{"id":"47ee1562-9737-4c1a-bb0d-9f7581d29476","arxiv_id":"2509.05802","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A long-time Ricci flow with Ric ≥ -ψ/t and sufficiently large injectivity radius forces the manifold to be diffeomorphic to R^n, improving dimension-4 small-curvature-concentration results.","lead":"A new criterion says a complete non-compact manifold is diffeomorphic to Euclidean space if a long-time Ricci flow keeps Ricci curvature bounded below by -ψ/t and injectivity radius growing at least like √t. The result upgrades earlier homeomorphism conclusions to diffeomorphism conclusions in dimension 4.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 1.1's hypothesis inj(g(t)) ≥ β/√t is weaker than the β√t bound the proof actually requires; the theorem as stated is not established.","rationale":"The reader's weakest_assumption pinpoints the exact gap: the proof uses balls of radius ~√t, but the stated injectivity radius is β/√t. I re-examined both methods. Method 1's exhaustion uses R_i ≈ (β/2)√t_i; Method 2's Ω is essentially {|v| ≤ (3β/4)√t}. In both cases the required radius exceeds β/√t for all sufficiently large t, so the stated hypothesis does not justify the exponential map being a diffeomorphism there. The paper's own applications and Remark 1.1 confirm that the intended lower bound is β√t (or βt^κ, κ>0), so the defect is likely a typo rather than a deep flaw. Since the main theorem as printed is unsupported, the reader's CONDITIONAL verdict is appropriate; no verdict change is needed. I do not see a second load-bearing concern: the applications' logic is standard and the corrected theorem would imply them.","tokens_in":7208,"tokens_out":8314,"duration_ms":82290,"concrete_test":"At Section 2, Method 2, set β=1 and t=4 in the stated hypothesis: inj(g(4)) ≥ 1/2, while Ω requires exp to be a diffeomorphism on |v| ≤ (3/4)·2 = 1.5. Since 1.5 > 0.5, the proof's assertion that (ii) makes Exp|_Ω a diffeomorphism is not valid. Replace (ii) by inj(g(t)) ≥ β√t, and the same numbers give 1.5 ≤ 2, so the step holds; the applications already establish this stronger bound. This single numerical check settles that the theorem statement must be corrected.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 2, Method 2 defines M' = {d_{g(t)}(x,x0) = (β/2)√t} and Ω = {d ≤ (3β/4)√t}, then asserts 'by the injectivity condition (ii) again, Exp|_Ω is a diffeomorphism.' This assertion requires inj(g(t)) ≥ (3β/4)√t, since Exp maps (v,t) to (exp_{x0,g(t)} v, t) and must be injective on all |v| ≤ (3β/4)√t. The stated hypothesis inj(g(t)) ≥ β/√t only guarantees this for |v| < β/√t. For t > 4/3, (3β/4)√t > β/√t, so the claim is unjustified. Method 1 has the same problem: R_i = (1/2 − 1/(2i+1))β√t_i is used as a radius on which exp_t is a diffeomorphism, far exceeding β/√t_i. The applications (proofs of Corollaries 1.1 and 1.2) all produce inj(g(t)) ≥ β√t, and Remark 1.1 says the condition can be β t^κ with κ>0 — never β/√t. Thus the intended theorem is with β√t (or β t^κ, κ>0); the printed β/√t is a statement–proof mismatch that makes Theorem 1.1 as written an overclaim.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a diffeomorphism-to-R^n criterion for complete non-compact manifolds that admit a long-time Ricci flow. Theorem 1.1 states that if (M^n, g(t)) is a smooth complete solution on [0,∞) with Ric(g(t)) ≥ −ψ/t for some 0<ψ<1/2 and inj(g(t)) ≥ β/√t for some β>0, then M is diffeomorphic to R^n. Two proofs are given: Method 1 adapts He–Lee's isotopy argument via time-dependent exponential maps, and Method 2 adapts Wang's space-time hypersurface construction. The paper then derives three applications: Corollary 1.1 upgrades the homeomorphism conclusion of Chan–Huang–Lee and Martens to diffeomorphism in dimension 4; Corollary 1.2 gives a Ricci-flow proof of the fact that a 3-dimensional complete non-compact manifold with nonnegative Ricci curvature and maximal volume growth is diffeomorphic to R^3; Corollary 1.3 gives a Sobolev-constant version of the small-curvature-concentration criterion. The central claim is therefore a useful black-box criterion, but the version stated in Theorem 1.1 is not what the proofs establish.","tokens_in":7569,"tokens_out":8598,"duration_ms":92457,"significance":"If Theorem 1.1 is corrected to the injectivity-radius bound actually used, the paper provides a clean and potentially widely applicable diffeomorphism criterion for long-time Ricci flows. The applications are nontrivial and would be valuable: Corollary 1.1 answers the dimension-4 question left open in [3] and [17], and Corollary 1.3 sharpens a known integral-curvature rigidity result. The proofs are assembled from standard tools (distance distortion under Ricci flow, exponential-map embeddings, isotopy by cutoff vector fields, space-time hypersurfaces) and the applications rely on previously established existence and injectivity-radius estimates in [3], [17], [22], [4], [7]. The argument is not circular: Theorem 1.1 is proved from standard geometric-analysis ingredients, and the applications independently supply the needed flow with the strong injectivity bound. The main deficiency is a statement–proof mismatch in the injectivity-radius hypothesis, which affects the central theorem as printed.","major_comments":[{"comment":"The stated hypothesis inj(g(t)) ≥ β/√t is weaker than what both proofs require. In Method 2, the set Ω is defined as {d_{g(t)}(x,x0) ≤ (3β/4)√t} and the text asserts 'by the injectivity condition (ii) again, Exp|_Ω is a diffeomorphism.' This requires exp_{x0,g(t)} to be injective on the ball of radius (3β/4)√t, i.e. inj(g(t)) ≥ (3β/4)√t. For t > 4/3 the stated bound β/√t is strictly smaller than (3β/4)√t, so the assertion is unjustified. In Method 1, R_i = (1/2 − 1/(2i+1)) β√t_i is used as a radius on which exp_t is a diffeomorphism, which again requires inj(g(t_i)) ≳ β√t_i/2, far above β/√t_i for large i. The proof therefore establishes the theorem only under the stronger condition inj(g(t)) ≥ β√t (or a comparable positive-power lower bound). The theorem statement must be corrected accordingly.","section":"Theorem 1.1, condition (ii); Section 2, Methods 1 and 2"},{"comment":"Remark 1.1 says that if condition (ii) is changed to inj(g(t)) ≥ βt^κ for some κ > 0, then the curvature condition only needs 0 < ψ < κ. This is internally inconsistent with the printed condition (ii), which corresponds to κ = −1/2 and cannot satisfy 0 < ψ < κ for any positive ψ. Moreover, as the proof stands, Method 1 chooses R_i proportional to √t_i, not to t_i^κ, so for 0 < κ < 1/2 the stated proof would still fail unless R_i and the inclusions in (2.1) are adapted. The intended and actually used assumption appears to be κ = 1/2, i.e. inj(g(t)) ≥ β√t. The remark should be rewritten to match the corrected theorem and proof.","section":"Remark 1.1"},{"comment":"The applications all invoke Theorem 1.1 with the stronger bound inj(g(t)) ≥ const·√t, which is exactly the bound the proofs require. Thus the applications survive the correction of Theorem 1.1. However, as the paper stands, the final steps 'the result follows from Theorem 1.1' inherit the overclaim. This is not an independent flaw, but it should be checked that the corrected Theorem 1.1 is quoted consistently throughout.","section":"Corollaries 1.1–1.3"}],"minor_comments":[{"comment":"The notation E_t = exp_t^{-1} is used as if exp_t were globally invertible, whereas on a complete non-compact manifold exp_t is surjective but not injective. The argument only needs E_t to be a local inverse on balls of radius below inj(g(t)); this should be stated explicitly.","section":"Section 2, Method 1"},{"comment":"The set M' = ∪_{t≥0} ∂B_{g(t)}(x0, β/2√t) is claimed to be a smooth manifold 'by the injectivity condition (ii)'. At t = 0 this set is the single point (x0,0), not a sphere; smoothness near the tip t = 0 deserves a brief justification (e.g. writing the hypersurface locally as (v, t) = (v, c|v|^2)).","section":"Section 2, Method 2"},{"comment":"There are several minor typographical issues: 'PIC1' should presumably be 'PIC_1'; 'bi-holomorphic' should be 'biholomorphic'; and the fraction in condition (i) of Theorem 1.1 is occasionally rendered without the slash in the extracted text. These do not affect the mathematics.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The statement–proof mismatch in Theorem 1.1 is load-bearing but readily fixable: the proofs and all applications use inj(g(t)) ≥ β√t, so correcting the theorem to that hypothesis preserves the paper's contributions. I do not see grounds for rejection, but the theorem cannot be published as stated. A careful rewrite of Theorem 1.1 and Remark 1.1 is necessary."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First, the useful core: the paper gives a sufficient condition for a complete long-time Ricci flow to force the underlying manifold to be diffeomorphic to R^n. Theorem 1.1 is not in the literature in this form, and the proof is a transparent adaptation of He–Lee and Wang. The main payoff is Corollary 1.1, which upgrades the known homeomorphism conclusion in dimension 4 to diffeomorphism; that is a real, if subfield-level, advance. The exposition is honest about Remark 1.2 being known to experts, and the citation pattern is clean — the self-citation to [3] is to the result being improved, not a load-bearing dependency. No circularity.\n\nNow the soft spot, and it is load-bearing: the stated hypothesis (ii), inj(g(t)) ≥ β/√t, is not what the proofs use. Method 1 needs the exponential map to be a diffeomorphism on balls of radius comparable to β√t_i; Method 2 explicitly asserts Exp|_Ω is a diffeomorphism for Ω = {d ≤ (3β/4)√t}, which again requires an injectivity radius of order √t. The printed β/√t is far smaller than these radii for large t. Remark 1.1 compounds the issue by saying the condition can be β t^κ with κ > 0 — which is also incompatible with β/√t. So the intended theorem is with inj ≥ β√t (equivalently β t^κ, κ > 0, and ψ < κ). As written, Theorem 1.1 is not established.\n\nThat said, this looks like a typo rather than a deep flaw. Every application in Section 3 produces inj ≥ β√t, and the proofs go through verbatim once condition (ii) is corrected. The stress-test concern is accurate; I do not see a way to make the stated β/√t condition suffice. A referee should catch this immediately, and the fix is a one-line change to the theorem statement.\n\nWho is this for? Geometric analysis readers interested in Ricci-flow proofs of topological rigidity. It is a solid extension of known techniques, not a paradigm shift. Corollary 1.2 is essentially a known fact (Liu's classification), so the novelty there is thin, but Corollary 1.1 carries the paper.\n\nVerdict: deserve a serious referee. The central idea is sound, the applications are meaningful, and the flaw is fixable. I would engage with it, but I would not cite Theorem 1.1 in its current form — only after the hypothesis is corrected.\n\nRecommendation: send to peer review, with the clear instruction that the injectivity-radius condition in Theorem 1.1 must be corrected to β√t (or β t^κ, κ>0, with ψ<κ) before acceptance.","headline":"Genuinely new diffeomorphism criterion with a clean dimension-4 application, but Theorem 1.1 as printed states the wrong injectivity-radius hypothesis (β/√t instead of β√t), so the theorem overclaims until that typo is fixed.","tokens_in":8014,"tokens_out":1657,"would_cite":true,"duration_ms":18396,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["53E20"],"pacs":[],"model":"deepseek-v4-flash","headline":"A long-time Ricci flow with 1/t Ricci decay and √t injectivity growth is diffeomorphic to R^n.","keywords":["Ricci flow","long-time solution","diffeomorphism criterion","injectivity radius","small curvature concentration","nonnegative Ricci curvature","maximal volume growth","space-time exponential map"],"falsifier":"Perform the containment check in (2.1) with inj(g(t_i)) = β/√t_i: Method 1's first inclusion requires B_{t_i}(R_i) to lie inside the domain where the exponential map is a diffeomorphism, but R_i ≈ (β/2)√t_i is a factor of t_i larger than the guaranteed injectivity radius for large i, so the inequality fails. That calculation settles whether the proof as written establishes the stated theorem.","tokens_in":7109,"feed_emoji":"📐","tokens_out":11795,"duration_ms":123294,"temperature":0.7,"pith_summary":"The note proves a topological rigidity criterion: a complete manifold that admits a long-time Ricci flow whose Ricci curvature decays like -ψ/t (ψ<1/2) and whose injectivity radius grows like √t must be diffeomorphic to R^n. The theorem statement writes the injectivity condition as β/√t, but the proofs and Remark 1.1 show the operative assumption is the square-root growth (or βt^κ with κ>ψ). Two self-contained proofs are given, one by matching exponential-coordinate charts through time and one through a space-time exponential map on the level set of the normalized distance. The criterion upgrades known small-curvature-concentration results from homeomorphism to diffeomorphism in dimension 4, gives a Ricci-flow proof that 3-dimensional nonnegative-Ricci manifolds with maximal volume growth are R^3, and yields a Sobolev-constant version in all dimensions n≥3.","feed_headline":"Long-time Ricci flow with 1/t curvature decay forces R^n topology","feed_subtitle":"A square-root injectivity-radius bound upgrades small-curvature results to diffeomorphism in dimension 4.","key_machinery":"The space-time exponential map Exp(v,t) = (exp_{x0,g(t)} v, t), together with the time-dependent distance distortion estimate d_{g(t')}(x,x0) ≤ d_{g(t)}(x,x0)(t'/t)^ψ. Because ψ<1/2, the normalized distance d/√t decreases in t along each fixed point, so the hypersurface M' = {d_{g(t)}(x,x0) = (β/2)√t} is a graph over M; injectivity-radius control makes Exp a diffeomorphism near M', and radial projection identifies M' with R^n. Method 1 instead uses the vector field ∂_t (exp_{x0,g(t)}^{-1}(x)) and a cutoff flow to build explicit diffeomorphisms between time slices.","core_discovery":"The paper's central claim is Theorem 1.1: if (M^n,g(t)) is a smooth complete long-time Ricci flow with Ric(g(t)) ≥ -ψ/t for 0<ψ<1/2 and inj(g(t)) ≥ β/√t for β>0, then M^n is diffeomorphic to R^n. Both proofs, however, invoke the injectivity-radius condition at the scale of √t: Method 1 needs the exponential map to be a diffeomorphism on balls of radius comparable to β√t, and Method 2 needs it on {|v| ≤ (3β/4)√t}; Remark 1.1 states the intended general form inj(g(t)) ≥ βt^κ with ψ<κ. The key mechanism is that the normalized distance d_{g(t)}(x,x0)/√t is monotone in time: the Ricci lower bound yields d_{g(t')}(x,x0) ≤ d_{g(t)}(x,x0)(t'/t)^ψ, so with ψ<1/2 each point x≠x0 crosses the hypersurfa","pith_inferences":["The scaling mismatch between the printed β/√t and the β√t used in both proofs means the theorem as literally stated is not established by the arguments given: if the injectivity radius truly decays like t^{-1/2}, the space-time exponential map on the required region is not guaranteed to be a diffeomorphism.","The same two methods should work for any time-dependent radius function a(t) with a(t)/√t strictly monotone and a(t) below the injectivity radius at each time, so the criterion is likely a special case of a more general monotonicity principle.","The κ>0 variant indicated in Remark 1.1 suggests the criterion extends to flows with Ricci ≥ -ψ/t and inj ≥ βt^κ for any κ>ψ, which would cover algebraic curvature decay rates other than exactly 1/t.","Corollary 1.2 uses only nonnegative Ricci and maximal volume growth in dimension 3; the Eguchi-Hanson example noted by the authors shows the dimensional ceiling is real, so the criterion is not vacuous."],"forward_implications":["Corollary 1.1: for n≥4, a complete non-compact manifold with bounded volume growth, a lower bound on local entropy, Ricci bounded below, and sufficiently small L^{n/2} curvature concentration is diffeomorphic to R^n, upgrading the homeomorphism conclusion of [3] and [17] to diffeomorphism in dimension 4.","Corollary 1.2: every 3-dimensional complete non-compact manifold with nonnegative Ricci curvature and maximal volume growth is diffeomorphic to R^3, proved here by Ricci flow rather than minimal-surface classification.","Corollary 1.3: if a complete bounded-curvature manifold satisfies a W^{1,2}-Sobolev inequality and has L^{n/2} curvature small relative to the inverse Sobolev constant, then it is diffeomorphic to R^n.","Remark 1.1: the same proof works for injectivity radius lower bounds of the form βt^κ with κ>0, provided the curvature decay exponent ψ is less than κ.","Remark 1.2: the 3-dimensional nonnegative-Ricci conclusion is sharp in dimension: the Eguchi-Hanson metric shows the analogous statement fails in higher dimensions without additional assumptions."],"supporting_citations":[{"why":"Supplies the Method-1 construction: time-dependent exponential coordinate charts and a cutoff-flow argument producing diffeomorphisms between time slices.","marker":"[13]"},{"why":"Supplies the Method-2 construction: the space-time exponential map and the projection from the level set M' to M.","marker":"[24]"},{"why":"Provides the small-curvature-concentration theorem whose dimension-4 conclusion Corollary 1.1 upgrades, and the long-time flow estimates with inj ≥ C^{-1}√t used in its proof.","marker":"[3]"},{"why":"Removes the scalar-curvature assumption in [3], yielding the flow estimates quoted in Corollary 1.1.","marker":"[17]"},{"why":"Provides short-time Ricci flow existence and the estimates used to build the long-time flow in Corollary 1.2.","marker":"[22]"},{"why":"Supplies the long-time Ricci flow existence and curvature decay used in the proof of Corollary 1.3.","marker":"[4]"},{"why":"Supplies the volume-growth lemma used to convert the Sobolev/curvature assumptions into a lower volume bound in Corollary 1.3.","marker":"[18]"},{"why":"Supplies the injectivity-radius estimate used at the end of Corollary 1.3.","marker":"[7]"}],"fun_headline_variants":["Ricci flow criterion upgrades small-curvature results to diffeomorphism","Curvature decay and injectivity radius force R^n diffeomorphism","Dimension 4: small curvature implies R^n topology","Long-time Ricci flow proves R^n diffeomorphism","Injectivity radius bound yields diffeomorphism to R^n"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The proof needs the space-time exponential map to be a diffeomorphism on balls of radius comparable to √t; the injectivity-radius lower bound β/√t stated in the theorem does not provide that, and if the true injectivity radius only decays like t^{-1/2}, the construction collapses.","fun_headline_variants_meta":{"raw":{"variants":["Ricci flow criterion upgrades small-curvature results to diffeomorphism","Curvature decay and injectivity radius force R^n diffeomorphism","Dimension 4: small curvature implies R^n topology","Long-time Ricci flow proves R^n diffeomorphism","Injectivity radius bound yields diffeomorphism to R^n"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000434,"raw_usage":{"total_tokens":2030,"prompt_tokens":706,"completion_tokens":1324,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":450,"completion_tokens_details":{"reasoning_tokens":1239}},"tokens_in":450,"tokens_out":1324,"duration_ms":13114,"temperature":1.0,"reasoning_tokens":1239,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T04:58:13.463365+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Perform the containment check in (2.1) with inj(g(t_i)) = β/√t_i: Method 1's first inclusion requires B_{t_i}(R_i) to lie inside the domain where the exponential map is a diffeomorphism, but R_i ≈ (β/2)√t_i is a factor of t_i larger than the guaranteed injectivity radius for large i, so the inequality fails. That calculation settles whether the proof as written establishes the stated theorem.","supporting_citations":[{"cited_title":"PDE10(2024), no","cited_arxiv_id":null,"evidence_quote":"Provides the small-curvature-concentration theorem whose dimension-4 conclusion Corollary 1.1 upgrades, and the long-time flow estimates with inj ≥ C^{-1}√t used in its proof."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Removes the scalar-curvature assumption in [3], yielding the flow estimates quoted in Corollary 1.1."},{"cited_title":"Topping,Local mollification of Riemannian metrics using Ricci flow, and Ricci limit spaces, Geom","cited_arxiv_id":null,"evidence_quote":"Provides short-time Ricci flow existence and the estimates used to build the long-time flow in Corollary 1.2."},{"cited_title":"Long-time Ricci flow existence and topological rigidity from manifolds with pinched scale-invariant integral curvature","cited_arxiv_id":"2403.02564","evidence_quote":"Supplies the long-time Ricci flow existence and curvature decay used in the proof of Corollary 1.3."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the volume-growth lemma used to convert the Sobolev/curvature assumptions into a lower volume bound in Corollary 1.3."},{"cited_title":"Differential Geometry17(1982), no","cited_arxiv_id":null,"evidence_quote":"Supplies the injectivity-radius estimate used at the end of Corollary 1.3."}],"review_version":1}