{"id":"81d2c760-6be4-441f-a158-40437d327c5e","arxiv_id":"2508.08103","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"For any finite set of points in the half-plane of axial symmetry, there exists a 3D Keller-Segel solution whose mass concentrates at those points with a precisely quantified finite-time blow-up rate.","lead":"This paper proves the existence of solutions to the three dimensional Keller-Segel chemotaxis equations that blow up in finite time, with density concentrating along several rings that shrink to points. It extends a known single-ring construction and sharpens the blow-up rate.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 5.1's barrier absorbs the intermediate-region O(|ln|x−ξ||) error only by verbal assertion; a rigorous check of that absorption is needed before the gluing construction is secure.","rationale":"The paper presents a serious and largely coherent construction of single-ring type II blow-up in the 3D axially symmetric Keller–Segel system. Section 2's elliptic integral expansion is detailed and plausible, the blow-up rate is internally consistent, and the inner theories imported from [10] are stated with explicit propositions. The main risk is not the physical picture but the rigor of the outer theory: Theorem 5.1 is the hinge that connects the inner two-dimensional regime to the three-dimensional far field, and its proof is largely a barrier construction with verbal assertions about error absorption. The reader's weakest_assumption identifies exactly this transition. I find the concern genuine: the O(|ln|x−ξ||) term in Proposition 2.10 could in principle dominate the barrier in the intermediate region if constants do not align, and the proof does not provide a written verification. However, my own estimates suggest the ratio is indeed ζ|lnζ| ≤ √T|lnT|, so the gap is likely fillable rather than fatal. I therefore recommend no change to the reader's CONDITIONAL verdict, with the concrete check above as the natural next step toward unconditional acceptance.","tokens_in":44008,"tokens_out":22181,"duration_ms":250940,"concrete_test":"Re-derive the barrier inequalities (5.10) and the absorption (5.8)–(5.9) using the full expansion from Proposition 2.10 rather than the simplified form (5.3). Concretely, for δ√(T−t) ≤ |x−ξ| ≤ √T, bound |(∇v0−∇Γ0)·∇φ0| and |(∇v0−∇Γ0)·∇φ3| by e^{−a√...}/D² and check that the implied constants are uniform with the stated δ as T→0. Run the computation at ζ=3δ√(T−t), ζ=√(T−t), and ζ=ε/2 for T=10^{-4},10^{-8},10^{-12}; if the error term ever dominates the leading barrier, Theorem 5.1 fails and the construction collapses.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim depends on Theorem 5.1, whose proof replaces ∇v0 by the piecewise approximation (5.3) and asserts that the discarded terms are absorbable. The dangerous region is the intermediate zone 2√δ(T−t) ≤ |x−ξ| ≤ ε, where Proposition 2.10 gives ∇v0 = ω1∇Γ0 + O(|ln|x−ξ||). For the first barrier term φ0, |∇φ0| ≈ |x−ξ| e^{−a√(2|lnD|)}/D² with D=|x−ξ|²+T−t, and the leading operator term is e^{−a...}/D². The error contribution is ≲ |ln|x−ξ||·|x−ξ|·e^{−a...}/D², whose ratio to the main term is |x−ξ||ln|x−ξ|| ≤ √T|lnT|, small as T→0 — provided the O(|ln|) error really is a pointwise vector bound and the barrier gradient is evaluated with the full drift. However, the proof does not display the needed estimate: (5.4)–(5.9) are asserted rather than verified, signs are dismissed as 'negligible', and the proof ends with 'rescaling and applying standard elliptic estimates'. If this intermediate error is not dominated uniformly, the outer fixed point in Section 6 collapses. This is a gap, not a demonstrated contradiction, but it is load-bearing.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper constructs axially symmetric finite-time Type II blow-up solutions to the three-dimensional Keller–Segel system, concentrating mass along an arbitrary number of rings. The main result, Theorem 1.1, asserts that for any k distinct points in R_+ × R there is an initial datum whose solution has the multi-ring expansion u(r,z,t) = Σ_j λ_j(t)^{-2} U(((r,z)-(r_j(t),z_j(t)))/λ_j(t))(1+o(1)) with λ_j(t) ≈ 2e^{-(γ+2)/2} √(T-t) e^{-√(|ln(T-t)|/2)}. The proof combines a precise asymptotic expansion for the three-dimensional axially symmetric inverse Laplacian of U (Section 2, Theorems 2.9 and Proposition 2.10), a tailored correction φ_λ (Section 3), an inner–outer gluing system (Section 4), an outer theory with a four-term barrier (Theorem 5.1), and a fixed-point argument (Section 6). The multi-ring case is only sketched in Section 6.1, and several key propositions are imported from the authors' previous works [10] and [29].","tokens_in":44418,"tokens_out":6750,"duration_ms":81218,"significance":"If the construction is complete, the result is significant: it gives the first rigorous multi-ring Type II blow-up for the three-dimensional axisymmetric Keller–Segel system, with a refined blow-up rate consistent with the two-dimensional theory, thereby generalizing the single-ring result of Hou–Nguyen–Song. The single-ring expansion in Section 2 is carefully developed and contains original elliptic-integral analysis; the derivation of Theorem 2.9 and Proposition 2.10 is detailed and appears sound. The paper also provides explicit estimates for the gradient of the Newtonian potential in cylindrical coordinates, which are of independent use. However, the proof of the main theorem as written is incomplete at several load-bearing points: the outer barrier verification is partly verbal, the parameter choice in Proposition 4.2 is deferred to [10], and the multi-ring statement is only sketched. These gaps currently prevent the paper from meeting the standard of a fully verified construction.","major_comments":[{"comment":"The outer theory is load-bearing for the fixed-point argument in Section 6, but its proof is incomplete. In the intermediate region 2√δ(T−t) ≤ |(r,z)−ξ| ≤ ε, Proposition 2.10 gives ∇v0 = ω1 ∇Γ0 + O(|ln|(r,z)−ξ||). The proof of Theorem 5.1 replaces ∇v0 by the piecewise expression (5.3) and states that the discarded terms 'can be absorbed into parts of the barrier'. The estimate below (5.3) bounds the drift-error contribution by √T |ln T| times the main singularity, but it does not display a uniform pointwise domination of the full remainder, including the O(|ln|) term and the (ω1−1)∇Γ0 contribution, in the overlap with the region where φ3 is active. The proof ends with 'rescaling and applying standard elliptic estimates'. A rigorous, explicit absorption estimate in this intermediate zone is needed; without it, the linear outer operator T_o^p and hence the contraction argument in Section 6","section":"Section 5, Theorem 5.1 and (5.3)–(5.9)"},{"comment":"Theorem 1.1 is stated for arbitrary k distinct points, but the multi-ring proof is only sketched. The text says 'We omit most of the details' and discusses only the case k=2. For k=2, the interaction terms E_{1,2} and E_{2,1} are asserted to be negligible without estimates, and the resulting two inner equations, two outer equations, orthogonality conditions, and parameter corrections are not written down. No induction or general construction for k distinct rings is provided. Since the multi-ring statement is the paper's main advertised generalization, this is a central gap that must be addressed by giving the full construction or a precise reduction showing that the k-ring case follows from the single-ring one with explicit control of inter-ring interactions.","section":"Section 6.1"},{"comment":"Proposition 4.2 is the mechanism that selects λ0(t) and α0(t) and produces the refined blow-up rate quoted in Theorem 1.1. Its proof is almost entirely deferred: the text states that after some reductions 'the detailed proof follows directly from the arguments presented in Section 4 of [10]'. The orthogonality conditions (4.15)–(4.16) are central to the construction, and the asymptotic expression for λ0(t) is a claimed improvement over [52]. At minimum, the nonlocal equation for λ0˙λ0(t), the role of the initial time −ε(T), and the solvability argument should be stated explicitly in the present setting. As written, the reader cannot verify the validity of (4.15)–(4.16) or the refined rate without consulting an external preprint.","section":"Section 4.1, Proposition 4.2"},{"comment":"The inner theory is imported from [10] as black boxes in Propositions 4.3–4.5, and Proposition 3.2 is proved only with the sentence 'For brevity, we omit the detailed computations.' These estimates are used in Section 6 to control the nonlocal terms involving ∇ψλ, so the adaptation to the axisymmetric three-dimensional setting must be demonstrated or precisely referenced. If the authors intend [10] to be used as a black box, they should state which specific statements and which modifications are needed; otherwise the present manuscript is not self-contained at a load-bearing point.","section":"Sections 3 and 4.5"}],"minor_comments":[{"comment":"The hypothesis reads 'Assume that (2.6) and (2.6) are satisfied.' Presumably the second reference should be (2.7).","section":"Section 5, Proposition 5.2"},{"comment":"There are several typographical errors: 'ecounter' near equation (2.17), 'iste ad' near the end of Section 5, and the reference [52] lists 'T. Y. How' instead of 'T. Y. Hou'.","section":"Throughout"},{"comment":"The norms ∥g∥⋆,o, ∥ϕ∥⋆⋆,o, and ∥g∥o are used with slightly inconsistent notation; unify the definitions so the statements of Theorem 5.1 and the fixed-point setup in Section 6 are directly comparable.","section":"Sections 5 and 6"},{"comment":"The expression uses φλ(T) although φλ is defined for t < T; clarify the intended meaning, e.g., a limit or an evaluation at a cut-off time.","section":"Equation (4.37)"},{"comment":"The O(|ln(|(r,z)−ξ|)|) term is singular as ξ is approached; it would be helpful to state explicitly that this means a bound by C|ln(|(r,z)−ξ|)| for small nonzero distance and to specify the constant's dependence on T, so that the absorption in Section 5 can be checked.","section":"Proposition 2.10"}],"recommendation":"major_revision","confidential_remarks":"The paper is essentially an adaptation of the gluing machine developed in [10] to the three-dimensional axisymmetric Keller–Segel system. The editors may wish to verify that [10] is in a citable, public form before relying on it. The main theorem for arbitrary k is not fully proved in the manuscript, and the outer barrier verification contains verbal steps. These are gaps that can likely be repaired with additional detail, but as written the paper is not yet ready for acceptance."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The thing to know: this paper proves (in detail) single-ring type II blow-up in the 3D axisymmetric Keller-Segel system and gives a refined rate, and it sketches the multi-ring extension. The genuinely new piece is Section 2.2: the asymptotic expansion of the 3D Newtonian potential of the standard bubble in cylindrical coordinates, with the omega_1 weight and elliptic-integral machinery. That section is careful and looks correct. The refined rate lambda(t) approx 2e^{-(gamma+2)/2} sqrt(T-t) e^{-sqrt(|ln(T-t)|/2)} is consistent with the 2D multi-bubble work and presumably with the Hou-Nguyen-Song single ring. So there is real new mathematical content here.\n\nThe soft spots are real but not fatal, in my view. First, the main theorem says \"any k distinct points,\" but the proof of Theorem 1.1 for arbitrary k is not written: Section 6.1 says \"We omit most of the details\" and only sketches k=2. That is a large gap in the advertised result. A serious referee should ask for a full multi-ring proof or a revised statement limiting the result to k=1. Second, the outer theory (Theorem 5.1) uses a four-term barrier and asserts that the O(|ln|x-xi||) error in the intermediate region is absorbed. The stress-test check shows the ratio is small: |x-xi||ln|x-xi|| <= sqrt(T)|ln T| -> 0, so the claim is plausible. But the proof does not actually display the absorption; (5.4)-(5.9) are stated and then constants are chosen verbally. This is a load-bearing gap in the gluing construction. I do not see a contradiction, but the estimate must be written out before the construction is secure.\n\nThe reliance on the authors' previous papers [10,28,29] is heavy: inner propositions are imported with \"see proof of Lemma X in [10].\" That is acceptable in this line of work, and the citations are to their own established methods, not an attempt to hide anything. The parameter lambda_0 is chosen by orthogonality conditions and the rate is derived from the nonlocal equation, so there is no circular fitting.\n\nBottom line: the paper deserves a serious referee. The single-ring case is a strong technical contribution, and the expansion in Section 2.2 is a useful tool on its own. The gaps are gaps, not demonstrated errors. I would send it to review, with the request that the referee insist on the multi-ring proof and the barrier verification. For me, I would bring it to reading group and probably cite the potential expansion, while being careful not to cite the k-ring theorem as established.","headline":"Substantial construction with a genuinely new potential expansion, but the advertised multi-ring theorem is only sketched and the outer barrier has a load-bearing gap; worth refereeing.","tokens_in":44908,"tokens_out":2640,"would_cite":true,"duration_ms":30198,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35B44","35K55","35Q92"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper constructs finite-time type II blow-up solutions of the three-dimensional axially symmetric Keller-Segel system in which mass concentrates along any prescribed finite collection of rings, with the refined two-dimensional blow-up","keywords":["Keller-Segel system","finite-time blow-up","type II blow-up","axially symmetric","ring concentration","gluing method","elliptic integrals","chemotaxis"],"falsifier":"Evaluate the quantities in (5.4)-(5.9) at points with $|(r,z)-\\xi|=(T-t)^{1/4}$ using the explicit expansion (2.28); if the $O(|\\ln(|(r,z)-\\xi|)|)$ term from the intermediate regime is not dominated uniformly by $e^{-a\\sqrt{2|\\ln((T-t)+|(r,z)-\\xi|^2)|}}\\big((T-t)+|(r,z)-\\xi|^2\\big)^{-2}$ for any admissible $a$, the barrier argument of Theorem 5.1 fails.","tokens_in":43861,"feed_emoji":"🪐","tokens_out":12734,"duration_ms":127200,"temperature":0.7,"pith_summary":"This paper proves that the three-dimensional axially symmetric Keller-Segel system admits finite-time type II blow-up solutions for any prescribed finite collection of rings. The main theorem states that for $k$ distinct points $(r_1,z_1),\\ldots,(r_k,z_k)\\in\\mathbb{R}_+\\times\\mathbb{R}$ there is an initial datum whose solution is, up to small errors, a sum of $k$ rescaled copies of the two-dimensional stationary profile $U(y)=8/(1+|y|^2)^2$, each centered at $(r_j(t),z_j(t))$ converging to the prescribed point, with the refined law $\\lambda_j(t)\\sim 2e^{-(\\gamma+2)/2}\\sqrt{T-t}\\,e^{-\\sqrt{|\\ln(T-t)|/2}}$. This generalizes the single-ring construction of [52] and matches the refined rates known in two dimensions. The proof uses an inner-outer gluing scheme whose key novelty is a precise asymptotic expansion of the three-dimensional axially symmetric Newtonian potential of $U$, expressed through complete elliptic integrals, together with a four-term outer barrier that absorbs the transition from the two-dimensional near field to the genuinely three-dimensional far field.","feed_headline":"3D chemotaxis blows up on any number of rings","feed_subtitle":"Inner-outer gluing yields multi-ring type II singularities with the refined 2D blow-up rate.","key_machinery":"The carrying mechanism is an inner-outer gluing scheme around the ansatz $u_1=(\\alpha/\\lambda^2) U((r,z)-\\xi)/\\lambda)\\,\\chi+\\phi_\\lambda$. The inner problem is a parabolic perturbation of the two-dimensional linearized operator $L[\\phi]=\\Delta\\phi-\\operatorname{div}(U\\nabla\\psi)-\\operatorname{div}(\\phi\\nabla\\Gamma_0)$, solved after the time rescaling $\\tau=\\int_0^t ds/\\lambda^2(s)$; the outer problem is solved by an explicit four-term barrier. The load-bearing three-dimensional input is the expansion of $\\nabla v_0$: near the ring it is $\\nabla\\Gamma_0$, in an intermediate shell it is $\\omega_1\\nabla\\Gamma_0+O(|\\ln|(r,z)-\\xi||)$, and in the far field it is $-4\\pi q_1 (r,z)/|(r,z)|^3$. The e","core_discovery":"The central claim is Theorem 1.1: for any $k$ distinct points $(r_1,z_1),\\ldots,(r_k,z_k)\\in\\mathbb{R}_+\\times\\mathbb{R}$ there is an initial datum whose solution is $u=\\sum_j \\lambda_j^{-2} U((r,z)-(r_j,z_j))/\\lambda_j)(1+o(1))$, uniformly on bounded sets, with $\\lambda_j\\sim 2e^{-(\\gamma+2)/2}\\sqrt{T-t}\\,e^{-\\sqrt{|\\ln(T-t)|/2}}$ and centers converging to the prescribed points. The blow-up is type II and each ring carries the two-dimensional stationary profile $U$. The novelty is the multi-ring generalization with the refined two-dimensional rate, enabled by a precise expansion of the 3D Newtonian potential of $U$ and a barrier controlling the 2D-to-3D crossover.","pith_inferences":["A natural test of the gluing's robustness is to track the first interaction between two rings at fixed separation as $T\\to0$: the paper treats mixed terms as negligible, so a sharper estimate of their contribution would reveal a spacing condition under which the construction still closes.","If the same barrier strategy transfers, analogous multi-bubble type II solutions should exist for other supercritical drift-diffusion systems whose linearized operator around the stationary profile has the same neutral modes.","The paper establishes existence but not stability; a natural follow-up, not addressed here, is whether each multi-ring configuration is stable or unstable under axially symmetric perturbations."],"forward_implications":["For every $k$ and every choice of $k$ distinct ring centers in $\\mathbb{R}_+\\times\\mathbb{R}$, there is an initial datum producing finite-time blow-up with one concentration ring at each prescribed center.","The blow-up rate at each ring is the refined two-dimensional one, $\\lambda_j(t)=2e^{-(\\gamma+2)/2}\\sqrt{T-t}\\,e^{-\\sqrt{|\\ln(T-t)|/2}}(1+o(1))$, so the singularity is of type II.","Near each ring the asymptotic profile is the two-dimensional stationary state $U$, not a three-dimensional self-similar profile.","The inverse-Laplacian expansion in cylindrical coordinates provides a quantitative bridge between the two-dimensional and three-dimensional regimes and can be reused in other axisymmetric problems, such as vortex ring dynamics."],"supporting_citations":[{"why":"Supplies the gluing scheme, inner solution operators, parameter selection and most of the fixed-point estimates that the present paper adapts.","marker":"[10]"},{"why":"The single-ring axisymmetric type II blow-up construction that the multi-ring result generalizes.","marker":"[52]"},{"why":"Provides the refined two-dimensional blow-up rate and spectral analysis that fix the expected lambda law.","marker":"[21]"},{"why":"Companion refined/stability analysis of two-dimensional blow-up used as baseline for the rate formula.","marker":"[22]"},{"why":"Source of the two-dimensional infinite-time blow-up framework, including the linear theory and Lemma 4.1 used for inner corrections.","marker":"[28]"},{"why":"Gluing methods for vortex dynamics, supplying the E and F operators and inner theories invoked in Section 4.5.","marker":"[29]"},{"why":"Handbook formulas for the complete elliptic integral K used in the expansion (2.12) of the 3D axially symmetric Newtonian potential.","marker":"[11]"},{"why":"Integral formula 4.224 used in Lemma 2.8 for the representation of the two-dimensional Newtonian potential.","marker":"[42]"}],"fun_headline_variants":["Multi-ring type II blow-up in 3D Keller-Segel","Refined blow-up rate for multi-ring chemotaxis","Arbitrary ring count for 3D chemotaxis blow-up","Gluing constructs multi-ring blow-up in 3D","Any number of rings: 3D chemotaxis blow-up"],"cache_read_input_tokens":2816,"weakest_assumption_plain":"The load-bearing premise is that the four-term outer barrier can absorb, for sufficiently small $T$, the $O(|\\ln(|(r,z)-\\xi|)|)$ error from the intermediate regime of Proposition 2.10 where the gradient of $v_0$ passes from the two-dimensional to the three-dimensional form; if that estimate fails, the outer fixed point and the whole construction collapse.","fun_headline_variants_meta":{"raw":{"variants":["Multi-ring type II blow-up in 3D Keller-Segel","Refined blow-up rate for multi-ring chemotaxis","Arbitrary ring count for 3D chemotaxis blow-up","Gluing constructs multi-ring blow-up in 3D","Any number of rings: 3D chemotaxis blow-up"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000769,"raw_usage":{"total_tokens":3183,"prompt_tokens":622,"completion_tokens":2561,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":366,"completion_tokens_details":{"reasoning_tokens":2471}},"tokens_in":366,"tokens_out":2561,"duration_ms":22290,"temperature":1.0,"reasoning_tokens":2471,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T21:38:14.268239+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Evaluate the quantities in (5.4)-(5.9) at points with $|(r,z)-\\xi|=(T-t)^{1/4}$ using the explicit expansion (2.28); if the $O(|\\ln(|(r,z)-\\xi|)|)$ term from the intermediate regime is not dominated uniformly by $e^{-a\\sqrt{2|\\ln((T-t)+|(r,z)-\\xi|^2)|}}\\big((T-t)+|(r,z)-\\xi|^2\\big)^{-2}$ for any admissible $a$, the barrier argument of Theorem 5.1 fails.","supporting_citations":[{"cited_title":"Axisymmetric type II blowup solutions to the three-dimensional Keller-Segel system","cited_arxiv_id":"2502.19775","evidence_quote":"The single-ring axisymmetric type II blow-up construction that the multi-ring result generalizes."},{"cited_title":"Collot, T","cited_arxiv_id":null,"evidence_quote":"Provides the refined two-dimensional blow-up rate and spectral analysis that fix the expected lambda law."},{"cited_title":"Collot, T","cited_arxiv_id":null,"evidence_quote":"Companion refined/stability analysis of two-dimensional blow-up used as baseline for the rate formula."},{"cited_title":"Existence and stability of infinite time blow-up in the Keller-Segel system","cited_arxiv_id":"1911.12417","evidence_quote":"Source of the two-dimensional infinite-time blow-up framework, including the linear theory and Lemma 4.1 used for inner corrections."},{"cited_title":"D´ avila, M","cited_arxiv_id":null,"evidence_quote":"Gluing methods for vortex dynamics, supplying the E and F operators and inner theories invoked in Section 4.5."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Handbook formulas for the complete elliptic integral K used in the expansion (2.12) of the 3D axially symmetric Newtonian potential."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Integral formula 4.224 used in Lemma 2.8 for the representation of the two-dimensional Newtonian potential."}],"review_version":1}