{"id":"ac72ce68-5c13-4b3a-aad0-d85f65726992","arxiv_id":"2504.19332","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For a nice contact form, some Reeb orbit intersects a given symplectic surface with frequency at least the surface area divided by the contact volume, without any genericity assumption.","lead":"This paper proves a quantitative lower bound on how often Reeb orbits cross a fixed symplectic surface in a three-dimensional contact manifold. The result needs no generic contact form, and it yields a new inequality between mean action of periodic orbits and the Calabi invariant for surface maps.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Lemma 4.6 integrates an almost-everywhere derivative bound for c_k without establishing absolute continuity; this step is not justified, and the main theorem relies on it.","rationale":"Central claim: for nice admissible surfaces, the sup of intersection-frequency over simple Reeb orbits is bounded below by Area(Σ,dλ)/vol(Y,λ). The proof builds inflated contact forms λδ, compares spectral invariants, and derives a contradiction from Lemma 4.6. The reader's concern about niceness is a scope limitation rather than a flaw; the paper explicitly asks in Question 1.18 whether it can be removed. The more serious issue is internal to the proof: equation (4.28) is obtained by integrating a derivative inequality that only holds almost everywhere. Since c_k is only known to be continuous and monotone, this does not follow without an additional absolute-continuity or finite-difference argument. This is not a question of genericity or of disagreement with consensus; it is a missing regularity argument in the central proof. If the missing AC/finite-difference lemma is standard and can be added, the theorem is likely correct; hence the reader's CONDITIONAL verdict is appropriate. I agree partially with the reader: the identified weak assumption (niceness) is real, but the most load-bearing concern is the unjustified integration step in Lemma 4.6.","tokens_in":25161,"tokens_out":23678,"duration_ms":236190,"concrete_test":"Check whether [22, Thm. 1.14], [25, Lem. 3.2], or the proof of Proposition 3.5 contains a finite-difference inequality of the form c_k(Y,λ_{δ+h})−c_k(Y,λδ) ≤ (F+ε)h·c_k(Y,λδ)+o(h), uniformly in δ, or an absolute-continuity statement for c_k along C^1 families e^{f_t}λ. If such a statement is present, add it to Lemma 4.6 and the integration step is valid. If not, construct a smooth one-parameter family of contact forms for which c_1 is not absolutely continuous (or prove it is always AC); this settles whether the 'integrating' step in Lemma 4.6 can be justified as written.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"Section 4.3, Lemma 4.6: after applying Lemma 3.8 to an orbit set realizing c_k(Y,λδ), the proof obtains, for almost every sufficiently small δ, the differential inequality d/dδ c_k(Y,λδ) < (F+ε)c_k(Y,λδ), and then says 'Integrating this inequality gives (4.28)'. The function δ↦c_k(Y,λδ) is shown in Proposition 3.5 only to be C^0-continuous, and monotonicity is noted in Remark 3.6; it is not shown to be absolutely continuous. For a continuous monotone function, an a.e. pointwise derivative bound does not imply the integrated bound (Cantor-type singular functions are counterexamples). The inequality (4.28) is then used in Lemma 4.8 and the proof of Theorem 1.10 to produce the required contradiction. The paper does not supply a finite-difference (one-sided Dini derivative) version of Lemma 3.8, nor a citation to a result giving absolute continuity of elementary spectral invariants along C^1 families of conformal factors. A related full-measure/differentiability subtlety appears in Lemma 4.7, where δ is slightly increased to make A0δ/V rational, which may leave the full-measure set of δ where c_k is differentiable. This is a technical but load-bearing gap in the central argument; it does not by itself disprove the theorem, but the proof as written is incomplete.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves that if (Y, λ, u) is an admissible symplectic surface in a closed contact three-manifold satisfying a 'nice' boundary hypothesis, then the supremum over simple Reeb orbits γ of (γ · Σ)/A(γ) is at least Area(Σ, dλ)/vol(Y, λ). The proof inflates λ in a slab over Σ, tracks the elementary spectral invariants c_k during the inflation, and compares them with alternative ECH capacities of a symplectization domain via Weyl laws. If the boundary ratio inequality (1.5) holds, the orbit can be chosen to meet int(Σ), giving Corollary 1.11. Examples 1.12 and 1.13 show the bound is sharp, and Section 2 derives a generalization of mean-action/Calabi-type inequalities for area-preserving surface diffeomorphisms. No genericity of the contact form is assumed.","tokens_in":25447,"tokens_out":21627,"duration_ms":231560,"significance":"If the main theorem is correct, it is a significant quantitative statement for Reeb flows without genericity, complementing Irie's equidistribution results and giving a new proof route to mean-action/Calabi inequalities. The proof is largely explicit: Lemma 4.4 gives a detailed inflation construction, Lemma 4.6 records the spectral growth estimate, and Lemma 4.7 performs the Weyl-law comparison; the examples verify sharpness. The main caveats are the heavy reliance on the author's own spectral machinery from [21,22] and the technical gap in Lemma 4.6 discussed below; these appear local and fixable rather than fatal.","major_comments":[{"comment":"The proof of Lemma 4.6 obtains, for almost every sufficiently small δ, the differential inequality d/dδ c_k(Y, λδ) < (F+ε)c_k(Y, λδ), and then says 'Integrating this inequality gives (4.28)'. However, Proposition 3.5 gives only C^0-continuity of δ ↦ c_k(Y, λδ), and Remark 3.6 gives monotonicity; neither implies absolute continuity. A continuous monotone function can have an almost-everywhere derivative bound without satisfying the integrated bound (Cantor-type singular functions are counterexamples). Inequality (4.28) is then used in Lemma 4.8 and in the proof of Theorem 1.10, so this is load-bearing. The manuscript needs an additional argument, for example proving that for nondegenerate λ and small δ the function δ ↦ c_k(Y, λδ) is piecewise given by finitely many smooth action functions of persistent orbit sets and hence is absolutely continuous, or replacing the integration by a finite-difference (Dini derivative) argument.","section":"§4.3, Lemma 4.6 and Eq. (4.28)"},{"comment":"The proof of Lemma 4.7 begins 'By increasing δ slightly if necessary, we can assume without loss of generality that A0δ/V is rational.' Lemma 4.8 later applies Lemma 4.6 at exactly this δ, and Lemma 4.6 as written requires δ to lie in the full-measure set of Lemma 4.4(d) where δ ↦ c_k(Y, λδ) is differentiable. An arbitrary full-measure set need not contain any rational numbers, so this maneuver is not justified as written. One can avoid changing δ by choosing positive integers k, l with l/k → 2A0δ/V and controlling the O(k^{1/4}) errors in the Weyl laws, or one can first establish an endpoint-independent form of Lemma 4.6 using absolute continuity; the paper should state which repair is intended.","section":"§4.4, Lemma 4.7 and Lemma 4.4(d)"}],"minor_comments":[{"comment":"The sentence 'Then by Lemma 4.6, there exists a simple Reeb orbit γ ∈ P(λ) with A(γ) ≤ L satisfying (4.48)' is logically misleading: Lemma 4.6 does not assert existence. The intended argument is a contradiction: if no such orbit exists, then F satisfies the hypothesis of Lemma 4.6, giving (4.28) with k replaced by k+l, which contradicts (4.50). Please rewrite this step accordingly.","section":"§4.4, Lemma 4.8"},{"comment":"There are small typographical errors where 'P(γ)' appears instead of 'P(λ)': in the proof of Theorem 1.10 ('simple Reeb orbit γ ∈ P(γ)') and in the proof of Lemma 4.6 ('α_i ∈ P(γ)').","section":"Throughout §4.4"},{"comment":"The statement that 'the return map along γ is an irrational rotation' should be justified from the nondegeneracy hypothesis: for a nondegenerate Reeb flow, a rational rotation number would create extra degenerate torus orbits in the neighborhood, so irrationality follows, but this is not immediate from Definition 1.9 alone.","section":"§4.2, Lemma 4.4, Step 2"}],"recommendation":"major_revision","confidential_remarks":"The central construction is sound in outline, and the gap in Lemma 4.6 appears repairable by adding an absolute-continuity argument or a finite-difference version. I would send the paper back for revision rather than reject. The paper leans heavily on the author's own prior spectral machinery, which is standard in this line of work; the new inflation construction and the Weyl-law comparison are the substantive contributions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The key novelty is real: Theorem 1.10 removes the C-infinity generic hypothesis from the recent circle of equidistribution and quantitative closing lemma results, giving a lower bound on Reeb orbit intersection frequency for every nice contact form. The inflation construction in Section 4.2 is explicit and the comparison between the elementary spectral Weyl law and the alternative capacity Weyl law in Lemma 4.7 is a neat idea. Corollary 1.11 and the mean-action/Calabi application in Theorem 2.7 are genuinely new generalizations, and the examples in Section 1.1 do a good job showing sharpness and why the extra hypothesis (1.5) is needed. No fitted parameters appear anywhere; the main inequality (1.3) is clean and plausible.\n\nThat said, there is one soft spot that matters. Lemma 4.6 derives an almost-everywhere derivative bound for the function delta -> c_k(Y, lambda_delta), which is only known to be C^0-continuous and monotone, and then says 'integrating this inequality gives (4.28)'. That inference is not justified: a continuous monotone function can have a Cantor-type singular part, so an a.e. pointwise derivative bound does not imply the integrated bound. This is load-bearing, since (4.28) feeds directly into Lemma 4.8 and the proof of Theorem 1.10. I don't think the underlying theorem is false, and the gap is likely fixable by proving a finite-difference version of Lemma 3.8 or by establishing absolute continuity of the spectral invariant along this particular deformation, but as written the proof is incomplete at this step.\n\nOther weaknesses are minor. The paper depends heavily on the author's own prior machinery ([19], [21], [22]); that is legitimate, but it means the paper is not self-contained. The 'nice' hypothesis is restrictive and excludes many natural examples, as the paper openly acknowledges in Question 1.18. The nondegenerate approximation at the end of Section 4.4 is handled a bit tersely, but the continuity argument there is standard.\n\nThe paper deserves a serious referee. The main result is important enough, the proof strategy is mostly transparent, and the single technical gap is localized and probably repairable. I would bring it to reading group and would cite it once the Lemma 4.6 issue is resolved. A peer review should focus on that integration step and on whether the monotone spectral invariant can be shown absolutely continuous along the inflation family.","headline":"A genuinely new no-genericity result in contact dynamics, with a real but likely repairable analytic gap in Lemma 4.6 that a referee should push on.","tokens_in":25954,"tokens_out":3299,"would_cite":true,"duration_ms":36066,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["53D10","53D35","53D40","37C27"],"pacs":[],"model":"deepseek-v4-flash","headline":"A symplectic surface in any nice contact 3-manifold is crossed by Reeb orbits at least at the area–volume frequency.","keywords":["Reeb orbits","contact three-manifolds","symplectic surfaces","embedded contact homology","spectral invariants","Calabi invariant","mean action","Weyl law"],"falsifier":"A concrete disproof would be a contact three-manifold with a nice admissible symplectic surface satisfying inequality (1.5) for which every simple Reeb orbit meeting $\\mathrm{int}(\\Sigma)$ has $\\gamma\\cdot\\Sigma/A(\\gamma)<\\mathrm{Area}(\\Sigma,d\\lambda)/\\mathrm{vol}(Y,\\lambda)$; one could search for this numerically in a one-parameter family of tight contact forms on a lens space or prequantization bundle by computing periods and intersection numbers of low-action Reeb orbits.","tokens_in":24955,"feed_emoji":"🔄","tokens_out":16355,"duration_ms":134715,"temperature":0.7,"pith_summary":"This paper proves a quantitative recurrence statement for Reeb flows on three-dimensional contact manifolds: given a symplectic surface whose boundary consists of Reeb orbits and whose boundary behavior is 'nice' (a perfectly linear rotation), there must exist Reeb orbits crossing the surface almost as often as the ratio of the surface's symplectic area to the manifold's contact volume. More precisely, for every $\\varepsilon>0$, some simple Reeb orbit $\\gamma$ satisfies $\\gamma\\cdot\\Sigma/A(\\gamma) \\ge \\mathrm{Area}(\\Sigma)/\\mathrm{vol}(Y) - \\varepsilon$, where $A(\\gamma)$ is the orbit's period. If in addition every boundary orbit has intersection frequency strictly below that ratio, the orbit can be chosen to meet the interior of the surface. The theorem needs no genericity of the contact form. A corollary generalizes known inequalities relating the mean action of periodic orbits of area-preserving surface diffeomorphisms to their Calabi invariant.","feed_headline":"Reeb orbits hit every symplectic surface at least area/volume often","feed_subtitle":"No genericity needed: some Reeb orbit crosses the surface within epsilon of the area-over-volume frequency.","key_machinery":"The argument is carried by three objects: the elementary spectral invariants $c_k(Y,\\lambda)$ (spectral numbers built from holomorphic curves in the symplectization, satisfying a Weyl law $c_k^2/k\\to 2\\,\\mathrm{vol}(Y,\\lambda)$); the alternative ECH capacities $c^{\\mathrm{Alt}}_l$ of four-dimensional symplectic domains (min–max symplectic capacities with their own Weyl law); and the inflation construction, which replaces $\\lambda$ by $\\lambda_\\delta=e^{\\delta\\zeta(s)\\beta_\\Sigma(z)}\\lambda$ in a thin slab $[0,s_0]\\times\\Sigma$. The inflation makes the volume grow by $2\\delta\\,\\mathrm{Area}(\\Sigma)$ while shifting the action of an interior-crossing orbit by $\\delta(\\gamma\\cdot\\Sigma)$; the capacity bound $c_{k+l}(Y,\\lambda_\\delta)\\ge c_k(Y,\\lambda)+c^{\\mathrm{Alt}}_l(M_{\\Sigma_0,\\delta})$ then combines the two Weyl laws to force an orbit with frequency at least area/volume. The 'nice' hypothesis is what makes the boundary Reeb orbits behave like linear rotations, so the action shift and the counting near the boundary are under control.","core_discovery":"On the paper's own terms, the central discovery is Theorem 1.10: if $(Y,\\lambda,u)$ is nice in the sense of Definition 1.9, then $\\sup_{\\gamma\\in P(\\lambda)} (\\gamma\\cdot\\Sigma)/A(\\gamma) \\ge \\mathrm{Area}(\\Sigma,d\\lambda)/\\mathrm{vol}(Y,\\lambda)$. Because the supremum may be attained only on boundary orbits, Corollary 1.11 adds inequality (1.5), under which the same lower bound holds for the supremum restricted to orbits that meet $\\mathrm{int}(\\Sigma)$. The proof proceeds by inflating the contact form near $\\Sigma$ and studying how the elementary spectral invariants $c_k(Y,\\lambda)$ grow; if all Reeb orbits had frequency below area/volume, the growth forced by the Weyl laws would be contradicted. An application reformulates this as a statement about mean action and Calabi invariants for area-preserving surface diffeomorphisms.","pith_inferences":["An extension the author leaves open: if the niceness hypothesis can be removed, the same inflation-and-Weyl-law strategy would likely give frequent intersections with Birkhoff sections and with geodesics in reversible Finsler metrics; the paper states this as an open question, so this consequence is speculative.","The action bound in the proof suggests a quantitative return time: one should be able to say how large an action is needed before an $\\varepsilon$-frequent orbit appears; the optimal dependence on $\\varepsilon$ is identified in the paper as open.","The equality cases in the ellipsoid indicate that inequality (1.5) is not merely technical: when boundary orbits sit exactly at the area/volume ratio, interior orbits need not exist at all. A natural test is to perturb such ellipsoid boundary rotations slightly and check whether interior-intersecting orbits appear with frequency near the ratio.","The same mechanism may yield the reverse inequality stated as an open question by running the inflation with the opposite sign (deflating near $\\Sigma$); if the interval property also holds, the set of crossing frequencies would accumulate at area/volume from both sides."],"forward_implications":["For every $\\varepsilon>0$ there is a simple Reeb orbit whose crossing frequency $\\gamma\\cdot\\Sigma/A(\\gamma)$ is within $\\varepsilon$ of $\\mathrm{Area}(\\Sigma)/\\mathrm{vol}(Y)$, with no genericity assumption on the contact form.","If every boundary Reeb orbit has frequency strictly below that ratio (inequality (1.5)), the orbit can be chosen to meet the interior of $\\Sigma$, and Lemma 4.8 gives an explicit upper bound on its action in terms of the spectral invariant used in the proof.","For area-preserving surface diffeomorphisms satisfying zero flux and rigid-rotation boundary conditions, the mean action of some periodic orbit is bounded above by the Calabi invariant when the Calabi invariant is below the boundary value of the primitive; a dual corollary gives the reverse bound.","The frequency bound is sharp: in the irrational ellipsoid examples every orbit has frequency exactly area/volume, and the interior version of the theorem fails without inequality (1.5).","If the Reeb flow is not dense over the surface, the denominator can be replaced by the smaller volume of the closure of $\\Phi(\\mathbb{R}\\times\\Sigma)$, giving a stronger bound."],"supporting_citations":[{"why":"Supplies the elementary spectral invariants and their Weyl law, spectrality, continuity, and capacity bound, which are the main tools of the proof.","marker":"[22]"},{"why":"Defines the alternative ECH capacities and proves their monotonicity and Weyl law, used to estimate the volume of the inflated slab.","marker":"[21]"},{"why":"Provides the construction turning an area-preserving surface diffeomorphism into a contact three-manifold with an admissible surface, used in the application, and the boundary intersection-number convention extended in Definition 1.8.","marker":"[19]"},{"why":"Gives the generic equidistribution theorem for Reeb orbits that the main theorem extends and that supplies the reverse inequality in the generic case.","marker":"[25]"},{"why":"Supplies the Weyl-law asymptotic for embedded contact homology spectral invariants that underlies the two Weyl laws used in the proof.","marker":"[8]"},{"why":"Defines the embedded contact homology spectral invariants from which the elementary invariants are derived and to which they are compared for subleading estimates.","marker":"[17]"}],"fun_headline_variants":["Reeb orbits cross symplectic surfaces at frequency ≥ area/volume","No genericity: Reeb orbits intersect symplectic surfaces at least area/volume","Reeb orbits hit symplectic surfaces at least area/volume frequency","Reeb orbits pierce symplectic surfaces at least area/volume rate","Every symplectic surface has a Reeb orbit crossing at area/volume"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof collapses if the Reeb vector field is not exactly a linear rotation near each boundary Reeb orbit: the 'nice' hypothesis (Definition 1.9) is what makes the inflation construction and the action-shift estimates hold.","fun_headline_variants_meta":{"raw":{"variants":["Reeb orbits cross symplectic surfaces at frequency ≥ area/volume","No genericity: Reeb orbits intersect symplectic surfaces at least area/volume","Reeb orbits hit symplectic surfaces at least area/volume frequency","Reeb orbits pierce symplectic surfaces at least area/volume rate","Every symplectic surface has a Reeb orbit crossing at area/volume"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00197,"raw_usage":{"total_tokens":7667,"prompt_tokens":886,"completion_tokens":6781,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":502,"completion_tokens_details":{"reasoning_tokens":6680}},"tokens_in":502,"tokens_out":6781,"duration_ms":46278,"temperature":1.0,"reasoning_tokens":6680,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T05:54:45.448325+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A concrete disproof would be a contact three-manifold with a nice admissible symplectic surface satisfying inequality (1.5) for which every simple Reeb orbit meeting $\\mathrm{int}(\\Sigma)$ has $\\gamma\\cdot\\Sigma/A(\\gamma)<\\mathrm{Area}(\\Sigma,d\\lambda)/\\mathrm{vol}(Y,\\lambda)$; one could search for this numerically in a one-parameter family of tight contact forms on a lens space or prequantization bundle by computing periods and intersection numbers of low-action Reeb orbits.","supporting_citations":[{"cited_title":"Hutchings, Elementary spectral invariants and quantitative closing l emmas for contact three-manifolds , J","cited_arxiv_id":null,"evidence_quote":"Supplies the elementary spectral invariants and their Weyl law, spectrality, continuity, and capacity bound, which are the main tools of the proof."},{"cited_title":"Hutchings, An elementary alternative to ECH capacities , PNAS Vol","cited_arxiv_id":null,"evidence_quote":"Defines the alternative ECH capacities and proves their monotonicity and Weyl law, used to estimate the volume of the inflated slab."},{"cited_title":"Hutchings, Mean action and the Calabi invariant , J","cited_arxiv_id":null,"evidence_quote":"Provides the construction turning an area-preserving surface diffeomorphism into a contact three-manifold with an admissible surface, used in the application, and the boundary intersection-number convention extended in Definition 1.8."},{"cited_title":"Irie, Equidistributed periodic orbits of C ∞-generic three-dimensional Reeb flows , J","cited_arxiv_id":null,"evidence_quote":"Gives the generic equidistribution theorem for Reeb orbits that the main theorem extends and that supplies the reverse inequality in the generic case."},{"cited_title":"Cristofaro-Gardiner, M","cited_arxiv_id":null,"evidence_quote":"Supplies the Weyl-law asymptotic for embedded contact homology spectral invariants that underlies the two Weyl laws used in the proof."},{"cited_title":"Hutchings, Quantitative embedded contact homology , J","cited_arxiv_id":null,"evidence_quote":"Defines the embedded contact homology spectral invariants from which the elementary invariants are derived and to which they are compared for subleading estimates."}],"review_version":1}