nThe genuinely new content is the exact Fredholm index formula for the Jacobi operator of a stationary varifold with strongly isolated singularities: minus the sum of effective Morse indices of the links. That is clean, self-contained modulo Lockhart–McOwen theory, and I found no circularity. The generic non-negativity theorem (Theorem 4.1) is the high-stakes extension of White/Edelen to arbitrary codimension and unstable cones. The applications to the Clifford football and Hsiang hyperspheres are real corollaries, not window dressing.\n\nThe soft spot is exactly where the reader put it. Theorem 4.23 (countable covering by canonical pseudo-neighborhoods) is proved in Appendix E by adapting Edelen's cone decomposition. The paper itself says Theorem E.9 follows \"verbatim\" but only Case 1 arises, and Theorem E.15 is \"essentially the same\" as a prior theorem. Yet the introduction warns that new techniques are needed for the combinatorics of cascades because the normal bundle has no order structure. Those two statements sit in some tension. If the finite branching, compactness of smooth models, or countability of the cover fails in higher codimension, the Baire-category argument for generic non-negativity has no foundation. That is a load-bearing premise, not a cosmetic gap.\n\nBut note: this is a failure mode, not a demonstrated error. The index formula does not depend on the covering argument. The generic sign theorem does, and a referee needs to check the adaptation carefully. I would not reject the paper for that; I would send it to a referee who knows Edelen's construction and ask for a verdict on Appendix E.\n\nWho is this for? Geometric measure theorists working on generic regularity, and anyone studying persistence of singular minimal submanifolds. The paper is long and technical, but the main theorem and the index formula are worth the investment. Yes, it deserves peer review.","headline":"Strong paper with a clean index formula and a substantial generic-regularity theorem; the load-bearing countable-covering step in arbitrary codimension is the piece to have a referee check carefully.","tokens_in":77055,"tokens_out":1325,"would_cite":true,"duration_ms":18505,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["49Q20","53C42","58J20","58E12"],"pacs":[],"model":"deepseek-v4-flash","headline":"For a generic background metric, stationary integral submanifolds cannot have only strongly isolated singularities unless a very special index condition holds; in codimension one, only non-isolated singularities can persist.","keywords":["generic regularity","stationary varifolds","strongly isolated singularities","Fredholm index","Jacobi operator","minimal hypersurfaces","Morse index","nearly round four-sphere"],"falsifier":"Construct an open set of smooth metrics arbitrarily close to the round metric on the four-sphere such that every metric in the set admits a stationary integral 3-varifold with exactly two strongly isolated singularities modelled on the cone over the minimal torus in the three-sphere; the theorem predicts such an open family cannot exist. Equivalently, find a metric for which the Jacobi operator of such a varifold has negative index and show that negativity persists on an open neighborhood of that metric in the smooth topology.","tokens_in":76017,"feed_emoji":"","tokens_out":10096,"duration_ms":95028,"temperature":0.7,"pith_summary":"Generically in the space of Riemannian metrics, singularities of stationary integral submanifolds cannot be of the simplest kind: a generic metric forces every stationary integral n-varifold with n at least 2 to be either smooth, or to possess at least one singular point that is not strongly isolated. In codimension one this becomes an unconditional alternative—smooth or with a non-isolated singularity—so that only \"more complicated\" singularities may persist. The proof centres on an index formula: the Fredholm index of the Jacobi operator on augmented weighted Sobolev spaces equals minus the sum of the effective Morse indices of the conical links at the singular points, and for generic metrics this index is non-negative. A reader should care because this settles a long-standing question in the negative: in nearly round four-spheres, the infinite family of minimal hyperspheres and their singular limit are generically non-persistent, and only finitely many small minimal hypersurfaces survive.","feed_headline":"Typical metrics forbid isolated singularities","feed_subtitle":"An exact index formula makes isolated conical singularities nongeneric, limiting minimal spheres near the round four-sphere.","key_machinery":"The central object is the Jacobi (stability) operator $L_{\\Sigma,g}$ of a stationary integral submanifold with only strongly isolated singularities, acting on weighted Sobolev spaces whose weights are measured by distance to the singular points, augmented by finitely many 'translation-like' sections that encode the singular points. The load-bearing identity is the index-counting formula of Theorem 3.2: $$\\hat{\\mathrm{index}}_\\tau(L_{\\Sigma,g}) = -\\sum_{p\\in\\operatorname{Sing}\\Sigma} I(C_p),$$ with $I(C) = \\operatorname{index}(L_{C\\cap S^{N-1}}) - N$. This formula converts the spectral Morse-index data of the conical links into a Fredholm index, which is then shown to be non-negative for a generic metric via a local Sard–Smale theorem and a countable cover of the space of metric–submanifold pairs by canonical neighborhoods.","core_discovery":"The paper's central claim is a residual-set regularity theorem: for a countable-intersection-of-open-dense set of smooth metrics on a closed N-manifold, every stationary integral n-varifold with 2 ≤ n < N satisfies one of three alternatives—it is entirely smooth; it has a singular point that is not strongly isolated; or all its singularities are strongly isolated and every link has Morse index exactly N. In the hypersurface case N = n+1 the third alternative cannot occur, so generically any stationary integral varifold is smooth or has a non-strongly-isolated singularity. The engine is an exact index formula identifying the Fredholm index of the Jacobi operator acting on augmented weighted Sobolev spaces as the negative of the sum of the links' effective Morse indices, combined with a theorem asserting that this index is non-negative for generic metrics. From this the dichotomy follows, and with it the generic finiteness of closed minimal hypersurfaces of area below 4π²−ε in nearly round four-spheres.","pith_inferences":["A likely sharpness statement, implicit in the paper's remarks, is that the 4π² threshold is optimal: for generic nearly round metrics one expects finitely many but arbitrarily many minimal hyperspheres with areas accumulating at 4π²; a numerical search could test this conjecture.","The index-gap principle suggests a transferable recipe: whenever a geometric variational problem has conical singularities and a computable Fredholm index, the sign of that index should control which singularity types are generically persistent.","The covering theorem and local Sard–Smale argument may adapt to stationary varifolds with higher-dimensional singular strata by replacing link Morse indices with a normal-index of the stratum, possibly yielding generic regularity for more general singular sets.","One could attempt to push the method from smooth metrics to metrics of low regularity, or to other ambient geometries such as manifolds with boundary, using the same weighted Sobolev framework."],"forward_implications":["In codimension one, generic metrics make the class of stationary integral varifolds either smooth or singular with a non-isolated singularity, so any isolated conical point is generically unstable.","For every ε > 0, a generic metric near the round one on the four-sphere contains only finitely many closed embedded minimal hypersurfaces of area below 4π² − ε.","The infinite family of embedded minimal hyperspheres in round four-sphere geometry—and its singular limit, the two-point suspension of a minimal torus—cannot be continuously deformed into nearby metrics as minimal hypersurfaces of area below 4π² − ε; at most finitely many members of the family can persist.","If an isolated conical singularity is to survive a generic perturbation, its link must have Morse index exactly equal to the ambient dimension N; the paper identifies candidate examples for which this equality is open.","The index formula gives a new obstruction: any regular minimal cone with nonzero effective Morse index cannot appear as the sole singularity type of a stationary varifold in a generic metric."],"supporting_citations":[{"why":"Supplies uniqueness of the tangent cone at a strongly isolated singularity and the resulting finiteness of singular points.","marker":"[40]"},{"why":"Provides the weighted Fredholm index theory on conically singular manifolds used in the index computation.","marker":"[31]"},{"why":"Gives the conifold weighted Sobolev-space framework that defines the operator's domain.","marker":"[37]"},{"why":"Constructs canonical neighborhoods and a countable covering in the hypersurface case, adapted here to arbitrary codimension.","marker":"[18]"},{"why":"Supplies the local Sard–Smale style technology for meagerness of bad singularities, extended to unstable cones.","marker":"[29]"},{"why":"The classical generic-metric regularity argument that this paper recasts in a Banach-manifold-free setting.","marker":"[49]"},{"why":"Supplies the regularity and compactness theorems used throughout for convergence and graphicality.","marker":"[2]"},{"why":"The motivating example of infinitely many minimal hyperspheres in the round four-sphere whose persistence is answered negatively.","marker":"[23]"},{"why":"Provides the mod-2 cyclicity property that turns an area bound into the strongly-isolated structure.","marker":"[50]"},{"why":"Supplies the degeneracy analysis for converging minimal hypersurfaces used in the finiteness proof.","marker":"[39]"}],"fun_headline_variants":["Isolated singularities fade under generic metrics","Generic metrics rule out isolated singular points","Index formula dooms isolated singularities generically","No isolated singularities for generic metrics","Generic metrics make isolated singularities vanish"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole genericity conclusion rests on the existence of a countable family of well-controlled model neighborhoods covering every possible minimal submanifold with only strongly isolated singular points; if that covering cannot be built for arbitrary codimension and unstable cones, the non-negativity of the index for generic metrics collapses.","fun_headline_variants_meta":{"raw":{"variants":["Isolated singularities fade under generic metrics","Generic metrics rule out isolated singular points","Index formula dooms isolated singularities generically","No isolated singularities for generic metrics","Generic metrics make isolated singularities vanish"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000798,"raw_usage":{"total_tokens":3553,"prompt_tokens":1027,"completion_tokens":2526,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":643,"completion_tokens_details":{"reasoning_tokens":2462}},"tokens_in":643,"tokens_out":2526,"duration_ms":17546,"temperature":1.0,"reasoning_tokens":2462,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T17:15:53.490308+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Construct an open set of smooth metrics arbitrarily close to the round metric on the four-sphere such that every metric in the set admits a stationary integral 3-varifold with exactly two strongly isolated singularities modelled on the cone over the minimal torus in the three-sphere; the theorem predicts such an open family cannot exist. Equivalently, find a metric for which the Jacobi operator of such a varifold has negative index and show that negativity persists on an open neighborhood of that metric in the smooth topology.","supporting_citations":[{"cited_title":"Simon, Asymptotics for a class of nonlinear evolution equations, w ith applications to geometric problems , Ann","cited_arxiv_id":null,"evidence_quote":"Supplies uniqueness of the tangent cone at a strongly isolated singularity and the resulting finiteness of singular points."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the weighted Fredholm index theory on conically singular manifolds used in the index computation."},{"cited_title":"Pacini, Desingularizing isolated conical singularities: Uniform estimates via weighted sobolev spaces , Comm","cited_arxiv_id":null,"evidence_quote":"Gives the conifold weighted Sobolev-space framework that defines the operator's domain."},{"cited_title":"Edelen, Degeneration of 7-dimensional minimal hypersurfaces whic h are stable or have a bounded index , Arch","cited_arxiv_id":null,"evidence_quote":"Constructs canonical neighborhoods and a countable covering in the hypersurface case, adapted here to arbitrary codimension."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The classical generic-metric regularity argument that this paper recasts in a Banach-manifold-free setting."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the regularity and compactness theorems used throughout for convergence and graphicality."},{"cited_title":"Hsiang, Minimal cones and the spherical Bernstein problem","cited_arxiv_id":null,"evidence_quote":"The motivating example of infinitely many minimal hyperspheres in the round four-sphere whose persistence is answered negatively."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the mod-2 cyclicity property that turns an area bound into the strongly-isolated structure."},{"cited_title":"Sharp, Compactness of minimal hypersurfaces with bounded index , J","cited_arxiv_id":null,"evidence_quote":"Supplies the degeneracy analysis for converging minimal hypersurfaces used in the finiteness proof."}],"review_version":1}