{"id":"1f5d73e0-5b43-43f4-ae15-620e700998f0","arxiv_id":"2607.06508","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":5,"one_line_summary":"Complete 3-manifolds with scalar curvature lower bound, finitely many ends, and finite first Betti number satisfy a sharp bottom spectrum upper bound and are parabolic under positive scalar curvature.","lead":"This paper proves that 3-manifolds with scalar curvature bounded below, finitely many ends, and finite first Betti number have bottom spectrum bounded above by the curvature bound, and are parabolic when scalar curvature is uniformly positive. This extends Cheng's classical eigenvalue estimate from Ricci curvature to scalar curvature under necessary topological constraints.","discovery_kind":"unclear","skeptic_critique":{"model":"glm-5.2","headline":"The 'without loss of generality' one-end reduction is under-justified: the divergence theorem step in Theorem 3.1 and the Dirichlet problem in Theorem 4.1 both require D_R to be bounded, which fails when M has multiple ends. The fix is straightforward but unstated.","rationale":"The reader correctly identified the one-end reduction as the weakest point. I traced the algebraic steps in detail (the Young's inequality chain from (3.5)–(3.6), the coefficient computation -3δ/(2+δ)² ≤ -δ/2 which requires δ ≤ √6-2 ≈ 0.449 rather than the stated δ < 1/2 but is satisfied since ε can be taken arbitrarily small in the contradiction argument, the h(d) computation where sec²θ - tan²θ = 1 gives exactly 2π²/(δL²), and the final bound where (2K+1) ≥ 1 yields the clean estimate (3.9)) and found them correct. The μ-bubble existence and regularity in dimension 3 is well-established via [22, 5]. The Bochner formula argument in Theorem 4.1 is also correct: the Kato-type cancellation gives Δ|∇u| ≥ (½S - ½S_t)|∇u| without needing strengthened Kato. The level-set connectedness cited from [13] (Li-Tam) and [15] is standard. The one-end reduction is the real gap: the divergence theorem and Dirichlet problem both require bounded domains, which fails with multiple ends. However, the fix—running the argument in each end separately and summing—is straightforward and standard in geometric analysis. The boundary terms on the inner cross-sections are compact and uniformly bounded. This is a presentation gap, not a fundamental error. The results are very likely correct as stated. ACCEPT with HIGH confidence is defensible, though noting the gap would strengthen the paper.","tokens_in":11032,"tokens_out":18970,"duration_ms":949984,"concrete_test":"Explicitly construct the multi-end version of the volume bound in Theorem 3.1: for M with k ends, for each end E_i define D_R^i as the bounded domain between a fixed compact cross-section Γ_i ⊂ E_i and the μ-bubble surface Σ_R^i. Verify that the divergence theorem gives λ₁(M) Vol(D_R^i) ≤ ∫_{Σ_R^i} |∇ln w| + ∫_{Γ_i} |∇ln w|, that the Γ_i boundary term is uniformly bounded (independent of R), and that summing over i = 1,…,k and letting R → ∞ yields Vol(M) ≤ kC/λ₁(M) < ∞. If the boundary terms on Γ_i cannot be uniformly controlled, the one-end reduction would be a genuine obstruction rather than a presentation gap.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Both proofs state 'we may assume without loss of generality that M has one end' without justifying the reduction. This is not merely cosmetic. In Theorem 3.1, the final volume bound uses the divergence theorem on D_R = M \\ N_R (where N_R is the unbounded component of M \\ Σ_R), yielding λ₁(M) Vol(D_R) ≤ ∫_{Σ_R} |∇ln w| ≤ C. If M has k ≥ 2 ends and Σ_R lies in one end E₁, then M \\ Σ_R has two unbounded components (one containing the rest of E₁, another containing E₂,…,E_k), so there is no unique unbounded component N_R, and D_R is not a bounded domain. The divergence theorem on an unbounded domain does not yield the finite volume bound. The same issue arises in Theorem 4.1, where the Dirichlet problem (4.4) is posed on D with boundary Γ ∪ Σ; if D contains other ends, it is unbounded and the problem is ill-posed as stated. The fix is to run the μ-bubble construction in each end separately: for each end E_i, construct Σ_R^i and the bounded annular domain D_R^i between a fixed inner boundary Γ_i and Σ_R^i, apply the divergence theorem there (picking up an additional boundary term on Γ_i that is uniformly bounded since Γ_i is compact), and sum over the finitely many ends. This gives Vol(M) < ∞ in Theorem 3.1 and the energy bound in Theorem 4.1. The argument is standard and the authors are experienced, but the reduction is not self-evident and should be spelled out. The reader correctly flagged this as the weakest assumption.","agreement_with_reader":"agree"},"referee_report":{"model":"glm-5.2","summary":"This paper establishes two results for complete noncompact 3-manifolds with scalar curvature lower bounds under the topological assumptions of finitely many ends and finite first Betti number. Theorem 1.1 proves a sharp upper bound on the bottom of the spectrum, λ₁(M) ≤ K, given S ≥ −6K, paralleling Cheng's classical Ricci curvature bound. Theorem 1.2 proves that if S ≥ 1, the manifold is parabolic (admits no positive Green's function). The proofs use the warped μ-bubble technique, applying the second variation formula to surfaces constructed in annular regions of the manifold to derive uniform area and energy estimates, leading to contradictions under the assumption of positive bottom spectrum or non-parabolicity.","tokens_in":11401,"tokens_out":1399,"duration_ms":618409,"significance":"The results are sharp and address natural questions in the interface of scalar curvature geometry and spectral theory. The constant K in Theorem 1.1 exactly matches the scalar curvature lower bound, and the topological exclusion of the (S²×S¹)#(S²×S¹) universal cover example is well-motivated. The proofs build on established μ-bubble machinery (Chodosh-Li, Zhu) and the Schoen-Yau rearrangement idea, applying them in a novel way to eigenvalue and parabolicity problems. The paper provides falsifiable, sharp predictions and the argument is largely self-contained modulo standard references for μ-bubble existence and regularity.","major_comments":[{"comment":"§3, Theorem 3.1 proof: The 'without loss of generality' reduction to one end is not justified and is load-bearing for the final volume estimate. The proof defines N_R as the unbounded component of M∖Σ_R and D_R := M∖N_R, then applies the divergence theorem on D_R to conclude λ₁(M)Vol(D_R) ≤ ∫_{Σ_R} |∇ln w| ≤ C. If M has k ≥ 2 ends and Σ_R lies in one end E₁, then M∖Σ_R has at least two unbounded components (one in E₁, another containing E₂,…,E_k), so there is no unique unbounded component N_R, and D_R as defined is not a bounded domain. The divergence theorem on an unbounded domain does not yield the finite volume bound. The fix (running the μ-bubble construction in each end separately, summing the resulting bounds) is standard but must be stated.","section":null},{"comment":"§4, Theorem 4.1 proof: The same one-end reduction issue arises here. The Dirichlet problem (4.4) is posed on D with boundary Γ ∪ Σ. If M has multiple ends and D contains other ends beyond the one being analyzed, D is unbounded and the Dirichlet problem as stated is ill-posed. The reduction to one end or the per-end construction with summation needs to be explicitly justified.","section":null},{"comment":"§3, Theorem 3.1 proof, Eq. (3.6)→(3.7): The transition from (3.6) to (3.7) replaces |∇_Σ u|²/u² with |∇u|²/u². Since u = w^γ and w is defined on all of M, |∇u|² = |∇_M u|² ≥ |∇_Σ u|². The sign in (3.6) is negative (−(8+δ)/12 · |∇_Σ u|²/u²), so replacing with the larger quantity |∇u|²/u² preserves the inequality direction. This step is correct but the justification (that |∇_Σ u|² ≤ |∇u|² and the coefficient is negative) should be stated explicitly for the reader's benefit.","section":null}],"minor_comments":[{"comment":"Title and running header: 'P ARABOLICITY' and 'CUR V A TURE' contain stray spaces (likely a formatting artifact).","section":null},{"comment":"§2, line below Eq. (2.3): 'K_{∂*Ω}' is introduced as the Gauss curvature of ∂*Ω but the subscript formatting could be clearer; consider K_Σ for consistency with later usage.","section":null},{"comment":"§3, Theorem 3.1 proof: The choice δ = ε/(K+1) is stated without motivation. A brief remark that this normalization ensures δ < 1/2 when ε < 1/2 (for any K ≥ 0) would help the reader.","section":null},{"comment":"§3, Theorem 3.3 proof: The statement 'we may assume without loss of generality that M has one end and its first Betti number is zero' conflates two reductions. The one-end reduction has the same issue as in Theorem 3.1. The b₁ = 0 reduction is a separate step (the finite b₁ case requires the connectedness result from [13]); this should be clarified.","section":null},{"comment":"§4, Theorem 4.1 proof, Eq. (4.5)–(4.9): The constant C₀ depends on the boundary Γ and the solution u_R on Γ. When passing to the limit R→∞, the text claims {u_R} converges to a harmonic function w on M₀. The convergence of C₀(R) to C₁ (defined analogously with w) should be briefly justified (e.g., by uniform convergence on compact sets containing Γ).","section":null},{"comment":"§4, Theorem 4.1 proof: The Bochner formula computation cites [15] for the Gauss curvature equations on level sets. Since the level-set connectedness is also cited from [15], a brief statement of what exactly is being used from [15] would improve self-containedness.","section":null},{"comment":"References: [19] lists 'arXiv:2408.0824' — this appears to be missing a digit (likely 2408.08240 or similar).","section":null}],"recommendation":"minor_revision","confidential_remarks":"The core arguments are sound and the algebraic manipulations check out. The one-end reduction is the main gap, but it is a standard fix that the authors (who are experienced) will certainly be able to address. I do not view this as a fundamental obstruction. The paper is a good fit for the journal."},"author_rebuttal":{"model":"glm-5.2","summary":"We thank the referee for a careful reading and for identifying a genuine gap in our one-end reduction argument, as well as for suggesting a clarifying remark. We address each comment below.","responses":[{"response":"The referee is correct. The statement 'we may assume without loss of generality that M has one end' is not justified as written. The issue is that while the μ-bubble construction is local to a single end, the final volume estimate relies on D_R exhausting all of M, which fails when there are multiple ends. We will revise the proof as follows. For each end E_i (i = 1, ..., k), we run the warped μ-bubble construction in the annular region of E_i, obtaining a surface Σ_R^{(i)} with the uniform bound (3.9). The bounded domain D_R^{(i)} enclosed by Σ_R^{(i)} (the component containing the fixed point p) is indeed bounded, and the divergence theorem yields λ₁(M) Vol(D_R^{(i)}) ≤ C for each i. Summing over all k ends gives λ₁(M) Vol(M) ≤ kC, which is still a uniform bound (k is finite by hypothesis), yielding the same contradiction. We will spell this out explicitly in the revision.","revision_made":"yes","referee_comment":"§3, Theorem 3.1 proof: The 'without loss of generality' reduction to one end is not justified. If M has k ≥ 2 ends, D_R as defined may be unbounded, and the divergence theorem does not yield the finite volume bound."},{"response":"Again the referee is correct. The same issue arises in Theorem 4.1. When M has multiple ends, the domain D bounded by Γ and Σ may contain other ends and thus be unbounded, making the Dirichlet problem (4.4) ill-posed as stated. We will revise by running the construction in each end separately. For each end E_i, we construct the surface Σ_R^{(i)} and solve the Dirichlet problem on the bounded domain D_R^{(i)} with boundary Γ ∪ Σ_R^{(i)}. The energy estimate (4.9) holds for each end with the same constant C (since C depends only on Γ, which is fixed). The convergence argument then applies to each end, and the co-area formula contradiction is obtained in any single end. Alternatively, one can note that it suffices to derive the contradiction in one end, since the harmonic function w obtained as the limit is defined on M_0 (the unbounded component of M ∖ B_p(R_0)), and the co-area formula argument in (4.11) only requires working in one end. We will clarify this in the revision.","revision_made":"yes","referee_comment":"§4, Theorem 4.1 proof: The same one-end reduction issue arises. The Dirichlet problem (4.4) on D with boundary Γ ∪ Σ is ill-posed if D contains other ends and is unbounded."},{"response":"We agree that the justification should be stated. The step is as follows: since u = w^γ is defined on all of M, we have |∇u|² = |∇_M u|² ≥ |∇_Σ u|², where ∇_Σ denotes the tangential gradient along Σ. In equation (3.6), the term appears with a negative coefficient −(8+δ)/12. Replacing |∇_Σ u|²/u² by the larger quantity |∇u|²/u² preserves the inequality (since we are subtracting a larger or equal quantity). We will add an explicit sentence to this effect in the revision.","revision_made":"yes","referee_comment":"§3, Theorem 3.1 proof, Eq. (3.6)→(3.7): The transition replaces |∇_Σ u|²/u² with |∇u|²/u². This is correct but the justification should be stated explicitly."}],"tokens_in":10855,"tokens_out":1754,"duration_ms":119532,"standing_objections":[]},"desk_editor":{"model":"glm-5.2","letter":"Bottom line: this paper proves two genuinely new results — a Cheng-type upper bound on the bottom spectrum for 3-manifolds with scalar curvature lower bounds (Theorem 3.1), and a parabolicity theorem under uniformly positive scalar curvature (Theorem 4.1). Both remove the Ricci curvature assumption from the authors' prior work [15], replacing it with scalar curvature plus the topological conditions (finitely many ends, finite first Betti). The parabolicity result is the more surprising of the two and is a strong global analytic consequence of positive scalar curvature in dimension three. The sharpness of the constant K in Theorem 3.1, matching the scalar curvature lower bound exactly, is a clean parallel to Cheng's classical result. The proofs are well-structured. The contradiction argument, the warped μ-bubble construction, and the second variation manipulations from (3.3) to (3.8) are traceable. I checked the key algebraic step: the inequality −(γ − (4+δ)/12)γ² ≤ −δ/2 with γ = 6/(2+δ) checks out. The choice L = 2(K+1)π/ε that makes the h² and |∇h| terms cooperate is clean. The iteration argument in Theorem 3.3 (volume → eigenvalue → volume, iterated to contradiction) is a nice trick. The parabolicity proof in Section 4 follows a parallel strategy but with different technical details — the Bochner formula application and the co-area formula contradiction at the end are both standard and correctly executed. The one soft spot is the 'without loss of generality' one-end reduction, which appears in both Theorem 3.1 and Theorem 4.1. The stress-test note flags this correctly: when M has multiple ends, the domain D_R = M ∖ N_R is not bounded because M ∖ Σ_R has multiple unbounded components, so the divergence theorem application and the Dirichlet problem (4.4) need adjustment. The fix is straightforward — run the μ-bubble construction in each end separately, pick up a uniformly bounded boundary term on each compact inner boundary, and sum over finitely many ends. This is standard and the authors clearly know how to do it, but it is not self-evident from the text as written. This is a gap in exposition, not in mathematics. The self-citation to [15] for the connectedness of level sets of harmonic functions is appropriate — that is a real result the authors established earlier, and the current paper builds on it legitimately. The μ-bubble existence and regularity citations to [22, 5] for n ≤ 7 are standard and correct. This paper is for geometric analysts working on scalar curvature, spectral geometry, and parabolicity. It deserves a serious referee who can verify the second variation computations and the one-end reduction. My recommendation: send to review. The one-end issue should be flagged to the authors for clarification, but it does not threaten the results.","headline":"Clean, sharp results; the one-end reduction needs a sentence but the proofs are solid","tokens_in":11912,"tokens_out":690,"would_cite":true,"duration_ms":648301,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["53C20","58J50","35P15"],"pacs":[],"model":"glm-5.2","headline":"Scalar curvature caps the spectrum in 3-manifolds","keywords":["scalar curvature","bottom spectrum","Laplacian","parabolicity","Green's function","mu-bubbles","3-manifolds","eigenvalue bound"],"falsifier":"A complete noncompact 3-manifold with scalar curvature S ≥ −6K, finitely many ends, finite first Betti number, but λ₁(M) > K would falsify the main theorem. Conversely, if the μ-bubble regularity or the second variation rearrangement fails for the specific warping and height functions used, the proof collapses.","tokens_in":11307,"feed_emoji":"🌊","tokens_out":1087,"duration_ms":219191,"temperature":0.7,"pith_summary":"This paper proves that for complete noncompact three-dimensional Riemannian manifolds with scalar curvature bounded below by −6K (K ≥ 0), the bottom of the Laplacian spectrum satisfies λ₁(M) ≤ K, provided the manifold has finitely many ends and finite first Betti number. This parallels Cheng's classical eigenvalue bound under a Ricci curvature lower bound, but here the hypothesis is strictly scalar curvature — a weaker geometric condition. The constant K is sharp, matching the scalar curvature bound exactly. The topological assumptions are necessary: without them, counterexamples exist (manifolds with infinitely many ends or nonamenable fundamental group can have positive bottom spectrum even with positive scalar curvature). A second result shows that when the scalar curvature is bounded below by a positive constant under the same topological hypotheses, the manifold admits no positive Green's function — it is parabolic.","feed_headline":"Scalar curvature caps the spectrum in 3-manifolds","feed_subtitle":"Under mild topology assumptions, the bottom Laplacian eigenvalue of a 3-manifold is bounded by its scalar curvature lower bound — matching a","key_machinery":"Warped μ-bubbles (surfaces stationary for a prescribed mean curvature functional), the second variation formula rearranged via a Schoen-Yau trick, the positive eigenfunction of the Laplacian as warping function, and Gauss-Bonnet applied to the separating surfaces.","core_discovery":"The central mechanism is the use of warped μ-bubbles — surfaces that are stationary for a prescribed mean curvature functional — to extract compact separating surfaces in the annular region of the manifold's end. By choosing the warping function as a power of the positive eigenfunction corresponding to the bottom spectrum, and choosing the prescribed mean curvature function h to blow up at the annular boundaries, the second variation formula for these μ-bubbles yields uniform area and energy bounds on the separating surfaces. These bounds force the manifold to have finite volume, contradicting the positivity of the bottom spectrum. The argument is a contradiction scheme: assuming λ₁ > K, one","pith_inferences":["The restriction to dimension 3 is tied to the regularity of μ-bubbles (guaranteed for n ≤ 7) and the Schoen-Yau rearrangement of the second variation formula; extending to higher dimensions would require either analogous regularity or a different variational framework, which the paper does not address.","The gap between the scalar curvature bound −6K and the eigenvalue bound K suggests that the factor 6 is specific to dimension 3 (where S = 2·Ric + curvature terms), and analogous results in higher dimensions would need to account for the different algebraic relationship between scalar and Ricci curvature.","If the finite first Betti number assumption could be relaxed to an analytic condition (e.g., amenability of the fundamental group), the result would subsume both the topological and the Brooks-type spectral obstructions in a single statement."],"forward_implications":["Manifolds with nonnegative scalar curvature, finitely many ends, and finite first Betti must have zero bottom spectrum, meaning they support no L² harmonic functions and are spectrally degenerate at the bottom.","The parabolicity result implies such manifolds with uniformly positive scalar curvature cannot sustain nonconstant positive harmonic functions, constraining the potential theory and heat flow behavior on these spaces.","The volume growth estimate (Theorem 3.3) shows that geodesic balls in such manifolds must have first Dirichlet eigenvalues decaying at least as fast as C/R² along a subsequence, which is the same rate as Euclidean space.","The technique provides a template for relating scalar curvature lower bounds to spectral and potential-theoretic properties in dimensions where direct minimal surface methods face topological obstructions."],"fun_headline_variants":["Warped μ-bubbles cap the bottom spectrum of 3-manifolds","Scalar curvature lower bound limits Laplacian spectrum in 3D","No positive Green's functions when scalar curvature stays positive","μ-bubble method bounds bottom eigenvalue via scalar curvature","3-manifolds with positive scalar curvature admit no Green's functions"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The entire argument depends on the existence and regularity of smooth warped μ-bubbles in the annular regions, which is guaranteed by citing prior work for dimensions at most 7. If the second variation formula fails to apply cleanly in any edge case of the construction, or if the reduction to one end does not hold independently for each end, the eigenvalue bounds would not hold.","fun_headline_variants_meta":{"raw":{"variants":["Warped μ-bubbles cap the bottom spectrum of 3-manifolds","Scalar curvature lower bound limits Laplacian spectrum in 3D","No positive Green's functions when scalar curvature stays positive","μ-bubble method bounds bottom eigenvalue via scalar curvature","3-manifolds with positive scalar curvature admit no Green's functions","Bottom spectrum of 3-manifolds controlled by scalar curvature bound","Stationary surfaces force finite volume in scalar-curved 3-manifolds","Scalar curvature governs spectral geometry of 3-manifolds","μ-bubbles extract separating surfaces that bound the spectrum","Positive scalar curvature forbids positive Green's functions"]},"model":"glm-5.2","effort":"high","cost_usd":0.0,"raw_usage":{"total_tokens":568,"prompt_tokens":365,"completion_tokens":203,"prompt_tokens_details":null},"tokens_in":365,"tokens_out":203,"duration_ms":16136,"temperature":1.0,"reasoning_tokens":54,"cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-08T03:21:57.063205+00:00","model_set":{"reader":"glm-5.2"},"falsifier":"A complete noncompact 3-manifold with scalar curvature S ≥ −6K, finitely many ends, finite first Betti number, but λ₁(M) > K would falsify the main theorem. Conversely, if the μ-bubble regularity or the second variation rearrangement fails for the specific warping and height functions used, the proof collapses.","supporting_citations":[],"review_version":1}