{"id":"7d35f00b-d16c-41d6-b015-f4cf77733a8e","arxiv_id":"2608.04553","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For complete locally conformally flat manifolds with nonnegative Ricci curvature, the space of polynomial-growth harmonic functions is no larger than in Euclidean space, and maximal dimension forces flatness.","lead":"This paper proves that on locally conformally flat spaces with nonnegative Ricci curvature, the number of harmonic functions growing at most polynomially is never larger than in flat Euclidean space. It answers a question Yau posed for this class, and shows equality forces the space to be Euclidean.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Proposition 4.1's L1 convergence rests on an unproved domination estimate for the spherical exponential average F(t); without it the blow-down to the cone and all of Section 5 collapse.","rationale":"The reader's weakest assumption identified the same load-bearing point: the proof depends on Ma's refined radial asymptotics and convergence assertions from [26] in Section 4, Proposition 4.1. My stress-test sharpens this to a specific missing domination estimate in the proof of (4.3)/(4.4). The manuscript states F(t)->1 in (4.12) and an annular estimate (4.15) by citing [26], but does not prove or state the quantitative bounds needed for dominated convergence in (4.11). Since Theorem 4.2's step 3 and step 5 rely on L1 convergence on the full ball, and Proposition 5.4 uses Theorem 4.2 to propagate the cone's convexity, this is the most load-bearing assumption. I found no internal contradiction in the rest of the argument, and the classification-based completion is standard. The verdict remains CONDITIONAL: the central claim is plausible and well-structured, but its correctness is contingent on the cited estimates from [26]. I did not find a reason to raise or lower the reader's confidence, so the verdict should be unchanged.","tokens_in":27475,"tokens_out":17720,"duration_ms":193927,"concrete_test":"Check whether [26, Lemma 4.7] provides a domination bound or a rate for F(t)-1. Re-derive (4.11) and the Q_R analogue using only what [26] proves; if F(t)-1 is only pointwise, construct a w with w>=wbar on a sequence of large radii for which the right side of (4.11) fails to vanish. Also confirm that (4.15) gives L1 convergence on B_Lambda, not just on annuli.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The decisive step is Proposition 4.1, where (4.3) and (4.4) assert P_R -> P_inf and Q_R -> Q_inf in L1(B_Lambda). The proof splits P_R into a radial part via (4.11), whose integrand contains F(Rs)-1, where F(t) is the spherical exponential average. Equation (4.12) cites [26, Lemma 4.7] for F(t)->1. But to pass from pointwise convergence F(Rs)-1 -> 0 to vanishing of the integral in (4.11), one needs a uniform integrable bound on s^{n-1} exp((n-2)(wbar(Rs)-wbar(R))) |F(Rs)-1|. The paper supplies no such domination, and the pointwise radial bound (4.10) controls only the radial factor, not the spherical excess w(tθ)-wbar(t) that enters F. Similarly, (4.15) is imported from [26, Page 25] and is needed to get Q_R -> Q_inf; without it, the weighted masses in Theorem 4.2, Step 5, cannot be identified with the cone averages. These L1 convergences are load-bearing: Theorem 4.2 uses (4.3) to pass to the limit in the equation and (4.4) to identify the limiting measure, and Proposition 5.4 then transfers the cone's log-convexity to the manifold. If either fails, the dimension bound in Theorem 5.8 and the main theorem lose their foundation.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript proves that on a complete noncompact locally conformally flat manifold (M^n,g), n≥3, with Ric≥0, the space H_d(M) of harmonic functions of polynomial growth of degree at most d satisfies h_d(M)≤h_d(R^n) for every integer d≥0, with equality for some d≥1 forcing M isometric to R^n. The proof splits into the cases β=0 and β>0 for the asymptotic volume ratio. In the β=0 case, the authors use Carron's diameter criterion together with a radial ODE argument. In the β>0 case, they prove a distance comparison, introduce rescaled weights P_R and Q_R, establish weighted G-convergence of rescaled harmonic functions to harmonic functions on a limiting cone, and transfer log-convexity of the cone spectrum back to the manifold via a uniform one-step propagation. The argument is heavily dependent on refined asymptotics of the conformal factor from Ma's preprint [26] and on potential-theoretic results of Ma-Qing [27].","tokens_in":27761,"tokens_out":14609,"duration_ms":151241,"significance":"If the cited estimates from [26] are valid, the theorem resolves Yau's sharp Euclidean comparison question in a substantial class of manifolds and supplies the expected Euclidean rigidity. The proof strategy is well organized, and the reduction to the four classification cases is clean. The manuscript is transparent about its dependence on external preprints, but the key L1-convergence steps in Section 4 are not sufficiently justified as written, and the β=0 case needs a clearer justification of the diameter criterion. There is no machine-checked or computational component; the verification is analytic and relies on unpublished results.","major_comments":[{"comment":"The proof that P_R converges to P_infty in L1(B_Lambda) is incomplete. Equation (4.12) gives only pointwise convergence F(t)-1 -> 0, while the integral in (4.11) requires a uniform integrable bound on s^(n-1) exp((n-2)(wbar(Rs)-wbar(R))) |F(Rs)-1|. Estimate (4.10) controls the radial exponential factor but gives no control on the spherical average excess F(Rs)-1. Without an explicit domination or rate from [26, Lemma 4.7], the conclusion ||P_R - Pbar_R||_{L1(A_{delta,Lambda})} -> 0 does not follow. This is load-bearing because Theorem 4.2 uses (4.3) in Step 3 and (4.4) in Step 5, and Proposition 5.4 transfers the resulting estimates to the manifold.","section":"Section 4.1, Eq. (4.11)-(4.12)"},{"comment":"The convergence of Q_R is imported from an unnumbered statement on page 25 of [26]. The quotient in (4.15) is asserted but not proved in the manuscript and is not tied to a numbered lemma. Since (4.15) is used to obtain ||Q_R - Qbar_R||_{L1(A_{delta,Lambda})} -> 0 and hence (4.4), and since (4.4) is needed for the weighted-mass convergence (4.22) and for the finite-family product convergence (4.23), this dependence must be made precise. Please state the full result with hypotheses and a precise reference, or include a proof in an appendix.","section":"Section 4.1, Eq. (4.15)"},{"comment":"The beta=0 case has a gap in the application of Carron's criterion. The argument computes a(rho)/rho -> 0 and hence diam_{gbar}(partial B^{gbar}_R)=o(R) for the radial metric, and then asserts that (2.1) implies diam_g(partial B^g_R)=o(R). Inclusions of balls alone do not control the g-diameter of the g-spheres; one needs either the stronger statement from [26, Page 29] quoted as a numbered result or an independent proof. Without this, Proposition 2.1 is not applicable, so the conclusion H_d(R^n,g)=R in the beta=0 case is not fully justified.","section":"Section 2, proof of Theorem 1.1"}],"minor_comments":[{"comment":"The displayed computation after the volume estimate contains a malformed expression involving ln(e^{n(1-m)i(r)} - C); it should be rewritten as a clear chain of inequalities with correct parentheses.","section":"Section 3, proof of Proposition 3.7"},{"comment":"The citation to [24] for the three-dimensional nonnegative sectional curvature result has the title \"Eigenvalues on spheres,\" which does not match the claimed result about dimensions of harmonic functions; please verify the reference.","section":"Introduction, reference [24]"},{"comment":"The notation H_p(R^n,g) and h_p(R^n,g) for non-integer p is used without a definition; also, the assertion that no spherical mode of degree at least two can occur in a harmonic function of growth at most p should be justified by a short argument or a reference.","section":"Remark 6.3"},{"comment":"There are many typographical and OCR-style formatting issues, including inconsistent rendering of the underline in quantities such as fbar and wbar, a corrupted title on the first page, and missing spaces in inline formulas. A careful copyedit is needed.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The central theorem is conditional on results in the unpublished preprint [26] (and, to a lesser extent, [24] and [2]). If the journal accepts papers relying on such preprints, the authors should be asked either to make [26] publicly available in a verifiable form or to include the needed statements as appendices. The editor may also wish to check whether reference [24] is correctly cited, since its title appears unrelated to the claimed result."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Right now, this is the strongest available answer to Yau's sharp Euclidean comparison on locally conformally flat manifolds with nonnegative Ricci curvature. The main theorem — h_d(M) ≤ h_d(R^n), with rigidity — is new, and the paper's three-part strategy (beta=0 Liouville, distance comparison, weighted blow-down to a cone) is coherent and mostly well executed. The beta=0 case and the non-integer d example in Remark 6.3 are useful extras.\n\nThe proof leans heavily on Ma's preprint [26] for the refined radial asymptotics, the exact volume ratio formula, the spherical exponential-average estimate, and the annular exceptional-set estimate. That is the real caveat. The paper is not self-contained at the load-bearing points; it depends on results that have not yet appeared in a refereed venue. If those results hold, the argument goes through. The reader's CONDITIONAL verdict is fair.\n\nThe stress-test's specific concern about Proposition 4.1 does not hold up. To get L1 convergence of P_R on an annulus, you need a domination of the integrand in (4.11). You have it: F(t)→1 implies |F(Rs)-1| is bounded for large Rs, and (4.9) or (4.10) bounds the radial factor by an integrable function of s on [δ,Λ]. The paper should have said \"dominated convergence\" but the gap is expository, not mathematical.\n\nThere are smaller issues. (4.10) is stated with m+ε for all s, but for s>1 the integration of r w'(r) → -m gives m-ε; the proof only uses (4.10) for s<1, so this does not damage the argument. The definition of f in Proposition 3.2 is repeated with a harmless factor change. The classification and the cylindrical/flat cases are handled cleanly.\n\nThis paper deserves a serious referee. The referee's main job is to verify the imported statements from [26] and [27]; the authors should be asked to state the needed results precisely, or include proofs in an appendix. I would send it out.","headline":"Strong new result on Yau's sharp Euclidean comparison for locally conformally flat manifolds with nonnegative Ricci curvature, with a real but explicit dependence on a recent preprint.","tokens_in":28284,"tokens_out":12931,"would_cite":true,"duration_ms":131342,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["53C21","53C20","31C05"],"pacs":[],"model":"deepseek-v4-flash","headline":"On complete locally conformally flat manifolds with nonnegative Ricci curvature, polynomial-growth harmonic functions obey the Euclidean dimension bound, and equality for positive degree forces Euclidean space.","keywords":["harmonic functions of polynomial growth","locally conformally flat manifolds","nonnegative Ricci curvature","asymptotic volume ratio","Euclidean volume growth","weighted compactness","metric cone at infinity","sharp dimension bound"],"falsifier":"Compute the asymptotic volume ratio $\\beta$ and the asymptotic exponent $m$ of the conformal factor on a candidate locally conformally flat metric on $\\mathbb{R}^n$ with nonnegative Ricci curvature and $\\beta>0$: the paper predicts $\\beta=(1-m)^{n-1}$ and $\\dim H_d(\\mathbb{R}^n,g)\\le \\dim H_{d-1}(\\mathbb{R}^n)$. A metric for which $\\beta\\ne (1-m)^{n-1}$, or for which the dimension exceeds $\\dim H_{d-1}(\\mathbb{R}^n)$, would falsify the argument. In the rotationally symmetric warped-product family $g=d\\rho^2+\\psi(\\rho)^2 g_{S^{n-1}}$ with $\\psi(\\rho)=a\\rho+b$ at infinity, the explicit dimension formula given in the paper can be checked by direct ODE analysis for each degree $d$.","tokens_in":27254,"feed_emoji":"📐","tokens_out":20761,"duration_ms":179066,"temperature":0.7,"pith_summary":"Two central questions about harmonic functions on complete manifolds with nonnegative Ricci curvature are whether the space of polynomial-growth harmonic functions of degree at most $d$ is finite-dimensional, and whether its dimension is bounded by the Euclidean one. The first is known; this paper settles the second affirmatively for locally conformally flat manifolds of dimension at least three. When the asymptotic volume ratio vanishes, every such harmonic function is constant. When the ratio is positive, the dimension is at most $h_{d-1}(\\mathbb{R}^n)$, strictly below $h_d(\\mathbb{R}^n)$. Equality for any integer $d\\ge 1$ forces the manifold to be isometric to Euclidean space.","feed_headline":"Nonnegative-Ricci conformally flat spaces obey the Euclidean bound","feed_subtitle":"The degree-d space of polynomial-growth harmonic functions is bounded by the Euclidean one; equality forces the manifold to be Euclidean.","key_machinery":"The carrying mechanism is the blow-down of the conformal metric to a metric cone at infinity. For $g=e^{2w}g_0$ with positive asymptotic volume ratio, the asymptotic exponent $m$ of the conformal factor controls the geometry through two facts established in [26]: the refined radial asymptotic decomposition $w(x)=\\underline{w}(|x|)+o(1)$ off a strong exceptional set, and the exact volume-ratio identity $\\beta=(1-m)^{n-1}$. Setting $a=1-m$, the rescaled weights $P_R(y)=e^{(n-2)(w(Ry)-\\underline{w}(R))}$ and $Q_R(y)=e^{n(w(Ry)-\\underline{w}(R))}$ converge in $L^1$ on balls to $|y|^{-m(n-2)}$ and $|y|^{-mn}$, whose Laplace–Beltrami operator is the limiting equation on the cone $C_a=(0,\\infty)\\times S^{n-1}$ with metric $d\\rho^2+a^2\\rho^2 g_{S^{n-1}}$, $\\rho=|y|^a$. Harmonic functions on this cone expand as sums of $r^{\\sigma_\\ell}$ times spherical harmonics, where $\\sigma_\\ell$ is the positive root of $\\sigma(\\sigma+a(n-2))=\\ell(\\ell+n-2)$. The inequality $\\sigma_d>ad$ creates a spectral gap, and the log-convexity of the weighted $L^2$ mass on the cone yields a uniform one-step doubling propagation that transfers the gap back to the original manifold. A weighted compactness theorem, built on [26] and a mean-value inequality [17], allows simultaneous blow-down of a finite-dimensional family of harmonic functions while preserving linear independence.","core_discovery":"The Main Theorem states that if $(M^n,g)$, $n\\ge 3$, is a connected complete noncompact locally conformally flat manifold with nonnegative Ricci curvature, then $h_d(M)\\le h_d(\\mathbb{R}^n)$ for every integer $d\\ge 0$, and equality for some $d\\ge 1$ implies $(M,g)$ is isometric to $\\mathbb{R}^n$. The proof reduces the general case to globally conformal metrics $g=e^{2f}g_0$ on $\\mathbb{R}^n$ via a classification of locally conformally flat manifolds with nonnegative Ricci curvature. In the vanishing-volume-ratio case, Theorem 1.1 gives $H_d(\\mathbb{R}^n,g)=\\mathbb{R}$ for every finite $d$. In the Euclidean-volume-growth case, Theorem 1.2 gives $\\dim H_d(\\mathbb{R}^n,g)\\le \\dim H_{d-1}(\\mathbb{R}^n,g_0)<\\dim H_d(\\mathbb{R}^n,g_0)$, so the gap to Euclidean space is at least one polynomial degree. The equality analysis treats the four classified cases and singles out Euclidean space. The dimension bound also holds for non-integer degrees, but the rigidity statement does not, as explicit rotationally symmetric conformal metrics demonstrate.","pith_inferences":["A natural extension is to ask whether the strict drop $h_d(M)\\le h_{d-1}(\\mathbb{R}^n)$ persists for other classes of manifolds with a non-flat tangent cone at infinity, since the proof only needs the cone's spectral gap, not a full classification.","The weighted compactness statement of the paper does not require uniqueness of the tangent cone at infinity; only $L^1$ convergence of the rescaled weights is used. That suggests the same blow-down template could apply to metrics that are merely conformally asymptotic to a cone.","The non-integer counterexamples built from warped products show the integer restriction in the rigidity statement is essential; for real degrees $p$ between 1 and 2 one can have equality without the manifold being Euclidean.","Because the proof reduces the theorem to the four classified shapes of locally conformally flat manifolds, a testable next step is whether the same four-case reduction, combined with a similar blow-down, yields the Euclidean comparison for other curvature sign conditions where the bound is still open."],"forward_implications":["On every complete noncompact locally conformally flat manifold of dimension $n\\ge 3$ with nonnegative Ricci curvature, $h_d(M)\\le h_d(\\mathbb{R}^n)=\\binom{n+d-1}{n-1}+\\binom{n+d-2}{n-1}$ for each integer $d$; equality for $d\\ge 1$ forces $(M,g)$ to be isometric to $\\mathbb{R}^n$.","In the globally conformal case with positive asymptotic volume ratio, the stronger bound $h_d(M)\\le h_{d-1}(\\mathbb{R}^n)$ holds, so the dimension is at least one polynomial degree below the Euclidean value.","If the asymptotic volume ratio is zero, all polynomial-growth harmonic functions are constant, so $h_d(M)=1$ for every degree.","The Euclidean comparison also holds for non-integer growth degrees, but equality rigidity does not; there exist non-flat conformal metrics on $\\mathbb{R}^n$ with $h_p(M)=h_p(\\mathbb{R}^n)$ for $1<p<2$.","The uniform doubling estimate $S_h(2R)\\le 2^{2N}S_h(R)$ implies $S_h(R)\\le 2^{2N}S_h(R/2)$ for all large dyadic radii simultaneously for every $h\\in H_d(g)$, giving the common growth control needed for the simultaneous blow-down."],"supporting_citations":[{"why":"Supplies the refined radial asymptotic decomposition of the conformal factor off a strong exceptional set, the exact volume-ratio formula $\\beta=(1-m)^{n-1}$, and the $L^1$ convergence of rescaled weights used in the blow-down.","marker":"[26]"},{"why":"Provides the asymptotic exponent $-m$ of the conformal factor, the exceptional-set behavior, and nonlinear potential-theoretic estimates used in the upper distance bound.","marker":"[27]"},{"why":"Supplies the asymptotically conic blow-down and spectral/log-convexity template that Sections 4 and 5 adapt to conformally flat metrics.","marker":"[23]"},{"why":"Provides the strategy of comparing the conformal distance with the Euclidean distance, which Section 3 adapts to the present setting.","marker":"[20]"},{"why":"Part of the classification of complete locally conformally flat manifolds with nonnegative Ricci curvature into the four cases handled in Section 6.","marker":"[4]"},{"why":"Completes the classification used to reduce the theorem to the globally conformal, spherical, cylindrical, and flat cases.","marker":"[30]"},{"why":"Establishes finite dimensionality of $H_d(M)$, so the dimensions $h_d(M)$ in the theorem are finite and the comparison is meaningful.","marker":"[8]"},{"why":"Supplies the mean-value inequality for subharmonic functions on manifolds with nonnegative Ricci curvature, used to obtain uniform local $L^\\infty$ bounds in the weighted compactness theorem.","marker":"[17]"},{"why":"Provides the linear-growth rigidity fact that $h_1(M)=n+1$ forces $M$ to be isometric to $\\mathbb{R}^n$, used to exclude equality in the classified cases.","marker":"[7]"}],"fun_headline_variants":["Sharp Euclidean bound for harmonic functions on conformally flat Ricci-nonnegative spaces","Conformally flat Ricci-nonnegative: Euclidean bound on harmonic growth","Euclidean bound for harmonic functions on nonnegative-Ricci conformally flat","Harmonic growth bound hits Euclidean limit on nonnegative-Ricci conformally flat","Euclidean bound sharp; equality forces Euclidean manifold"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that, off a sparse exceptional set, the conformal factor is asymptotically radial and satisfies the exact volume-ratio identity $\\beta=(1-m)^{n-1}$; if that radial approximation fails, the distance comparison and the blow-down to the cone both collapse.","fun_headline_variants_meta":{"raw":{"variants":["Sharp Euclidean bound for harmonic functions on conformally flat Ricci-nonnegative spaces","Conformally flat Ricci-nonnegative: Euclidean bound on harmonic growth","Euclidean bound for harmonic functions on nonnegative-Ricci conformally flat","Harmonic growth bound hits Euclidean limit on nonnegative-Ricci conformally flat","Euclidean bound sharp; equality forces Euclidean manifold"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.002427,"raw_usage":{"total_tokens":9338,"prompt_tokens":962,"completion_tokens":8376,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":578,"completion_tokens_details":{"reasoning_tokens":8279}},"tokens_in":578,"tokens_out":8376,"duration_ms":62459,"temperature":1.0,"reasoning_tokens":8279,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T22:11:03.052870+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the asymptotic volume ratio $\\beta$ and the asymptotic exponent $m$ of the conformal factor on a candidate locally conformally flat metric on $\\mathbb{R}^n$ with nonnegative Ricci curvature and $\\beta>0$: the paper predicts $\\beta=(1-m)^{n-1}$ and $\\dim H_d(\\mathbb{R}^n,g)\\le \\dim H_{d-1}(\\mathbb{R}^n)$. A metric for which $\\beta\\ne (1-m)^{n-1}$, or for which the dimension exceeds $\\dim H_{d-1}(\\mathbb{R}^n)$, would falsify the argument. In the rotationally symmetric warped-product family $g=d\\rho^2+\\psi(\\rho)^2 g_{S^{n-1}}$ with $\\psi(\\rho)=a\\rho+b$ at infinity, the explicit dimension formula given in the paper can be checked by direct ODE analysis for each degree $d$.","supporting_citations":[{"cited_title":"Ma and J","cited_arxiv_id":null,"evidence_quote":"Provides the asymptotic exponent $-m$ of the conformal factor, the exceptional-set behavior, and nonlinear potential-theoretic estimates used in the upper distance bound."},{"cited_title":"Lin,Asymptotically conic elliptic operators and Liouville type theorems, Geometric analysis and the calculus of variations, 217–238","cited_arxiv_id":null,"evidence_quote":"Supplies the asymptotically conic blow-down and spectral/log-convexity template that Sections 4 and 5 adapt to conformally flat metrics."},{"cited_title":"Li and L.-F","cited_arxiv_id":null,"evidence_quote":"Provides the strategy of comparing the conformal distance with the Euclidean distance, which Section 3 adapts to the present setting."},{"cited_title":"Carron and M","cited_arxiv_id":null,"evidence_quote":"Part of the classification of complete locally conformally flat manifolds with nonnegative Ricci curvature into the four cases handled in Section 6."},{"cited_title":"Zhu,The classification of complete locally conformally flat manifolds of nonnegative Ricci curva- ture, Pacific J","cited_arxiv_id":null,"evidence_quote":"Completes the classification used to reduce the theorem to the globally conformal, spherical, cylindrical, and flat cases."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes finite dimensionality of $H_d(M)$, so the dimensions $h_d(M)$ in the theorem are finite and the comparison is meaningful."},{"cited_title":"Li and R","cited_arxiv_id":null,"evidence_quote":"Supplies the mean-value inequality for subharmonic functions on manifolds with nonnegative Ricci curvature, used to obtain uniform local $L^\\infty$ bounds in the weighted compactness theorem."},{"cited_title":"Cheeger, T.H","cited_arxiv_id":null,"evidence_quote":"Provides the linear-growth rigidity fact that $h_1(M)=n+1$ forces $M$ to be isometric to $\\mathbb{R}^n$, used to exclude equality in the classified cases."}],"review_version":1}