{"id":"c87876a7-e769-4896-920a-384df124c344","arxiv_id":"2510.07078","paper_version":2,"verdict":"UNVERDICTED","confidence":"MODERATE","novelty_score":0.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A survey of recent results: the multi-bubble isoperimetric conjecture is proved in Gaussian space for all k≤n and for up to five bubbles on Rⁿ and Sⁿ, with the remaining cases open.","lead":"This paper is a survey of recent mathematical proofs about soap-bubble-like partitions that minimize surface area for fixed bubble volumes, mostly obtained by the author and collaborators. It summarizes which conjectures are now proven (in Gaussian, Euclidean and spherical spaces) and which remain open (hyperbolic spaces, larger numbers of bubbles).","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Stability of standard k-bubbles is asserted as a result but rests on an unproven ADN boundary-regularity extension; this is the most load-bearing gap among the headline claims.","rationale":"The paper is a survey, and the main minimizer theorems on G^n and on R^n/S^n for k≤5 are quoted from published or arXiv-sourced papers with detailed proof sketches; I do not see a reason to doubt those. My concern is narrower and specific: the stability theorem is part of the headline, and the manuscript itself flags in §9(7) that the proof requires an open ADN-type boundary-regularity result. A claim that depends on an open problem is not a settled theorem, and this is a genuine missing-support issue under the review rules. The trace identity (8.5) is also an explicit limitation, but it is outside the claimed range of the headline results because the paper only claims the multi-bubble conjecture on S^n/R^n for k≤5; hence it is not the most load-bearing gap. The recommended verdict CONDITIONAL reflects that the stability assertions should be accepted only once the technical regularity step is either supplied or shown to be unnecessary for standard bubbles. This is not an accusation of misconduct; the authors are transparent about the gap, and the point is about the logical status of the claim as written.","tokens_in":21617,"tokens_out":9126,"duration_ms":79922,"concrete_test":"Read the proof of the stability theorem in [48] (arXiv:2504.11185) and determine whether standard k-bubbles, whose interfaces are subsets of generalized spheres with regular singular strata, require the ADN sector estimate from Open Problem 7, or whether their special geometry lets the test-function approximation be done directly. If direct approximation works, the concern fails for the specific §6 claim; if the proof explicitly relies on the unproven sector estimate, the stability assertion must be downgraded to 'conditional on Open Problem 7.'","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The survey's headline includes the claim that, for all 1≤k≤n+1 and n≥3, standard k-bubbles in R^n, S^n, and H^n are stable (§6). The proof sketch in §8.9 uses a conformally flattening potential and a multi-bubble Brascamp–Lieb inequality, but Open Problem 7 states that removing the technical assumptions on test functions requires extending Agmon–Douglis–Nirenberg boundary regularity from half-planes to convex sectors with aperture cos^{-1}(-1/3), the local model of T-singularities. This extension is not supplied and is listed as open. Thus the stability theorem, one of the three central claims, is conditional on an unresolved regularity result. The text is transparent about this, but it still reports stability of standard bubbles as a confirmed fact even though the publicly available proof either only covers restricted test functions or requires an open PDE estimate. By contrast, the reader's flagged trace identity (8.5) affects only the unclaimed extension to k>5 on S^n/R^n, not the stated range of the minimizer theorems.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This survey by E. Milman reviews recent progress on multi-bubble isoperimetric problems. It reports that the Gaussian multi-bubble conjecture holds for the full applicable range 2 ≤ k ≤ n, that the double-, triple-, quadruple-, and quintuple-bubble conjectures hold on R^n and S^n for n ≥ k, and that standard k-bubbles and more general Möbius-flat spherical Voronoi partitions are stable. The paper includes detailed proof sketches for the spherical Voronoi structure, connectedness of cells, the PDI approach to isoperimetric profiles, and the stability argument, and it closes with a list of open problems. Crucially, the text itself flags two limitations: the trace identity (8.5) needed for general k on S^n/R^n is unverified, and the full stability statement for all test functions requires an unresolved Agmon–Douglis–Nirenberg boundary regularity extension.","tokens_in":21776,"tokens_out":4152,"duration_ms":24941,"significance":"If the reported theorems are correct, they settle major conjectures in soap-bubble geometry for the stated ranges: the Gaussian case completely, and the Euclidean/spherical case up to five bubbles. The survey is valuable as a unified account of a substantial body of work by the author and collaborators. Its strengths include explicit proof sketches, a clear separation of proved results from open problems, and honest statements of technical gaps. The main caveat is that one of the headline claims—the stability theorem in §6—is presented as a confirmed result although the proof is explicitly conditional on an open PDE regularity question. The survey is nevertheless a useful synthesis and does not conceal its limitations.","major_comments":[{"comment":"The unqualified bullet 'For all 1≤k≤n+1 and n≥3, standard k-bubbles in R^n,S^n,H^n are stable' is stronger than what the proof sketch supports. §8.9 and Open Problem 7 state that removing the technical assumptions on test functions in the stability test requires extending Agmon–Douglis–Nirenberg regularity from half-planes to convex sectors with aperture cos^{-1}(-1/3), which is not supplied. Since Definition 7.3 defines stability for all smooth compactly supported vector fields, the theorem as stated is conditional. Please revise §6 to state exactly the class of test functions covered, and mark the full statement as open or as conditional on the ADN extension.","section":"§6 and Open Problem 7"},{"comment":"The survey correctly separates the k≤5 result from the general-k barrier posed by the trace identity (8.5). However, the text should be explicit that the proof of the quintuple result relies on a case distinction in §8.8 that verifies (8.5) in each alternative, while Open Problem 6 concerns proving (8.5) in general for k>5. As written, an uncareful reader could conclude that the PDI argument for all k≤n is blocked by (8.5), which is not the claimed state of affairs.","section":"§4 / §8.7–8.8"},{"comment":"The theorem that minimizing k-clusters in R^n and S^n are spherical Voronoi with connected cells is quoted from [47] with only a proof sketch. This is acceptable in a survey, but the survey should indicate more precisely which parts of the proof are fully carried out in [47] and which are being summarized at survey level, so that the reader can distinguish established results from the author's heuristic account.","section":"§4, Theorem 4.4"}],"minor_comments":[{"comment":"Typo: 'quituple' should be 'quintuple'.","section":"§4"},{"comment":"References [46] and [48] are arXiv preprints. For a journal survey, please mark these as 'arXiv' in the bibliography or state their publication status.","section":"References"},{"comment":"The phrase 'Modulo technicalities' is vague. A footnote specifying the precise test-function restriction and where the proof appears in [48] would improve precision.","section":"§6"},{"comment":"The notation D_con and the statement that LJac is self-adjoint and Fredholm in L2(Σ1, μ^{n−1}) would benefit from a precise reference to the corresponding proposition in [47] where the domain and spectral theory are developed.","section":"§8.6"}],"recommendation":"major_revision","confidential_remarks":"This is a survey by a leading researcher in the area, and most of the mathematical content is drawn from published or arXiv works. My main concern is the overstatement of the stability theorem relative to its caveat. This is fixable by precise rewording and does not require new mathematics, but it does affect a headline claim of the paper. The survey should also make clear that the extension beyond five bubbles on S^n/R^n is an open problem, as it already does in Open Problem 6."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear X,\n\nBottom line: this is a survey, not a research paper. No new theorem appears; the headline results are quoted from the author's previous papers. That is fine if you read it as a status report, and it is a good one: it collects the recent resolutions of Sullivan's multi-bubble conjectures in Gaussian space (all k≤n) and the double- to quintuple-bubble cases on Rn and Sn, plus the stability results for standard partitions. The proof sketches are useful, and the survey is unusually honest about what remains open. The author explicitly marks the stability results as 'modulo technicalities' and lists the two specific unresolved points: the trace identity (8.5) that would extend the Sn/Rn results beyond k=5, and the ADN boundary-regularity extension needed to remove test-function restrictions in the stability proofs (Open Problems 6 and 7 in the text).\n\nWhere the soft spots are: the stability claims in Section 6 are presented as results, but the proof is conditional on that ADN extension. The text says so, but a reader skimming the boxed statement could come away thinking the theorem is fully proved. That is a real caveat, not a manufactured one. The minimizer theorems for k≤5 on Sn/Rn do not depend on the trace identity, so that gap is only relevant to extending to general k≤n, which the survey does not claim. The proof sketches in Section 8 are too compressed to verify line by line; you need the original papers, and one of those (Milman–Xu) is only on arXiv. Still, the cited papers include Annals and Acta, so the burden is on the original sources, and the survey is fair about that.\n\nFor whom: graduate students or researchers who want a quick, accurate map of what is known and what is not, and a list of concrete open problems. It is not the place to learn the proofs. It deserves a serious referee: an editor should send it out, mostly to check that the quoted results match the cited papers and that the open problems are stated correctly.\n\nMy advice: treat it as a useful survey, cite it for the open problems, but do not cite it as the proof of the stability theorems—cite [48] for that, with the caveat that the technical assumptions are not yet removed.\n\nBest","headline":"A clear, honest survey of recent multi-bubble isoperimetric results, but the stability claims rest on an open regularity extension and the paper itself proves no new theorems.","tokens_in":22355,"tokens_out":2494,"would_cite":true,"duration_ms":23103,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["49Q20","49Q10","53A10","51B10"],"pacs":[],"model":"deepseek-v4-flash","headline":"The survey establishes that the multi-bubble isoperimetric conjecture holds for all k≤n in Gaussian space and for k≤5 in Euclidean and spherical space, and that standard bubbles in R^n, S^n, H^n are stable for all k≤n+1.","keywords":["multi-bubble isoperimetric problem","soap bubbles","stability","clusters","partitions","spherical Voronoi","isoperimetric profile","conformal Jacobi fields"],"falsifier":"Take a candidate minimizing spherical Voronoi k-cluster in S^n with k≥6 that is neither full-dimensional nor Möbius-flat, compute the operator F from the conformal Jacobi fields, and check whether tr(F(½ Id + k⊗k)) = H^{n−1}(Σ1); any violation would block the PDI extension. Alternatively, compare F with its relaxation F0 obtained by conformal perturbation to test the required continuity F(lim Ω_t)=lim F(Ω_t).","tokens_in":21380,"feed_emoji":"🫧","tokens_out":8442,"duration_ms":62375,"temperature":0.7,"pith_summary":"The survey reports a sequence of results that pin down which soap-bubble clusters minimise perimeter for a given set of enclosed volumes. In Gaussian space the conjecture is settled completely: for every number k of bubbles up to the ambient dimension, the minimiser is a standard simplicial bubble with flat interfaces. In Euclidean and spherical space, the same conclusion holds for up to five bubbles (with uniqueness open on R^n in the five-bubble case), and the proof is reduced to a combinatorial statement about the adjacency graph of cells. The survey also reports that every standard bubble, and more generally every Möbius-flat spherical Voronoi partition, is stable in all three model spaces, meaning no infinitesimal volume-preserving deformation can lower its area. The remaining gap to the full conjecture in R^n and S^n is a single trace identity whose verification is listed as an open problem.","feed_headline":"Solved: Gaussian multi-bubble conjecture, plus five-bubble cases","feed_subtitle":"A new proof strategy settles the Gaussian conjecture in full and confirms minimizers up to five bubbles in Euclidean and spherical space.","key_machinery":"The central objects are the standard k-bubbles — stereographic projections of the equal-volume Voronoi partition of S^n by k+1 points — and the more general spherical Voronoi partitions. The load-bearing mechanism is the reduction theorem: a minimizer with k≤n must be spherical Voronoi with connected cells, turning the variational problem into a finite-dimensional one. The final comparison uses the multi-bubble isoperimetric profile and conformal Jacobi fields satisfying L_Jac f = (n−1)a; a trace identity for the operator F is what would push the argument beyond five bubbles, and the survey states this identity is not yet verified in general.","core_discovery":"The central claim: a minimizing k-cluster in R^n or S^n (k≤n) is necessarily a spherical Voronoi cluster with connected cells — interfaces lie on geodesic spheres and cells are cut out by half-space inequalities. This reduces the global problem to finite dimensions, governed by the cell-incidence graph, and rules out disconnected cells and empty chambers. The structure, plus a maximum-principle argument for the isoperimetric profile, proves the double- through quintuple-bubble conjectures on S^n and R^n for n≥k and the full Gaussian conjecture for all 2≤k≤n. Separately, standard k-bubbles in R^n, S^n and H^n are stable for all 1≤k≤n+1, as are regular Möbius-flat spherical Voronoi partitions","pith_inferences":["The trace identity (8.5) appears to be the single technical gate between the current k≤5 results and the full conjecture in Euclidean and spherical space; a counterexample to it would invalidate the PDI strategy rather than just the conjecture.","The stability results for Möbius-flat spherical Voronoi partitions suggest that local minimality holds for a much wider class of bubbles than standard ones, so global uniqueness is a matter of global comparison, not local stability.","The Gaussian case being fully closed indicates that the flat-interface variant is the most tractable; extending the PDI approach to H^n would require reversing the sign of a stability inequality, so a genuinely new idea is needed there.","One could test the strategy computationally for a k=6 or k=7 spherical Voronoi cluster in S^n by numerically computing the operator F and checking whether (8.5) holds; this would provide evidence either for or against the remaining conjecture."],"forward_implications":["The Gaussian multi-bubble conjecture is fully resolved: for every 2≤k≤n the standard simplicial k-bubble uniquely minimizes Gaussian perimeter.","In R^n and S^n, the double-, triple-, quadruple- and quintuple-bubble conjectures hold for n≥k; uniqueness on R^n in the quintuple case remains open.","Standard k-bubbles in R^n, S^n and H^n are stable for every 1≤k≤n+1, giving local-minimality evidence for the still-open cases.","The spherical Voronoi structure theorem shows that cells of a minimizer are connected and no empty chambers can be trapped, resolving a conjecture of Heppes.","If the trace identity (8.5) is verified, the same PDI argument would confirm the multi-bubble conjecture on S^n and R^n for all k≤n (without uniqueness on R^n)."],"fun_headline_variants":["Gaussian multi-bubble conjecture fully settled","Minimizers up to five bubbles proven in Euclidean, spherical spaces","Stable soap bubbles: new proof for standard partitions","Multi-bubble isoperimetric minimizers characterized","Five-bubble conjectures confirmed via Voronoi clusters"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"Everything beyond five bubbles on S^n and R^n hangs on one unverified identity: for a minimizing spherical Voronoi cluster in S^n, tr(F(½ Id + k⊗k)) must equal H^{n−1}(Σ1); the survey states this could not be verified in general and lists it as Open Problem 6.","fun_headline_variants_meta":{"raw":{"variants":["Gaussian multi-bubble conjecture fully settled","Minimizers up to five bubbles proven in Euclidean, spherical spaces","Stable soap bubbles: new proof for standard partitions","Multi-bubble isoperimetric minimizers characterized","Five-bubble conjectures confirmed via Voronoi clusters"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000501,"raw_usage":{"total_tokens":2195,"prompt_tokens":557,"completion_tokens":1638,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":301,"completion_tokens_details":{"reasoning_tokens":1561}},"tokens_in":301,"tokens_out":1638,"duration_ms":10802,"temperature":1.0,"reasoning_tokens":1561,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T11:01:06.621978+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a candidate minimizing spherical Voronoi k-cluster in S^n with k≥6 that is neither full-dimensional nor Möbius-flat, compute the operator F from the conformal Jacobi fields, and check whether tr(F(½ Id + k⊗k)) = H^{n−1}(Σ1); any violation would block the PDI extension. Alternatively, compare F with its relaxation F0 obtained by conformal perturbation to test the required continuity F(lim Ω_t)=lim F(Ω_t).","supporting_citations":[],"review_version":1}