{"id":"16380718-baac-4045-afa1-eacc593ca072","arxiv_id":"2506.12438","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The genus 1 divisor Gromov-Witten invariant of Hilb^n(C^2) is expressed through traces of quantum multiplication and equals the Eisenstein generating function that also governs Noether-Lefschetz cycles on A_g.","lead":"This paper computes a central genus 1 Gromov-Witten invariant of the Hilbert scheme of n points in the plane for every n. The answer matches a Noether-Lefschetz calculation on moduli spaces of abelian varieties and proves a homomorphism property for a tautological projection operator.","discovery_kind":"unification","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central reduction depends on the [46] correspondence, whose proof the paper itself flags as gapped; the analytic-continuation patch is not part of the manuscript.","rationale":"The reader identifies Proposition 7 as the weakest assumption. On inspection, Proposition 7 is supported by a short self-contained argument: the exact sequence (2.5) and the factorization (2.6) of lambda_g, together with the expansion (2.2) of the Euler class, give the vanishing directly. The remaining assertion of injectivity for n>0 is plausible and standard. By contrast, the paper itself flags a missing proof in the published source [46] of the correspondence that converts the Hilbert scheme invariant into the families invariant. That correspondence is the bridge between Theorems 1 and 2, so a failure there would invalidate the central formula even if every later argument is correct. The manuscript's statement that the gap does not affect the analytic continuation form of the result is not a proof, and the fix is not included. This is a genuine load-bearing concern about the central claim. The appropriate verdict remains CONDITIONAL, as the reader already concluded, so no verdict change is recommended. The concrete test is designed to settle whether the concern lands: either the revised [46] supplies the missing proof, or an independent low-genus computation reproduces the formula's predictions.","tokens_in":33682,"tokens_out":35221,"duration_ms":409160,"concrete_test":"Obtain the revised version of [46] from the authors' websites and verify that the proof of the triangle of correspondences used here is complete, in particular that the equality of the Hilbert-scheme and family invariants as rational functions after the variable change -q=e^{iu} is proven without assuming the unproven rationality statement. In parallel, check the first nontrivial case n=3: compute the genus 1 divisor invariant of Hilb^3(C^2) directly by localization on the moduli space of stable maps (or by the Hilb^3 methods of Hu-Qin) and compare the first three q-coefficients with the closed form in Theorem 1; a match for those coefficients would provide independent evidence that the bridge is valid, while any mismatch would falsify the reduction.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The proof of Theorem 1 reduces the Hilbert scheme series to the families invariant via Theorem A, quoted from [46]. The manuscript explicitly states that the published proof of the rationality of the genus 1 series in [46] has a gap, and says the gap is filled only in a revision available on the authors' websites, adding that it 'does not affect the proof in [46] of the triangle of correspondences in the form of analytic continuation from q=0 to q=-1'. This is an external, unverified patch. The version of Theorem A used here is an equality of rational functions in q; if rationality is not established, analytic continuation from an interval cannot be used to conclude equality of rational functions on their full domain. The bridge (0.3) is then not rigorously available. This is more load-bearing than Proposition 7: the vanishing in Proposition 7 follows from the exact sequence (2.5) and factorization (2.6), and the Euler-class algebra in (2.2) is self-contained. If the [46] correspondence fails, Theorem 1 is unsupported; if it holds, the rest of the paper is a substantial derivation. Since the patch is not included, the correctness of the central claim rests on an unpublished document.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper computes genus-1 T-equivariant Gromov-Witten invariants of the Hilbert scheme Hilb^n(C^2). Theorem 1 gives an explicit formula for the divisor 1-point invariant <D>_1^{Hilb^n(C^2)} as a universal coefficient -(t1+t2)^2/(24 t1 t2) times a combination of normalized traces Tr_k of quantum multiplication by D, with coefficients sigma_{-1}(n-k). The proof passes through the Pandharipande-Tseng correspondence to the Gromov-Witten theory of families of local elliptic curves over M_{1,1}. In that setting the paper proves Theorem 2, a closed formula for the families Hodge integral <tau_1(p_1) lambda_g lambda_{g-2}>_{g,n}^{pi,circ} whose generating series is (-1)^g/24 |B_{2g}|/(4g) |B_{2g-2}|/(2g-2)! E_{2g}(Q). From this the authors derive, via prior work [23], the homomorphism property for the tautological projection on CH^*(A_g) (Theorem 3). They also prove a reduction of multi-point genus-1 invariants to 1-point series (Theorem 4) and, conditional on a Wronskian nondegeneracy conjecture, a full reconstruction of the genus-1 theory (Theorems 5 and 6). Explicit tables for n up to 5 are included.","tokens_in":33924,"tokens_out":14947,"duration_ms":172888,"significance":"If the arguments are correct, Theorem 1 is the first complete closed-form evaluation of the fundamental genus-1 divisor invariant of Hilb^n(C^2) for all n, and Theorem 2 establishes an elegant Eisenstein-series structure for families Hodge integrals over M_{1,1}. The resulting connection to Noether-Lefschetz cycles on A_g is striking and consequential. The paper is also commendably explicit about its main caveats: Theorems 5 and 6 are conditional on Conjecture 13, verified only for n <= 7, and the key bridge Theorem A rests on a rationality statement whose published proof is acknowledged to have a gap filled only in an external revision. The computations are substantial and the exposition is generally clear, but the two caveats mean that the unconditional central claims are not fully self-contained as submitted.","major_comments":[{"comment":"The proof of Theorem 1 uses Theorem A to identify the full Hilbert scheme series <D>_1^{Hilb^n(C^2)} with the family series via (0.3). The footnote states that the published proof of rationality of the genus-1 series in [46] has a gap, filled only in an unpublished revision, and asserts that analytic continuation from q=0 to q=-1 is all that is needed. As written, this does not justify Theorem A as an equality of rational functions in q: if the two sides are only known as analytic germs near q=0 or along a continuation path, one cannot conclude equality of rational functions on their full domain without a rationality proof for both sides. Since the coefficient extraction in Section 3.3 uses the full series identity, this is load-bearing for Theorem 1. Please include the missing rationality argument or state precisely the weaker analytic-continuation statement that is actually used and prove that it suffices for (0.3).","section":"0.3, footnote 7 and Theorem A"},{"comment":"The exact sequence 0 -> eps_S^*(Tan P^1) -> E_g -> F -> 0 is the key input for the factorization (2.6), hence for Proposition 7 and for the reduction in Section 2.4 from the K3-family invariant to the fixed elliptic curve invariant. The injectivity of the pull-back map (2.4) is asserted with the single phrase \"Since n > 0\", and the equality c_{g-1}(F) = lambda_{g-1} is justified only by restricting the sequence to a fiber of eps_S. These identifications are not obvious globally on the moduli space of relative stable maps, especially over the expanded target with simple circuits over the 24 nodal fibers. Please provide a global argument or a precise reference for both statements, since Proposition 7 is the mechanism that makes Theorem 2 follow from the fixed-target evaluation (2.3).","section":"2.5, Eq. (2.5)"},{"comment":"The proof of Proposition 8, which is used in the connected/disconnected calculus leading to Theorem 1, relies on the assertion lambda_g lambda_{g-2} = 2 lambda_{g-1}^2 and on the claim that this insertion annihilates all degeneration graphs except the two configurations in Figure 1. The identity is not proved or cited, and the dimension analysis of the vertex list (3.15)--(3.22) is summarized rather than demonstrated. These steps control the right-hand side of the degeneration formula (3.13), so they are load-bearing for Theorem 1. Please provide a proof or reference for the Hodge-class identity and a more detailed verification of the vanishing and non-vanishing of the listed vertex contributions.","section":"3.5, proof of Proposition 8"}],"minor_comments":[{"comment":"The displayed formula for Tr_n is typeset without sufficient parentheses: the rational factors ((q)^{mu_i}+1)/((q)^{mu_i}-1) and ((q)+1)/((q)-1) are hard to distinguish from the surrounding terms. Please add parentheses or define the summand explicitly.","section":"0.5.1, Eq. (0.5)"},{"comment":"The equality of the integral over M_{1,1} with 1/48 times the integral over the P^1-base of the K3 fibration is stated without comment. Please justify compatibility of the relative virtual classes under this degree-48 base change or cite a standard statement.","section":"2.4"},{"comment":"The text defines triples (m,k,g) but a few lines later refers to \"triplets (e,k,g)\"; please correct this inconsistent notation.","section":"3.4"},{"comment":"In the n=0 evaluation, the Hodge integral used from [10, Theorem 4] is quoted without stating the exact formula; a short display of the quoted integral would make the numerical coefficient easier to verify.","section":"2.2"}],"recommendation":"major_revision","confidential_remarks":"The main risk is the dependence on the external rationality patch for [46]; if the authors include that argument as an appendix or state and prove the precise weaker correspondence needed, the paper would be considerably more self-contained. The Wronskian conjecture is clearly flagged, but it limits the unconditional scope of Theorems 5 and 6. I would not reject on the current evidence: the central formulas are explicit, checkable in examples, and the presented proofs are coherent apart from the flagged dependence."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The headline result is genuine and substantial. Theorem 1 gives an explicit formula for <D>_1 on Hilb^n(C^2) for all n in terms of traces of quantum multiplication, which was open for n >= 3. Theorem 2 evaluates a family Hodge integral as an Eisenstein series, and the triple equivalence with Noether-Lefschetz classes on A_g is a real structural insight. The authors are also honest about what is conditional: the full genus 1 reconstruction (Theorems 5-6) depends on a Wronskian nondegeneracy conjecture, verified for n <= 7. That is a clearly flagged conjecture, not a hidden assumption.\n\nThe main weakness is external to the manuscript. The paper quotes Theorem A from the published [46] to convert Hilbert scheme invariants into family invariants, and then notes in a footnote that the proof of rationality in [46] has a gap, filled only in a revision on the authors' websites. This matters because Theorem A is used as an equality of rational functions in q, and the analytic continuation from q=0 to q=-1 may not be enough to conclude equality on the full domain without rationality. The patch is not included here, so the central bridge in Theorem 1 rests on an unpublished document. That is an unusual and load-bearing reliance. The rest of the argument, including Proposition 7 and the vanishing that replaces the family invariant with a fixed elliptic curve invariant, is well-structured and self-contained.\n\nI do not think the paper should be rejected or distrusted wholesale. The authors are credible, the computations are detailed, and the claimed match with Noether-Lefschetz theory is the kind of result that deserves public scrutiny. But before accepting, a referee needs either the repaired proof of the rationality statement in [46] or a precise statement of which weaker form of Theorem A is actually needed and why it is unaffected by the gap. The paper would be improved by appending that patch or spelling out the argument in full.\n\nWho is this for? Anyone working on Gromov-Witten theory of Hilbert schemes, family GW theory, or tautological rings of A_g. It deserves a serious referee and a close reading, but the referee should insist on clarifying the status of Theorem A before the paper is final.","headline":"The genus 1 divisor invariant of Hilb^n(C^2) is computed in closed form for all n, with a striking match to Noether-Lefschetz theory; the main caveat is that one load-bearing bridge rests on a patch external to the manuscript.","tokens_in":34465,"tokens_out":1407,"would_cite":true,"duration_ms":18977,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["14N35","14C05","14K10"],"pacs":[],"model":"deepseek-v4-flash","headline":"Genus-1 Gromov-Witten invariants of the Hilbert scheme of points in the plane are determined by traces of quantum multiplication and match the Noether-Lefschetz theory of abelian varieties.","keywords":["Hilbert scheme","Gromov-Witten theory","Noether-Lefschetz theory","moduli of abelian varieties","Eisenstein series","quantum cohomology","Hodge integrals","tautological projection"],"falsifier":"A concrete check would be to compute the Proposition 7 integral by virtual localization for a specific elliptically fibered K3 (for instance, a Weierstrass fibration) at $g=2$ and $n=1$; a nonzero value would falsify the vanishing and hence Theorem 2. Independently, the Wronskian determinant $\\det(W)$ used in Theorems 5 and 6 can be evaluated symbolically for $n=8$; a zero would disprove the nondegeneracy conjecture on which the full reconstruction depends.","tokens_in":33484,"feed_emoji":"∑","tokens_out":9518,"duration_ms":97119,"temperature":0.7,"pith_summary":"The paper calculates the genus 1 Gromov-Witten theory of Hilb^n($C^{2}$), the space of (possibly coincident) n-point configurations in the plane. Its central result is a closed formula for the basic divisor insertion, expressed as a constant times a sum of traces of the operator of quantum multiplication by that divisor, with coefficients given by the divisor-sum function. The same generating function appears in the Noether-Lefschetz geometry of the moduli space of principally polarized abelian varieties, where it describes the tautological projection of special cycle classes. The proof goes through a families version of local Gromov-Witten theory over the moduli space of elliptic curves, where the answer is shown to be an Eisenstein series. It is a complete evaluation of a higher-genus invariant series, and it links curve counting in Hilbert schemes to modular forms and abelian varieties.","feed_headline":"Hilb^n(C^2) genus-1 counts equal Eisenstein series in disguise","feed_subtitle":"One divisor insertion controls genus-1 curve counts and links them to abelian varieties.","key_machinery":"The load-bearing object is the moduli stack of connected stable maps to the fibers of the universal elliptic curve $\\pi\\colon E \\to \\overline{\\mathcal{M}}_{1,1}$, together with the Hodge bundle on the domain. The argument is carried by three mechanisms: Theorem A of the paper's earlier correspondence, converting genus 1 invariants of Hilb^n($C^{2}$) into families invariants of local elliptic curves; a vanishing statement for maps to the fibers of an elliptically fibered K3 surface, which yields a factorization of the top Hodge class and lets the families Hodge integral be replaced by a fixed elliptic curve invariant; and a connected/disconnected calculus that relates the family invariant to traces of quantum multiplication via the Fock space expression for the quantum multiplication operator. For the full reconstruction, Givental's genus 1 formula for semisimple Frobenius manifolds is used, controlled by a Wronskian determinant whose nondegeneracy is conjectured.","core_discovery":"On the paper's own terms, the discovery is that three objects are governed by the same number: the genus 1 Gromov-Witten invariant of Hilb^n($C^{2}$) with a divisor insertion; a families Hodge integral over moduli of stable maps to the universal elliptic curve; and the tautological projection of Noether-Lefschetz cycle classes on the moduli space of principally polarized abelian varieties. Theorem 1 states $\\langle D \\rangle_1^{\\mathrm{Hilb}^n(\\mathbb{C}^2)} = -\\frac{1}{24}\\frac{(t_1+t_2)^2}{t_1 t_2}\\bigl( \\mathrm{Tr}_n + \\sum_{k=2}^{n-1} \\sigma_{-1}(n-k) \\mathrm{Tr}_k \\bigr)$, where $\\mathrm{Tr}_k$ is the normalized trace of quantum multiplication by the divisor $D$. Theorem 2 states that the generating series of the families invariants is $\\frac{(-1)^g}{24}\\frac{|B_{2g}|}{4g}\\frac{|B_{2g-2}|}{(2g-2)!} E_{2g}(Q)$, with $E_{2g}$ the Eisenstein series of weight $2g$. The two theorems are equivalent, and Theorem 3 derives the homomorphism property of the tautological projection for the Torelli cycle and the Noether-Lefschetz class.","pith_inferences":["One could expect that the Wronskian nondegeneracy conjecture is a manifestation of the irreducibility of the quantum connection of Hilb^n(C^2), similar in spirit to irreducibility results for quantum $D$-modules; this is not addressed in the paper.","The appearance of $E_{2g}(Q)$ suggests that the full higher-genus theory of Hilb^n(C^2) might be expressible in terms of quasimodular forms, as happens for elliptic curves; the paper only proves the genus 1 case.","A natural testable extension is to replace $\\mathbb{C}^2$ by the minimal resolution of $A_{n-1}$ singularities, where the quantum cohomology is known, and ask whether the same trace-plus-divisor-sum formula holds with coefficients replaced by a root-system dependent function.","If the Wronskian conjecture is true, the reconstruction theorem would give a practical algorithm to compute all genus 1 invariants of Hilb^n(C^2) to arbitrary $n$ from the genus 0 data alone."],"forward_implications":["The genus 1 Gromov-Witten potential of Hilb^n(C^2) is determined, up to genus 0 data and the Wronskian conjecture, by the single series $\\langle D \\rangle_1^{\\mathrm{Hilb}^n(\\mathbb{C}^2)}$ together with the quantum multiplication operator $M_D$.","The classes $\\mathrm{Tor}_*[\\mathcal{M}^{\\mathrm{ct}}_g]$ and $[\\mathrm{NL}_{g,n}]$ satisfy the multiplicative property with respect to the tautological projection: $\\mathrm{taut}(\\gamma)\\cdot\\mathrm{taut}(\\gamma') = \\mathrm{taut}(\\gamma\\cdot\\gamma')$ for these pairs.","The generating function of the families Hodge integrals equals an Eisenstein series, giving closed form evaluations in terms of Bernoulli numbers and divisor sums.","For $n \\le 5$, explicit formulas for all 1-point genus 1 series are listed, providing concrete data for the Hilbert scheme theory.","The degree 0 part of the formula matches the Carlsson-Okounkov vertex operator calculations."],"supporting_citations":[{"why":"Supplies the explicit matrix of quantum multiplication by $D$ in the Nakajima basis, used to define the traces $\\mathrm{Tr}_n$.","marker":"[39]"},{"why":"Provides Theorem A, the correspondence between genus 1 invariants of Hilb^n(C^2) and families local elliptic curve invariants that starts the proof.","marker":"[46]"},{"why":"Gives the Hodge integral evaluation on a fixed elliptic curve target used to prove Theorem 2.","marker":"[36]"},{"why":"Proves Theorem B describing $\\mathrm{taut}([\\mathrm{NL}_{g,n}])$ as a multiple of $\\lambda_{g-1}$, the input for the homomorphism statement.","marker":"[23]"},{"why":"Constructs the projection operator $\\mathrm{taut}$ whose homomorphism property is the subject of Theorem 3.","marker":"[5]"},{"why":"Provides the Hodge integral evaluation used in the $n=0$ case of Theorem 2.","marker":"[10]"}],"fun_headline_variants":["One divisor insertion ties Hilb^n(C^2) counts to Eisenstein series","Same pattern in Hilb^n(C^2) and A_g: Eisenstein series","Hilb^n(C^2) genus-1 invariant is an Eisenstein series","From Hilbert scheme points to Eisenstein series: one invariant"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is Proposition 7: for an elliptically fibered K3 surface, the integral against the Euler class $e\\bigl(E_g^\\vee \\otimes \\varepsilon_S^*(\\mathrm{Tan}\\,\\mathbb{P}^1)\\bigr)$ over the moduli space of stable maps to the fibers vanishes for genus at least 2 and positive degree $n$; if this vanishing fails, the Eisenstein formula for the families Hodge integral no longer follows.","fun_headline_variants_meta":{"raw":{"variants":["One divisor insertion ties Hilb^n(C^2) counts to Eisenstein series","Same pattern in Hilb^n(C^2) and A_g: Eisenstein series","Hilb^n(C^2) genus-1 invariant is an Eisenstein series","From Hilbert scheme points to Eisenstein series: one invariant"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001012,"raw_usage":{"total_tokens":4322,"prompt_tokens":1038,"completion_tokens":3284,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":654,"completion_tokens_details":{"reasoning_tokens":3200}},"tokens_in":654,"tokens_out":3284,"duration_ms":30134,"temperature":1.0,"reasoning_tokens":3200,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T00:50:05.402341+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A concrete check would be to compute the Proposition 7 integral by virtual localization for a specific elliptically fibered K3 (for instance, a Weierstrass fibration) at $g=2$ and $n=1$; a nonzero value would falsify the vanishing and hence Theorem 2. Independently, the Wronskian determinant $\\det(W)$ used in Theorems 5 and 6 can be evaluated symbolically for $n=8$; a zero would disprove the nondegeneracy conjecture on which the full reconstruction depends.","supporting_citations":[{"cited_title":"Okounkov, R","cited_arxiv_id":null,"evidence_quote":"Supplies the explicit matrix of quantum multiplication by $D$ in the Nakajima basis, used to define the traces $\\mathrm{Tr}_n$."},{"cited_title":"Pandharipande, H.-H","cited_arxiv_id":null,"evidence_quote":"Provides Theorem A, the correspondence between genus 1 invariants of Hilb^n(C^2) and families local elliptic curve invariants that starts the proof."},{"cited_title":"Iribar-L ´opez, Noether-Lefschetz cycles on the moduli space of abelian varieties (2024), arXiv:2411.09910","cited_arxiv_id":null,"evidence_quote":"Proves Theorem B describing $\\mathrm{taut}([\\mathrm{NL}_{g,n}])$ as a multiple of $\\lambda_{g-1}$, the input for the homomorphism statement."},{"cited_title":"Tautological projection for cycles on the moduli space of abelian varieties","cited_arxiv_id":"2401.15768","evidence_quote":"Constructs the projection operator $\\mathrm{taut}$ whose homomorphism property is the subject of Theorem 3."},{"cited_title":"Faber, R","cited_arxiv_id":null,"evidence_quote":"Provides the Hodge integral evaluation used in the $n=0$ case of Theorem 2."}],"review_version":1}