{"id":"45cb5ca5-3d51-42e1-8834-597c0320d618","arxiv_id":"1908.09958","paper_version":4,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"A closed Riemannian manifold whose curvature operator has positive average of its lowest n-p eigenvalues has zero p-th and (n-p)-th Betti numbers, with quantitative diameter-dependent estimates.","lead":"This paper proves new vanishing and estimation theorems for Betti numbers of closed Riemannian manifolds under a curvature condition that averages the lowest eigenvalues of the curvature operator. It generalizes earlier results by Meyer, Gallot-Meyer, and Gallot, and yields new consequences such as vanishing Betti numbers in dimensions 5 and 6 when the curvature operator is 3-positive.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified","rationale":"The reader's ACCEPT verdict is justified. The central claim depends on Lemma 2.2(c), Proposition 2.9(b), and Lemma 2.1, and these align: the sharp pointwise constant p is exactly what makes the relevant eigenvalue count n−p, and the sharpness examples in Section 4 concretely support that constant. The application of Theorem 1.12 is also sound: the eigenvalue-sum hypothesis yields the required Ricci lower bound, and the coefficient p(n−p) appears correctly in the exponent of the Betti number estimate. The only potential subtlety is the orientation double cover, but because the theorem's universal constant is not explicit, any diameter factor from the cover is absorbed. I therefore see no reason to change the verdict, and the reader's weakest-assumption identification matches the step I would scrutinize first.","tokens_in":22249,"tokens_out":25459,"duration_ms":234157,"concrete_test":"Independently re-verify the two hypotheses feeding the quantitative bound: (1) for p ≤ ⌊n/2⌋ and any unit L ∈ so(V), compute max_{ω ∈ Λ^p, |ω|=1} |Lω| and confirm it equals √p, so Lemma 2.2(c) is sharp; (2) confirm by the Ky Fan eigenvalue sum bound that the condition (λ1+...+λ_{n−p})/(n−p) ≥ κ implies Ric(X,X) ≥ (n−1)κ for every unit X, so Theorem 1.12's hypotheses are satisfied with the same κ.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Non-finding. Theorem A is supported by a clear, internally consistent chain: Lemma 2.2(c) supplies the sharp pointwise bound |Lω|^2 ≤ p|ω|^2|L|^2 for p ≤ n/2; Proposition 2.9(b) gives |\\hat ω|^2 = p(n−p)|ω|^2; and Lemma 2.1 turns this into g(R(\\hat ω),\\hat ω) ≥ κ|\\hat ω|^2 under the average-eigenvalue hypothesis. I checked the sign handling in Lemma 2.1, where the negative coefficients on the first ⌊C⌋ terms are the reason the pointwise bound is used as an upper bound, and I verified that the eigenvalue hypothesis implies Ric ≥ (n−1)κ, so Theorem 1.12 applies with the same κ. The non-orientable reduction via the orientation double cover can in principle increase the diameter, but since the constant C(n,κD^2) in Theorem A is non-explicit, any fixed diameter factor is absorbed without changing the claim. The only external input, Theorem 1.12, is quoted without proof, but it is a standard Li–Gallot estimate rather than a circular step. No load-bearing flaw was found.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a new curvature condition for the Bochner technique: instead of a pointwise lower bound on the curvature operator, it uses a lower bound on the average of the lowest n-p eigenvalues. Theorem A asserts that for closed connected n-manifolds, positivity of λ1+...+λ_{n-p} forces the Betti numbers b_p and b_{n-p} to vanish, and a normalized lower bound κ≤0 together with a diameter bound gives a quantitative exponential estimate for b_p. The proof relies on a new algebraic Lemma 2.1 that converts an estimate on the action of so(V) on p-forms into a curvature-term inequality, sharp estimates in Lemma 2.2 and Proposition 2.9, and the Li-Gallot estimates of Theorem 1.12. The paper also proves a Tachibana-type rigidity theorem for Einstein manifolds (Theorem B), a Weyl-tensor analogue, and constructs examples that demonstrate the sharpness of the constants and the independence of the new condition from Ricci-flow-invariant curvature conditions.","tokens_in":22476,"tokens_out":27390,"duration_ms":256715,"significance":"Modulo the flaw in the proof of Theorem B identified below, the paper makes a substantial contribution. The average-eigenvalue condition genuinely extends earlier work of Meyer, Gallot-Meyer, and Gallot, and it applies to 3-positive curvature operators in dimensions 5 and 6, a condition that is not preserved by the Ricci flow. The main Theorem A is a concrete, falsifiable statement with explicit dependence on κD², and the paper contains detailed, checkable computations and examples showing optimality of the estimates in Section 4. The proof of Theorem A is internally consistent and does not appear to rely on circular reasoning; the only external input is the standard Li-Gallot estimate.","major_comments":[{"comment":"The displayed inequality in the dimension-four part of the proof of Theorem B is false as stated. The exact Singer-Thorpe formula gives g(R(ˆRm), ˆRm)=16(E(λ1,λ2,λ3)+E(λ4,λ5,λ6)), where E(x,y,z)=x(y−z)^2+y(x−z)^2+z(x−y)^2. The proof claims that, after relabeling so that λ1+λ2≥0, λ1≤λ2≤λ3, and λ4,λ5,λ6≥0, one has g≥16{(λ1+λ2)(λ1−λ3)^2+λ3(λ1−λ2)^2}. Consider an Einstein algebraic curvature operator on R^4 whose Singer-Thorpe eigenvalues are (1,2,3) on one self-dual half and (2,2,2) on the other. It satisfies the first Bianchi identity (1+2+3=2+2+2), is Einstein, and is 2-positive since the two smallest global eigenvalues are 1 and 2 with sum 3>0. For this operator the left side equals 16·12=192, while the claimed right side equals 16·15=240. Thus the inequality is not valid. The conclusion of Theorem B in dimension 4 may still be correct, and one can instead analyze the equality cases of E directly, but the argument as written needs to be corrected.","section":"§3, proof of Theorem B, dimension n=4"}],"minor_comments":[{"comment":"The proof of the p-form estimate is quite terse; the sentence explaining why repeated coefficients in the sum force the corresponding wedge components to vanish would be easier to follow with a short example or a more detailed algebraic justification.","section":"§2, Lemma 2.2(c)"},{"comment":"When passing to the orientation double cover, the diameter bound may double; the statement of Theorem A is unaffected because C(n,κD²) is non-explicit and can absorb the factor, but this should be stated explicitly to avoid confusion.","section":"§3, proof of Theorem A"},{"comment":"The motivating assertion that CP² is 3-positive with b₂=1 would benefit from a pointer to the eigenvalue computation in Remark 2.7, so that the reader can verify it without reconstructing the Singer-Thorpe basis.","section":"Introduction, Question"},{"comment":"The notation 'C1-close' should be typeset as 'C^1-close', and the phrase 'arbitrary negative minimum' reads more clearly as 'arbitrarily negative minimum'.","section":"§4, Example 4.1"}],"recommendation":"major_revision","confidential_remarks":"The central Theorem A is sound and the paper deserves publication after the localized correction to the dimension-four case of Theorem B. The flaw does not affect Theorem A, but since Theorem B is advertised as a main result, the proof must be fixed. I would also encourage the authors to clarify Lemma 2.2(c) and the orientation-cover diameter point."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper is a genuine advance. Petersen and Wink replace pointwise lower bounds on the curvature operator with an average of its lowest eigenvalues, and they get both vanishing and quantitative Betti number estimates from that weaker condition. The key mechanism is Lemma 2.1, a neat eigenvalue-averaging argument, and Lemma 2.2, which supplies sharp constants for forms, symmetric tensors, and curvature tensors. Those two lemmas are the real content, and they are proved directly with explicit examples showing optimality. Theorem A is a true generalization of Meyer, Gallot-Meyer, and Gallot, and the applications—like vanishing for 3-positive curvature operators in dimensions 5 and 6—are genuinely new. Theorem B for Einstein manifolds is a nice bonus.\n\nThe proofs are careful and internally consistent. I checked the logic of Lemma 2.1: the step where the negative coefficients on the low eigenvalues are handled using the pointwise bound is correct, and the sign handling works. The reduction to the Li–Gallot theorem (Theorem 1.12) is legitimate because the eigenvalue hypothesis indeed implies the needed Ricci lower bound. The examples in Section 4 convince me that the constants in Lemma 2.2 are sharp, and Example 4.6 correctly shows that Theorem A cannot be weakened to (n−1)-positivity in general.\n\nSoft spots are minor. The quantitative part of Theorem A depends on the external Li–Gallot estimate, quoted without proof; that is standard, but it means the diameter-dependent constant is non-explicit and the bound is only as strong as that input. The non-orientable case is handled by passing to the orientation double cover, which could in principle double the diameter, but since the constant is non-explicit this does not affect the claim. Proposition 3.5 is more technical and somewhat orthogonal to the main theorems; it could be trimmed without hurting the paper. None of these issues threaten the central argument.\n\nThis is a solid paper that deserves serious refereeing. I would bring it to a reading group and would cite it in my own work. My verdict: accept after the usual checks, not because it is flashy but because the mathematics is correct, the generalization is meaningful, and the exposition is honest and detailed.","headline":"A clean, substantive generalization of Bochner-type vanishing and estimation theorems via average-of-lowest-eigenvalues conditions; the main arguments hold up on close reading.","tokens_in":22965,"tokens_out":1202,"would_cite":true,"duration_ms":14150,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["53B20","53C20","53C21","53C23","58A14"],"pacs":[],"model":"deepseek-v4-flash","headline":"On a closed Riemannian manifold, if the sum of the lowest $n-p$ eigenvalues of the curvature operator is positive, then the $p$-th and $(n-p)$-th Betti numbers vanish; under a curvature lower bound and diameter bound, a quantitative…","keywords":["Bochner technique","Betti numbers","curvature operator","eigenvalue sums","Lichnerowicz Laplacian","harmonic forms","Einstein manifolds","diameter estimates"],"falsifier":"Exhibit a closed Riemannian $5$-manifold whose curvature operator is $3$-positive and whose second Betti number is nonzero; Theorem A predicts $b_2=0$, so such a manifold would disprove the theorem. Alternatively, a direct computation of $|L\\omega|^2$ on a $p$-form with $L\\in\\mathfrak{so}(V)$ that exceeds $p|\\omega|^2|L|^2$ would overturn Lemma 2.2(c), the key estimate on which the proof rests.","tokens_in":22071,"feed_emoji":"📐","tokens_out":10848,"duration_ms":97502,"temperature":0.7,"pith_summary":"This paper proves that the Betti numbers of a closed Riemannian manifold are controlled by the average of the lowest eigenvalues of its curvature operator, not by pointwise positivity of the whole operator. The central theorem says that if the sum of the lowest $n-p$ eigenvalues is positive, then the $p$-th and $(n-p)$-th Betti numbers vanish, and under a normalized lower bound $\\kappa \\le 0$ with a diameter bound $D$, the $p$-th Betti number is at most $\\binom{n}{p}\\exp(C\\sqrt{-\\kappa D^2 p(n-p)})$. This generalizes classical Bochner-type vanishing results of Meyer, Gallot--Meyer, and Gallot to a curvature condition that is not preserved by the Ricci flow. The proof works by feeding a sharp estimate on how rotations act on $p$-forms into the Li--Gallot quantitative Bochner technique. A reader should care because it gives vanishing and quantitative topology from an average eigenvalue condition, and because in dimensions $5$ and $6$ it yields Betti number vanishing under a $3$-positive curvature operator.","feed_headline":"Averaging the lowest curvature eigenvalues kills Betti numbers","feed_subtitle":"A new Bochner-style proof: a positive sum of the lowest n-p eigenvalues makes b_p and b_{n-p} vanish.","key_machinery":"The central object is the hat tensor $\\hat T \\in \\Lambda^2 V \\otimes T^{(0,k)}(V)$, defined implicitly by $g(L,\\hat T(X_1,\\dots,X_k)) = (LT)(X_1,\\dots,X_k)$ for every rotation $L \\in \\mathfrak{so}(V)$; it packages how every rotation acts on a tensor. The key estimate is Lemma 2.1, which converts the bound $|LT|^2 \\le \\frac{1}{C}|\\hat T|^2|L|^2$ into the pointwise curvature-term inequality $g(R(\\hat T),\\hat T) \\ge \\kappa|\\hat T|^2$ whenever the average of the lowest $\\lfloor C\\rfloor$ eigenvalues of the curvature operator $R$ is at least $\\kappa$. For $p$-forms this holds with $C=n-p$, and the constant is sharp by the examples in Section 4. The Li--Gallot theorem (Theorem 1.12) then converts the pointwise Bochner inequality into a dimension bound on the kernel of the Hodge Laplacian, which is the $p$-th Betti number.","core_discovery":"The paper's discovery is that the curvature term in the Bochner formula for $p$-forms can be controlled by the sum of the lowest $n-p$ eigenvalues of the curvature operator. Concretely, Lemma 2.1 shows that if a tensor $T$ satisfies $|LT|^2 \\le \\frac{1}{C}|\\hat T|^2|L|^2$ for all $L \\in \\mathfrak{so}(V)$, then an average lower bound $\\frac{1}{\\lfloor C\\rfloor}(\\lambda_1+\\cdots+\\lambda_{\\lfloor C\\rfloor}) \\ge \\kappa$ forces $g(R(\\hat T),\\hat T) \\ge \\kappa|\\hat T|^2$. For $p$-forms with $p \\le n/2$, Lemma 2.2(c) gives exactly this with $C=n-p$: $|L\\omega|^2 \\le p|\\omega|^2|L|^2 = \\frac{1}{n-p}|\\hat\\omega|^2|L|^2$. Combined with Proposition 2.9(b), which identifies $|\\hat\\omega|^2 = p(n-p)|\\omega|^2$, and the Li--Gallot theorem bounding the kernel of the Lichnerowicz Laplacian, this yields Theorem A. The same mechanism, with Lemma 2.2(d) for curvature tensors, gives Theorem B: Einstein manifolds satisfying a low-eigenvalue sum condition have parallel curvature, and in the strict case constant sectional curvature.","pith_inferences":["Because the condition is an average of the lowest eigenvalues rather than pointwise positivity of the full operator, the method may extend to integral curvature bounds or to metric-measure spaces with a suitable Bochner inequality; a direct test would be to replace the eigenvalue sum by its $L^q$ average and check whether the exponential estimate survives.","The sharpness examples indicate that the eigenvalue count $n-p$ is the natural threshold; one concrete open problem the paper highlights is whether there are simply connected manifolds with $\\lambda_1+\\cdots+\\lambda_{n-1}>0$ and large $b_2$, and the method here gives a way to search for them.","Since $3$-positivity is not preserved by the Ricci flow, the theorem identifies a curvature condition that other flows or surgery constructions might be able to preserve; testing whether the condition is stable under connected sums would clarify how large the class of examples is."],"forward_implications":["If the average of the lowest $n-p$ curvature eigenvalues is positive, then both $b_p(M)$ and $b_{n-p}(M)$ vanish.","Under the normalized lower bound $\\kappa \\le 0$ and diameter bound $D$, the estimate $b_p(M) \\le \\binom{n}{p}\\exp(C(n,\\kappa D^2)\\sqrt{-\\kappa D^2 p(n-p)})$ holds, so slightly negative curvature--diameter product forces the Betti number to be no larger than the binomial coefficient.","In dimensions $n=5,6$, a $3$-positive curvature operator suffices for Betti number vanishing, even though $3$-positivity is not preserved by the Ricci flow.","For closed Einstein manifolds, the low-eigenvalue sum conditions of Theorem B force the curvature tensor to be parallel; strict inequality forces constant sectional curvature.","When $\\kappa=0$ in Theorem A, all harmonic $p$-forms are parallel, giving a rigidity statement at the threshold."],"supporting_citations":[{"why":"Introduces the Bochner technique that the paper extends to average eigenvalue conditions.","marker":"[Boc46]"},{"why":"Proves vanishing of Betti numbers for positive curvature operator, the classical result Theorem A generalizes.","marker":"[Mey71]"},{"why":"Handles nonnegative curvature operator and rigidity, setting the Gallot--Meyer context for the new average condition.","marker":"[GM75]"},{"why":"Provides the Sobolev-constant and spectrum estimates behind the dimension bound in Theorem 1.12.","marker":"[Li80]"},{"why":"Supplies the quantitative Bochner estimate and exponential Betti-number bound used as Theorem 1.12.","marker":"[Gal81]"},{"why":"The derivative-of-regular-representation viewpoint on the Hodge Laplacian that Lemma 2.1 generalizes.","marker":"[Poo80]"},{"why":"Singer--Thorpe bases for four-dimensional Einstein curvature operators are used in Theorem B and the sharpness examples.","marker":"[ST69]"},{"why":"Records that 3-positivity is not preserved by the Ricci flow, motivating the Bochner-proof independence from Ricci flow.","marker":"[BW08]"},{"why":"Tachibana's Einstein-manifold theorem that Theorem B extends.","marker":"[Tac74]"},{"why":"Source for the Lichnerowicz Laplacian formulas connecting Ricci terms to hat tensors.","marker":"[Pet16]"}],"fun_headline_variants":["Low eigenvalue sums force Betti number vanishing","Bochner trick: average of lowest eigenvalues kills cohomology","Curvature eigenvalues' average dictates Betti numbers","Vanishing Betti numbers via averaged curvature eigenvalues","Average low curvature eigenvalues, lose Betti numbers"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole theorem rests on the sharp constant in the estimate $|L\\omega|^2 \\le p|\\omega|^2|L|^2$ for every $p$-form $\\omega$ and rotation $L$; if that constant were smaller, the proof would need more eigenvalues in the sum and the vanishing statement would fail.","fun_headline_variants_meta":{"raw":{"variants":["Low eigenvalue sums force Betti number vanishing","Bochner trick: average of lowest eigenvalues kills cohomology","Curvature eigenvalues' average dictates Betti numbers","Vanishing Betti numbers via averaged curvature eigenvalues","Average low curvature eigenvalues, lose Betti numbers"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000212,"raw_usage":{"total_tokens":1426,"prompt_tokens":960,"completion_tokens":466,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":576,"completion_tokens_details":{"reasoning_tokens":392}},"tokens_in":576,"tokens_out":466,"duration_ms":5213,"temperature":1.0,"reasoning_tokens":392,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T10:56:31.829069+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Exhibit a closed Riemannian $5$-manifold whose curvature operator is $3$-positive and whose second Betti number is nonzero; Theorem A predicts $b_2=0$, so such a manifold would disprove the theorem. Alternatively, a direct computation of $|L\\omega|^2$ on a $p$-form with $L\\in\\mathfrak{so}(V)$ that exceeds $p|\\omega|^2|L|^2$ would overturn Lemma 2.2(c), the key estimate on which the proof rests.","supporting_citations":[],"review_version":1}