{"id":"44acd007-062c-4696-a46c-98c61d89e5dd","arxiv_id":"2411.17633","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A BV function is minimally singular iff its singular vertical distance to some point vanishes almost everywhere, and over a wide class of open connected domains this property is equivalent to rigidity for Steiner's inequality.","lead":"This paper introduces a new geometric quantity, the singular vertical distance, and a new class of functions called minimally singular BV functions. It uses these tools to characterize when equality in Steiner's perimeter inequality forces the extremal set to be a vertical translation of the symmetric set.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 1.9's 'rigidity ⇒ minimally singular' uses Proposition 1.6 as a bijection between extremals and admissible barycenters; as stated it is only an if-and-only-if for a fixed E, so non-constant admissible b may never be realized by any E.","rationale":"The reader correctly identified Proposition 1.6 as the weakest link, but the sharper issue is logical rather than merely a matter of verifying hypotheses: the deduction in Theorem 1.9 Step 1 requires the class of admissible barycenter functions to be exactly the set of barycenters of elements of MΩ(v). Proposition 1.6 as printed does not assert this surjectivity, and no proof is given. The central Theorem 1.8, which characterizes minimally singular functions via the singular vertical distance, appears internally coherent and is supported by the detailed construction of Lemma 4.5; my concern does not target that theorem. However, the paper's stated goal includes Theorem 1.9, and that application depends on the missing surjectivity. If the original results in [6] contain the missing construction, the concern is resolved; otherwise Theorem 1.9 should be weakened or supplemented. Because this is a checkable gap rather than a demonstrated falsehood, a conditional verdict is appropriate, consistent with the reader's assessment but for a more specific reason.","tokens_in":44487,"tokens_out":10043,"duration_ms":92222,"concrete_test":"Check [6, Theorems 1.7 and 1.9] for an explicit construction: for every b ∈ GBV(Ω) satisfying (1.18)–(1.20), is the set E_b := {(x,t) ∈ Ω×R : |t − b(x)| < v(x)/2} in MΩ(v) with barycenter b? If the construction requires an assumption beyond (1.16), Theorem 1.9 Step 1 fails. A minimal finite-dimensional test: take n = 1 and v with a single jump, so v is not minimally singular; write down the non-constant admissible b and verify that the associated E_b has P(E_b; Ω×R) = P(F[v]; Ω×R), thereby realizing the predicted non-constant barycenter as a genuine extremal. If no such E_b exists, the characterization is false; if it does, the surjectivity step is validated in the nontrivial case.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The most load-bearing gap is not merely that Proposition 1.6 is imported without proof; it is that the proof of Theorem 1.9 (Section 6, Step 1) silently upgrades Proposition 1.6 from a characterization of a fixed extremal to a surjective correspondence between MΩ(v) and the class of admissible barycenter functions. From rigidity over Ω we only learn that every E ∈ MΩ(v) has constant b_E,Ω. To conclude that (1/2)v is minimally singular, one must know that for every b ∈ GBV(Ω) satisfying (1.18)–(1.20), there exists E ∈ MΩ(v) with b_E,Ω = b. Proposition 1.6, as stated, supplies no such existence statement: it says 'E ∈ MΩ(v) if and only if the barycenter of that same E satisfies ...', not 'for every admissible b there is an E with that barycenter'. Unless the 'careful inspection' of [6, Theorems 1.7 and 1.9] establishes this surjectivity under exactly hypothesis (1.16), the implication rigidity ⇒ minimally singular does not follow. The reverse implication is safe: if v is minimally singular and E ∈ MΩ(v), Proposition 1.6 forces b_E,Ω to be constant. Thus Theorem 1.9, the flagship application, currently rests on an unstated construction of extremals with prescribed barycenter.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a class of functions of bounded variation called minimally singular functions on an open connected domain Ω of finite measure. A function u∈BV(Ω) is minimally singular if every GBV function b with zero approximate gradient, jump controlled by [u], and Cantor part controlled by D^c u must be constant. The author defines a pseudometric, the singular vertical distance SVD_{u,Ω}, by taking infima of the singular variation of restrictions of u to polygonal chains, and proves (Theorem 1.8) that u is minimally singular if and only if there is a point x̄ such that SVD_{u,Ω}(x̄,x)=0 for almost every x. This is then applied to the localized rigidity problem for Steiner's perimeter inequality: over a domain Ω satisfying (1.16), rigidity over Ω is claimed to be equivalent to minimal singularity of v (Theorem 1.9). The paper also discusses when such a domain exists (Lemma 1.11, Proposition 1.12, Lemma 1.13) and records two open problems about the new class.","tokens_in":44786,"tokens_out":10480,"duration_ms":98354,"significance":"If correct, Theorem 1.8 is a genuine geometric characterization of a new class of BV functions, and Theorem 1.9 would be the first characterization of localized rigidity for Steiner's inequality under hypothesis (1.16). The method is original: it adapts Vol'pert's theory of one-dimensional restrictions to a family of polygonal chains and uses the resulting singular variation as a pseudometric. The technical scaffolding in Propositions 3.4 and 3.5 is substantial, and the paper is commendably explicit about its limitations, especially Proposition 1.12 and Open Problems 4.9–4.10. However, the application to rigidity currently depends on an imported proposition and, more seriously, on a surjectivity step that the stated Proposition 1.6 does not provide. The central geometric characterization is promising, but the bridge from equality cases to arbitrary admissible barycenter functions must be made rigorous before the main rigidity theorem can be accepted.","major_comments":[{"comment":"The implication (i)⇒(ii) is not justified. Rigidity over Ω gives, for every E∈MΩ(v), a constant barycenter b_E,Ω. To conclude that (1/2)v is minimally singular, one must know that every admissible b∈GBV(Ω) satisfying (1.18)–(1.20) is realized as b_E,Ω for some E∈MΩ(v). Proposition 1.6, as stated in Section 1.4, is an equivalence for a fixed E and its own barycenter; it does not assert this surjectivity. The phrase \"Thanks to the generality of E∈MΩ(v)\" does not bridge the gap. If the \"careful inspection\" of [6, Theorems 1.7 and 1.9] establishes the surjectivity under hypothesis (1.16), this should be stated and proved; otherwise the main application remains incomplete.","section":"Section 6, Step 1 (proof of Theorem 1.9)"},{"comment":"This proposition is load-bearing and is imported without proof. The paper says only that \"a careful inspection of the proofs\" of [6, Theorems 1.7 and 1.9] leads to it, but the localized hypotheses (1.16) and the appearance of Ω in all four conditions are not shown to follow from the original statements. Since Theorem 1.9 inherits every hidden assumption from this proposition, a complete proof or a precise derivation from [6] is required. This is especially important because the constants in (1.19) and (1.20) differ from those in Definition 1.7 after the factor 1/2, and the reader must be able to verify the compatibility.","section":"Section 1.4, Proposition 1.6"}],"minor_comments":[{"comment":"Definition 4.1 defines SVD_{u,Ω} only for points in tildeΩ_u, while Theorem 1.8 asserts existence of x̄∈Ω with SVD_{u,Ω}(x̄,x)=0 for almost every x. The proof begins with x̄∈tildeΩ_u, and Remark 4.2 suggests an extension. Please reformulate the theorem on tildeΩ_u or explicitly invoke Remark 4.2 so that the statement is well-posed for arbitrary x̄∈Ω.","section":"Theorem 1.8 / Definition 4.1"},{"comment":"The displayed identity \"H^{n-1}({v∧ = 0} \\setminus {v∧ = 0}) = 0\" is evidently a typo; the first set should be the boundary set appearing in (1.26), namely ∂{v∧>0} \\setminus {v∧=0}. Please correct.","section":"Lemma 1.13"},{"comment":"There are several spelling inconsistencies: \"Schwartz\" should be \"Schwarz\" in the introduction and references, and \"Radon–Nykodim\" in Section 2.2 should be \"Radon–Nikodym\".","section":"Throughout"},{"comment":"The informal sentence \"if {v∧>0} ⊂ R^n is H^{n-1}-equivalent to an open and connected set Ω it can be proved that rigidity over R^n holds if and only if v∈BV(Ω) is minimally singular\" would benefit from a precise hypothesis or a forward reference to Theorem 1.9.","section":"Section 1.5"},{"comment":"The quantifier \"for L1-a.e. M>0\" appears after the integral equality; please state clearly that (1.20) is required to hold for every bounded Borel set B and for L1-a.e. M>0.","section":"Proposition 1.6(iv)"}],"recommendation":"major_revision","confidential_remarks":"This is a serious paper with a novel construction, and the central geometric characterization may well be correct. The skeptical objection in the reader's report is accurate: the implication rigidity ⇒ minimal singularity in Theorem 1.9 requires a surjectivity statement for barycenter functions that Proposition 1.6, as stated, does not contain. I recommend major revision rather than rejection because the missing statement may be recoverable from the proofs in [6] or by a direct construction. Please ask the author to provide a complete proof or precise reference for Proposition 1.6 and to address the surjectivity point explicitly."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The thing to know: this paper contains a real geometric characterization — Theorem 1.8 — that seems plausible and rests on genuinely new ideas (minimally singular functions, singular vertical distance, and a Vol'pert-style restriction of BV functions to polygonal chains). The proof is long and technical, and I did not verify every line, but the structure is coherent and the lemmas (especially Proposition 3.4 and Lemma 4.5) are worked out in detail. The author also honestly flags two open problems and, in Section 1.6, admits that the rigidity application does not cover all cases (Proposition 1.12). That is good mathematical citizenship.\n\nThe soft spot is central and, in my view, real. Theorem 1.9 claims that rigidity over Ω is equivalent to v being minimally singular. The easy direction is fine: if v is minimally singular and E is an extremal, then Proposition 1.6 forces b_E,Ω to satisfy (1.18)–(1.20), hence b_E,Ω is constant, giving rigidity. The problem is the other direction. From rigidity you only know that every barycenter actually realized by an extremal is constant. To conclude that 1/2 v is minimally singular, you need that every admissible b — every GBV function satisfying (1.18)–(1.20) — is realized as b_E,Ω for some E ∈ MΩ(v). Proposition 1.6, as stated, does not give that surjectivity; it is an if-and-only-if for a fixed E, not an existence statement. The proof of Theorem 1.9, Step 1, silently assumes that surjectivity. Unless a careful reading of [6, Theorems 1.7 and 1.9] actually implies it, the implication rigidity ⇒ minimally singular does not follow.\n\nA secondary concern is that Definition 1.7 is crafted exactly so that “every admissible b is constant,” which makes Theorem 1.9 almost a restatement of Proposition 1.6 plus surjectivity. The real content, again, is Theorem 1.8. I don’t think this is a fatal flaw in the whole paper — Theorem 1.8 stands on its own — but it means the advertised rigidity characterization is not proven as written.\n\nThe audience is researchers in geometric measure theory and BV theory, especially people working on symmetrization. I would send it to a serious referee, but the referee should insist on either a proof of the surjectivity statement or a corrected version of Theorem 1.9. If the author can fix that, this is a solid contribution.","headline":"Theorem 1.8 (the SVD characterization of minimally singular functions) looks like a genuine new result with a substantial proof, but the flagship application, Theorem 1.9, has a surjectivity gap: as written, rigidity over Ω does not imply minimal singularity.","tokens_in":45342,"tokens_out":2644,"would_cite":true,"duration_ms":27166,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["26B30","49Q20","28A75"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves that a function of bounded variation is minimally singular exactly when a new curve-based pseudometric, the singular vertical distance, vanishes from some point to almost every other point, and uses this to characterize…","keywords":["minimally singular functions","singular vertical distance","Steiner symmetrization","rigidity of perimeter inequality","functions of bounded variation","barycenter function","essential connectedness","equality cases"],"falsifier":"Compute the singular vertical distance for the Cantor function on $(0,1)$: the function $b(t)=|D^s u|((0,t))$ is nonconstant and satisfies conditions (1.22)-(1.24), so $u$ is not minimally singular; the theorem then forces $\\operatorname{SVD}_{u,(0,1)}(\\bar x,x)>0$ for positive-measure pairs, and a direct chain computation should confirm this. A computation giving $\\operatorname{SVD}=0$ almost everywhere for this $u$, or in higher dimensions any $u$ with $\\operatorname{SVD}=0$ almost everywhere that still admits a nonconstant controlled $b$, would refute Theorem 1.8.","tokens_in":44237,"feed_emoji":"📏","tokens_out":6015,"duration_ms":57814,"temperature":0.7,"pith_summary":"The paper tries to identify exactly when Steiner symmetrization has no nontrivial competitors: when the only v-distributed sets with the same perimeter as the Steiner symmetric set are vertical translations of it. It introduces a new class of functions of bounded variation, called minimally singular, and proves that on an open connected domain of finite measure, u is minimally singular precisely when a geometric quantity, the singular vertical distance from one fixed point to almost every other point, is zero. The author then uses this to characterize rigidity of equality cases for Steiner's perimeter inequality over such domains: rigidity over Ω holds if and only if v is minimally singular in Ω. If correct, this converts an analytic rigidity question into a curve-counting condition on the singular part of v's derivative.","feed_headline":"One pseudometric decides rigidity in Steiner's perimeter inequality","feed_subtitle":"A BV function is minimally singular exactly when its singular vertical distance vanishes from one point to almost every other.","key_machinery":"The named central object is the singular vertical distance $\\operatorname{SVD}_{u,\\Omega}$, defined in Definition 4.1 as a pseudometric on the good points of $\\Omega$: it records, for two points $x_1,x_2$, the infimum of $|D^s u_\\gamma|(I_\\gamma^\\circ)$ over all polygonal chains $\\gamma\\in\\Gamma_\\Omega(u)$ connecting them, where $u_\\gamma$ is the restriction of $u^\\wedge$ to the chain. The companion notion is minimal singularity: a BV function $u$ is minimally singular when the only GBV functions whose singular behaviour is controlled by $D^s u$ are constants. The theorem that carries the application is Theorem 1.9, which identifies rigidity over $\\Omega$ for Steiner's inequality with minimal singularity of $v$, using the established fact that equality cases in $\\mathcal M_\\Omega(v)$ are described by a barycenter function $b_{E,\\Omega}$ whose approximate gradient vanishes, whose jumps are bounded by $\\frac12[v]$, and whose Cantor part is controlled by $D^c v$.","core_discovery":"The central result characterizes minimal singularity geometrically. For an open connected Ω with $\\mathcal L^n(\\Omega)<\\infty$ and $u\\in BV(\\Omega)$, define $\\operatorname{SVD}_{u,\\Omega}(\\bar x,x)$ as the infimum of $|D^s u_\\gamma|(I_\\gamma^\\circ)$ over polygonal chains $\\gamma\\subset\\Omega$ joining $\\bar x$ to $x$, where $u_\\gamma(t)=u^\\wedge(\\gamma(t))$. Theorem 1.8 states that $u$ is minimally singular if and only if some $\\bar x$ has $\\operatorname{SVD}_{u,\\Omega}(\\bar x,x)=0$ for $\\mathcal L^n$-a.e. $x$. Minimal singularity means: every $b\\in GBV(\\Omega)$ with approximate gradient zero, jump $[b]\\le [u]$, and Cantor part controlled by $D^c u$ must be constant. Theorem 1.9 translates this into Steiner rigidity: under the domain condition (1.16), rigidity over $\\Omega$ holds if and only if $v$ is minimally singular in $\\Omega$; with a global-support condition, local rigidity upgrades to global rigidity.","pith_inferences":["The singular vertical distance pseudometric could be tested numerically on piecewise-linear approximations: if a BV function is approximated by functions with small singular variation along sampled chains, rigidity may be certified in applications where exact solutions are unavailable.","The definition of minimal singularity is not tied to Steiner symmetrization; analogous barycenter rigidity problems for spherical, Schwarz, or Gaussian symmetrization may admit the same curve-variation criterion.","A positive answer to the paper's Open Problem 4.10 would give a higher-dimensional analogue of the decomposition of a one-dimensional BV function into absolutely continuous plus singular parts, with 'minimally singular' replacing 'absolutely continuous'.","The paper's Proposition 1.12 implies that the characterization, though broad, does not settle the rigidity problem in full generality; a complete solution would need a criterion that also covers domains where $\\{v^\\wedge>0\\}$ is not essentially open."],"forward_implications":["Rigidity over $\\Omega$ becomes equivalent to a single geometric identity: existence of a point from which the singular vertical distance to almost every other point vanishes.","The equality-case problem in $\\mathcal M_\\Omega(v)$ reduces to checking the barycenter function $b_{E,\\Omega}$: since rigidity is $b_{E,\\Omega}$ constant, Theorem 1.9 identifies exactly which $v$ force this.","The one-dimensional rigidity theorem from prior work is recovered: $v$ is rigid iff its support is an interval and $v\\in W^{1,1}$ with positive lower limit.","The mismatched stairway property of earlier work coincides with minimal singularity when $\\{v^\\wedge>0\\}$ is open, connected, and finite-measure.","If $D^s v$ is concentrated on a set that essentially disconnects $\\Omega$, rigidity fails; the singular vertical distance condition quantifies precisely this obstruction."],"supporting_citations":[{"why":"Supplies the equality-case characterization of $\\mathcal M(v)$ (Proposition 1.6), the rigidity theorem for special functions of bounded variation, and the essential-connectedness framework this paper localizes.","marker":"[6]"},{"why":"Establishes Steiner's perimeter inequality, the localized version (1.10), and the first general sufficient conditions for rigidity.","marker":"[9]"},{"why":"Provides the one-dimensional restriction theory of BV functions (Theorems 3.103, 3.107, 4.35) used to define and control $u_\\gamma$ along polygonal chains.","marker":"[2]"},{"why":"Introduces the original Vol'pert theory of restricting BV functions to lines, which motivates the curve families and the singular vertical distance.","marker":"[19]"},{"why":"Provides the geometric characterization of localized extremals for Steiner symmetrization, used for the barycenter formulation (1.13) and (1.15).","marker":"[16]"},{"why":"Supplies the construction in Example 12.25 used in Proposition 1.12 and the De Giorgi structure theorem used in the proof of Theorem 1.9.","marker":"[14]"},{"why":"Provides the decomposition theorem for sets of finite perimeter, invoked to extract an indecomposable component in Proposition 1.12.","marker":"[1]"}],"fun_headline_variants":["One pseudometric settles Steiner's rigidity","Minimal singularity: the right condition for Steiner","Zero singular distance implies Steiner rigidity","Steiner's extremals are minimally singular functions"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole bridge to rigidity rests on Proposition 1.6, a cited equality-case characterization taken from prior work 'by careful inspection of the proofs'; if that characterization requires hypotheses stronger than the domain condition (1.16), Theorem 1.9 does not follow.","fun_headline_variants_meta":{"raw":{"variants":["One pseudometric settles Steiner's rigidity","Minimal singularity: the right condition for Steiner","Zero singular distance implies Steiner rigidity","Steiner's extremals are minimally singular functions"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000744,"raw_usage":{"total_tokens":3305,"prompt_tokens":916,"completion_tokens":2389,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":532,"completion_tokens_details":{"reasoning_tokens":2335}},"tokens_in":532,"tokens_out":2389,"duration_ms":18373,"temperature":1.0,"reasoning_tokens":2335,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T11:53:16.975918+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the singular vertical distance for the Cantor function on $(0,1)$: the function $b(t)=|D^s u|((0,t))$ is nonconstant and satisfies conditions (1.22)-(1.24), so $u$ is not minimally singular; the theorem then forces $\\operatorname{SVD}_{u,(0,1)}(\\bar x,x)>0$ for positive-measure pairs, and a direct chain computation should confirm this. A computation giving $\\operatorname{SVD}=0$ almost everywhere for this $u$, or in higher dimensions any $u$ with $\\operatorname{SVD}=0$ almost everywhere that still admits a nonconstant controlled $b$, would refute Theorem 1.8.","supporting_citations":[{"cited_title":"Cagnetti, M","cited_arxiv_id":null,"evidence_quote":"Supplies the equality-case characterization of $\\mathcal M(v)$ (Proposition 1.6), the rigidity theorem for special functions of bounded variation, and the essential-connectedness framework this paper localizes."},{"cited_title":"Chleb´ık, A","cited_arxiv_id":null,"evidence_quote":"Establishes Steiner's perimeter inequality, the localized version (1.10), and the first general sufficient conditions for rigidity."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Introduces the original Vol'pert theory of restricting BV functions to lines, which motivates the curve families and the singular vertical distance."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the geometric characterization of localized extremals for Steiner symmetrization, used for the barycenter formulation (1.13) and (1.15)."},{"cited_title":"Maggi , Sets of finite perimeter and geometric variational problems , vol","cited_arxiv_id":null,"evidence_quote":"Supplies the construction in Example 12.25 used in Proposition 1.12 and the De Giorgi structure theorem used in the proof of Theorem 1.9."},{"cited_title":"Ambrosio, V","cited_arxiv_id":null,"evidence_quote":"Provides the decomposition theorem for sets of finite perimeter, invoked to extract an indecomposable component in Proposition 1.12."}],"review_version":1}