Pith. sign in

REVIEW 2 major objections 3 minor 1 cited by

Admissible covers and stable maps

T0 review · 2 major / 3 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper proves a cycle-valued wall-crossing formula expressing every Gromov–Witten cycle as a sum of Hurwitz cycles with explicit I-function corrections, refining the ELSV formula and the Gromov–Witten/Hurwitz correspondence.

desk verdict Cycle-level GW/Hurwitz formula that looks true and is computationally supported, but the printed proof has a genuine circular dependency that must be fixed before publication. read the letter →

arxiv 2505.03487 v2 pith:AAQKJ4RG submitted 2025-05-06 math.AG

Denis Nesterov , Maximilian Schimpf , Johannes Schmitt This is my paper · ORCID
classification math.AG MSC 14N3514H1014N10
keywords admissiblecoversstablemapsHurwitzcyclesGromov-Wittenwall-crossingELSVformulaI-functionsmoduliofcurves
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Two compactifications of the space of degree-$d$ maps to a moving curve — stable maps and admissible covers — both yield cycles on the moduli space $\mathcal{M}_{g,n}$. The paper proves a cycle-valued wall-crossing formula: every Gromov–Witten cycle $GW_{g,\mu}(\gamma;\alpha)$ is equal to a finite sum, over star-shaped graphs, of Hurwitz cycles $H_{\mu,\eta}(\gamma_\Gamma;\alpha)$ glued to I-functions $I_{g_i,n_i,\eta_i}(-\tilde{\psi}_i)$ that encode contracted components and ramifications. The I-functions admit explicit closed forms in $\lambda$- and $\psi$-classes obtained by equivariant localization on $\mathbb{P}^1$. Because the equality holds for cycles rather than only for numbers, the paper recovers the ELSV formula and the Gromov–Witten/Hurwitz correspondence as refinements and extends them to arbitrary descendent insertions and moving targets. Low-genus computations for degree one and two maps to a relative $\mathbb{P}^1$ confirm the formula on tautological cohomology.

What carries the argument

The proof is driven by the master space for $\epsilon$-unramification, a stability parameter that interpolates from stable maps ($\epsilon\ll1$) to admissible covers ($\epsilon>1$); for each wall $\epsilon_0=1/d_0$ there is a proper master space with a $\mathbb{C}^*$-action, and its fixed loci are indexed by star-shaped graphs. Virtual localization on this master space, together with the entangled-tail residue identity, converts the wall-crossing into the graph sum. The I-function is the localized virtual class of the space of contracted components, \[ I_{g,n,\eta}(z)=z\prod_{j=1}^{\ell(\eta)}\eta_j\,\pi_*\left(\frac{[V_{g,n,\eta}^{\mathbb{C}^*}]}{e_{\mathbb{C}^*}($N^{{\mathrm{vir}}$})}\right), \] expanded in the range $|z|>1$; Proposition 2.6 evaluates it in terms of the Hodge bundle $\Lambda^\vee(z)$ and $\psi$-classes. The star-shaped graph carries the Hurwitz cycle on its root vertex and I-functions on its non-root vertices, and the gluing map $\mathrm{gl}_\Gamma$ assembles these factors by identifying labelled markings; dividing by $|\operatorname{Aut}(\Gamma)|$ removes overcounting.

What would settle it

Verify both sides of Theorem 2.7 as explicit classes in the tautological ring for degree $d=2$ maps to the relative $\mathbb{P}^1$ with profile $((2),(2))$ at source genus $g=4$, a case beyond the paper's table; any nonzero difference in a graph coefficient would settle that the cycle-valued equality is false.

Watch

Extended reading notes

Core claim

The central statement, Theorem 2.7, is that for $(h,m)\neq(0,1)$, \[ GW_{g,\mu}(\gamma;\$\alpha$)=\sum_{\Gamma}\frac{1}{|\operatorname{Aut}(\Gamma)|}(\mathrm{gl}_\Gamma)_*\left(H_{\mu,\eta}(\gamma_\Gamma;\$\alpha$)\boxtimes\prod_{i=1}^{k} I_{g_i,n_i,\eta_i}(-\tilde{\psi}_i)\right), \] where the sum is over star-shaped graphs whose edge labels are partitions $\eta_i\vdash d$, the genus is fixed by $g=\sum_i(g_i+\ell(\eta_i))+g_0-k$, and the degree relation reads $2g_0-2=d(2h-2)+\sum_i(d-\ell(\mu_i))+\sum_i(d-\ell(\eta_i))$; negative powers of $\psi$-classes are set to zero. The left-hand side is the pushforward of the virtual fundamental class of degree-$d$ stable maps to the universal curve over $\mathcal{M}_{h,m}$, with ramification profiles $\mu$ at target markings and cohomology insertions $\gamma$ and $\alpha$. The right-hand side is a graph sum of Hurwitz cycles — admissible covers with profiles $\mu$ and $\eta$ — with an I-function correction attached at every contracted component. The special case $(h,m)=(0,1)$ is governed by a separate polar wall-crossing formula, Theorem 6.1, which recovers the ELSV formula by extracting the coefficient of the minimal power of $z$.

Load-bearing premise

The proof relies on a residue identity on the master space whose derivation invokes the $(0,1)$ wall-crossing, while that $(0,1)$ wall-crossing is proved by applying the same formula; unless this circular dependency is broken by an independent argument, the main theorem is not established.

Editorial extensions

If this is right

  • The ELSV formula and the Gromov–Witten/Hurwitz correspondence hold as equalities of cycle classes, so they apply to arbitrary descendent insertions and to a moving target curve, not only to point insertions on a fixed curve.
  • The degree-one specialization gives an effective way to compute the Torelli pullback of $A_1\times A_{g-1}$ over $\mathcal{M}_g$, reproducing the known compact-type values $24\lambda_2$ in genus three and $20\lambda_3$ in genus four.
  • The degree-two specialization relates double-ramification cycles $DR_g(2,-2)$ to hyperelliptic cycles $Hyp_g$, giving a recursion that computes high-genus hyperelliptic cycles from lower-genus ones and from I-functions.
  • The numerical invariants in the formula — Hurwitz numbers, Fulton–MacPherson integrals, and Hodge integrals — are all explicitly computable, so the theorem yields concrete enumerative numbers for arbitrary insertions.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Beyond the paper, the same master-space residue calculus is likely reusable for other stability changes, producing cycle-level expansions for loci such as $k$-fold ramification cycles or spin Hurwitz loci whenever analogous polar I-functions exist.
  • Beyond the paper, reading the I-functions as universal classes of contracted components suggests an operator-valued graph sum in which the correspondence becomes a change of basis on the infinite wedge; the paper establishes the numerical version of this statement.
  • Beyond the paper, the recursion should produce graph-sum formulas for the hyperelliptic loci $Hyp_{g,2,0}$ for every genus, a case the paper leaves open; the resulting genus-two formula can be checked against existing computations.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 3 minor

Summary. Fix a degree d, a moving target curve of genus h, and ramification profiles assigned to m target markings. The paper claims a cycle-valued wall-crossing formula (Theorem 2.7): the Gromov–Witten cycle of degree-d stable maps, pushed forward to the moduli space of possibly disconnected source curves, equals a sum over star-shaped graphs of gluings of a Hurwitz cycle (at the root vertex) with products of explicit I-functions (at contracted vertices), the I-functions being tautological classes in lambda- and psi-classes given in Proposition 2.6. The claimed applications are: a cycle-level refinement of the ELSV formula (Theorem 6.1 and (6.3)); a recursion for hyperelliptic cycles (Corollary 5.1); effective computation of Torelli pullbacks of [A1 x A_{g-1}] (Corollary 4.1); and a new proof of the Okounkov–Pandharipande Gromov–Witten/Hurwitz correspondence (Section 8). The proof proceeds by localization on the epsilon-unramification master spaces of [Nes24] with Zhou's entangled-tail analysis [Zho22], and is verified computationally in Appendix A (Schmitt) for rubber P1 with two relative points, degrees 1 and 2, in low genus.

Significance. If Theorem 2.7 is valid, it is a strong and useful result: a geometric, cycle-theoretic explanation of the relation between admissible covers and stable maps, with all correction terms given by explicit, parameter-free lambda/psi formulas, subsuming ELSV and the OP06 correspondence. The degree-one and degree-two applications (Torelli pullbacks; hyperelliptic loci) are new and would yield effective recursions. Positive features: the I-functions are fully explicit (Proposition 2.6); Appendix A provides reproducible computational checks, with code linked in the paper, against independently computed double-ramification and hyperelliptic cycles, in tautological rings of rank up to 838; the checks are not circular, since the comparison cycles are computed via [JPPZ17] and [SvZ20]. The risk is concentrated exactly where the claimed novelty lies: the proof of the cycle-level residue identity (3.3) is delegated to numerical wall-crossing results, and Remark 3.3 introduces a self-referential dependency on Theorem 6.1 (see major comments). The verification covers only the (h,m)=(0,2) case with trivial insertions and does not exercise the (0,1) wall.

major comments (2)
  1. [§3.5 (Remark 3.3), §3.6 (3.5), §6.1] The proof of Theorem 2.7 contains a circular dependency at its engine. Remark 3.3 states that passing from (3.2) to (3.3) requires the (h,m)=(0,1) wall-crossing 'presented in the proof of Theorem 6.1.' The proof of Theorem 6.1 derives (6.2) from the explicit empty-chamber master space, but then finishes by applying formula (3.5) ('To finish the proof, we apply the formula (3.5) to the class on the right of (6.2)'), and (3.5) is obtained in §3.6 from (3.4), which is a consequence of (3.3). Hence the argument runs in a cycle: (3.3) depends on Theorem 6.1, Theorem 6.1 depends on (3.5), and (3.5) depends on (3.3). The polar-part identity needed in Remark 3.3 is the content of Theorem 6.1's formula, not merely the intermediate statement (6.2), so a charitable reading does not break the cycle. The (h,m)=(0,1) wall-crossing must be proved directly, independently of (3.3)/(3.5), before (3.3) is derived; without that, Theorem 2.7 rests on an unproved lemma.
  2. [§3.5, equations (3.2)–(3.4)] The passage from (3.2) to (3.3) in §3.5 is the load-bearing residue computation, and it is not carried out in the paper: (3.2) is imported from [Nes24, Prop. 6.19], (3.3) is attributed to 'the analysis presented in the proof of [Zho22, Theorem 7.3.3]', and the divisor classes D_i are deliberately left unspecified. The cited results are numerical wall-crossing statements (pushforwards to a point), whereas the paper needs (3.3) as an identity of pushed-forward cycle classes in H*(M_{g,n}(X,mu)), including the substitution z = -psi-tilde_i with negative powers set to zero. The b-sum in (3.3), the identification of the D_i contributions with polar parts of I-functions (Remark 3.3), and the cancellation of the b<0 terms leading to (3.4) all require justification at the level of cycles, not numbers; this is exactly the 'extra layer of complexity' announced in §1.2. Please give a complete derivation of (3.2) to (3.3), or state a precise lemma under which the [Zho22]/[Nes24] analysis applies verbatim to pushed-forward cycle classes and verify its hypotheses.
minor comments (3)
  1. [§5.3] The displayed list of degree-two I-functions should be proofread against (2.5) and Proposition 2.6: entries with a (0,1) connected component, such as the one over M^◦_{1,1} × M^◦_{0,1}, appear to omit the factor z^{-1} contributed by I^◦_{0,1,(1)}, and some genus subscripts do not match the component genera; since the verification in Appendix A depends on these formulas, please state the corrected table explicitly.
  2. [§6.2, equation (6.3)] The step from Theorem 6.1 to (6.3) appeals to a dimension constraint in order to discard all contributing I-functions except I_{0,(2)}; since this is the point where the ELSV formula is recovered, the dimension count, including the treatment of the stable no-marking I-functions of Lemma 7.3, should be written out rather than asserted in one sentence.
  3. [§2.3, §2.8] The class gamma_Gamma of (2.3) is built from source-marking insertions gamma_j that are first defined on the universal curve over M^◦_{h,m}, while the Hurwitz cycle H_{mu,eta} evaluates them at the k moving target markings, i.e., on the universal curve over M^◦_{h,m+k}; the base change implicit in this identification is never stated and should be clarified.

Circularity Check

1 steps flagged · score 6.0 of 10

Proof of Theorem 2.7 loops: residue formula (3.3) is said to need the (0,1) wall-crossing proved in Theorem 6.1, whose proof applies (3.5), which is derived from (3.3).

  1. other [Section 3.5, Remark 3.3; Section 3.6, eq. (3.5); Section 6.1, proof of Theorem 6.1]
    "Remark 3.3. To obtain (3.3) from (3.2), it is also necessary to consider the wall-crossing for (h,m) = (0, 1), which is presented in the proof of Theorem 6.1. The reason is that the contributions of Di are expressible in terms of polar parts of I-functions, as explained in [Zho22, Lemma 7.2.1]. ... [Theorem 6.1:] To finish the proof, we apply the formula (3.5) to the class on the right of (6.2), crossing all walls from ϵ+ ... to ϵ > 1. ... [Section 3.6:] By removing the order on graphs in (3.4), we obtain the following relation ... (3.5)."

    Equation (3.3) is the residue identity that drives Theorem 2.7: Section 3.1 says 'Theorem 2.7 is obtained by applying (3.5) to all walls,' and (3.5) follows from (3.4), which is described as (3.3) after cancellation of terms. But Remark 3.3 states that the passage from (3.2) to (3.3) requires the (h,m)=(0,1) wall-crossing 'presented in the proof of Theorem 6.1.' The proof of Theorem 6.1 then finishes by applying (3.5) to the right-hand side of (6.2). Hence the printed derivation has the mutual dependency (3.3) -> (3.5) -> Theorem 6.1 -> (3.3), and the (0,1) case is not proved independently of (3.5). Appendix A verifies only special rubber-P1 cases with trivial insertions, so it does not supply the general cycle-level residue identity.

full rationale

The paper is not circular at the level of definitions: Gromov-Witten cycles, Hurwitz cycles, I-functions, and the graph sum in Theorem 2.7 are independently defined, and the theorem is not obtained by renaming an input. The low-genus checks in Appendix A compare both sides in nontrivial double-ramification and hyperelliptic cases, so there is substantial independent content. However, the central proof is not self-contained as printed. The engine is the cycle-level residue identity (3.3). Remark 3.3 explicitly defers the (h,m)=(0,1) wall-crossing to the proof of Theorem 6.1, and Theorem 6.1 concludes by applying (3.5), which Section 3.6 derives from (3.3). This is a direct proof-cycle in the claimed derivation chain: Theorem 2.7 depends on (3.5), (3.5) depends on (3.3), and (3.3) depends on Theorem 6.1, whose proof depends back on (3.5). Because the problematic identity is load-bearing for the main theorem and is not supplied by the cited machine-checked or external numerical results, the proof as written has partial circularity. If the cited numerical wall-crossing results from [Nes24] and [Zho22] can be formally lifted to the pushed-forward cycle classes used here, the dependency would become non-circular; as printed, it is looped.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

No empirical constants are fitted. The theorem rests on the cited moduli-space, master-space, and localization results listed above; none of these is replaced by a derivation in this paper, so the axiom burden is high.

assumptions (6)
  • domain assumption Virtual fundamental classes of stable maps and admissible covers, and their pushforwards, are independent of the chosen compactification (Li, Harris-Mumford, orbifold, logarithmic).
    Remark 2.1 asserts this invariance; the theorem is stated for any of these constructions.
  • domain assumption The master space MM^{epsilon_0}_{g,n}(X,mu) of [Nes24, Section 6], based on Zhou's entangled tails, is proper, carries a perfect obstruction theory, and has the C*-fixed locus decomposition (3.1).
    Used in Section 3.4 to write the localization formula; no proof is reproduced in the paper.
  • domain assumption The fixed-component analysis of [Nes24, Prop. 6.19] yields the residue expression (3.2), and the combinatorial cancellation of polar terms follows [Zho22, Thm 7.3.3].
    These cited results carry the core of the wall-crossing computation in Section 3.5.
  • domain assumption Proposition 2.6: the equivariant localization formula for I-functions, I^circ_{g,n,eta}(z) = z^{ell(eta)-|eta|-1} product_j eta_j^{eta_j}/eta_j! times Lambda^vee(z)/product_j(z/eta_j - psi_j), including the unstable cases (g,n) = (0,0), (0,1), holds as stated.
    This formula is quoted from [GP99] and [OP06a]; every non-root vertex class in Theorem 2.7 is evaluated with it.
  • ad hoc to paper The (h,m)=(0,1) wall-crossing used in Remark 3.3 is valid and can be invoked to prove (3.3) without depending on the final statement of Theorem 6.1.
    The text does not isolate this as a lemma; if it only lives inside the proof of Theorem 6.1, which itself uses (3.5), the proof chain is circular.
  • domain assumption Operator formalism for Hurwitz and Hodge integrals: the Burnside formula, the infinite-wedge expression for Hodge integrals via ELSV, and the analytic continuation results of Okounkov and Pandharipande.
    Section 8.1 imports these external results to compute the completed cycles tau^d_k.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Admissible covers and stable maps." pith.science (2026). https://pith.science/paper/AAQKJ4RG

@misc{pith2026250503487,
  author       = {Pith},
  title        = {Pith review of: Admissible covers and stable maps},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/AAQKJ4RG}},
  note         = {Machine review of arXiv:2505.03487}
}
abstract

Moduli spaces of admissible covers and stable maps of target curves give rise to cycles on $\overline{M}_{g,n}$. We prove a formula relating these cycles. It recovers both the Ekedahl-Lando-Shapiro-Vainshtein formula and the Gromov-Witten/Hurwitz correspondence, providing a cycle-theoretic refinement thereof. The formula is verified computationally in low-genus cases with the help of Johannes Schmitt.

Figures

Figures reproduced from arXiv: 2505.03487 by the authors.

Figure 1
Figure 1. Star-shaped graph. together with a gluing map glΓ : MΓ → Mg,Pℓ(µi)+n, such that g = Pk i=1(gi +ℓ(η i ))+g0 −k, and markings are ordered (labelled) by the parts of partitions with a standard order and by labels of the leaves; curves are glued along markings with the same label. On unstable components M ◦ 0,1 and M ◦ 0,2 , the gluing map glΓ corresponds to the forgetful map and to the map which glues two marked points… view at source ↗
Figure 2
Figure 2. Star-shaped graph with leaves at the root. Let F ord Γ be the space associated to FΓ by putting a standard order (i.e., the order which is in agreement with genus labels of Γ) on rational tails of degree d0 and on the ramification points over nodes. Then the space F ord Γ is a Zk-gerbe over the following product, F ord Γ → Mfϵ+ g0,n0 (X , µ, η) × Y k i=1 V C∗ gi,ni,ηi . 3.4. Localisation on the master space. The mas… view at source ↗
Figure 3
Figure 3. Graphs for d = 1, (g, n) = (3, 0). There are just two I-functions in this case: I1,(1)(z) = z − λ1 z − ψ1 = 1 + O(z −1 ) I2,(1)(z) = z 2 − λ1z + λ2 z − ψ1 = z − λ1 + ψ1 + O(z −1 ), (4.1) By Corollary 4.1, GW3 = −[ψ˜ 1, 1]1,2 − [1, λ1]1,2 + [1, ψ1]1,2 + 1 2 [1, 1, 1]1 3 , [PITH_FULL_IMAGE:figures/full_fig_p014_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Graphs for d = 1, (g, n) = (4, 0). In addition to (4.1), there is one more I-function: I3,(1)(z) = z 3 − λ1z 2 + λ2z − λ3 z − ψ1 = z 2 − λ1z + λ2 + ψ1z − λ1ψ1 + ψ 2 1 + O(z −1 ). By Corollary 4.1, GW4 = [ψ˜2 1 , 1]1,3 + [ψ˜ 1, λ1]1,3 + [1, λ2]1,3 − [ψ˜ 1, ψ1]1,3 + [1, …
Figure 5
Figure 5. Figure 5: depicts the graphs involved in this wall-crossing. For simplicity, we do not include vertices associated to I0,(2)(z), as their number can be deduced from other labels. We also omit those graphs whose I-functions are purely polar, as they do not contribute to the wall-…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Torelli loci, product cycles, and the homomorphism conjecture for $\mathcal{A}_g$

    math.AG 2026-01 conditional novelty 7.0 of 10

    For 2≤g≤8, taut([J_g]·[A_2×A_{g-2}]) = taut([J_g])·taut([A_2×A_{g-2}]), and similarly for ([J_6],[A_3×A_3]); the paper also constructs new Gorenstein-kernel classes in compact-type moduli spaces.

Reference graph

Works this paper leans on

24 extracted references · 18 canonical work pages · cited by 1 Pith paper

  1. [1]

    Abramovich, A

    D. Abramovich, A. Corti, and A. Vistoli, Twisted bundles and admissible covers, Commun. Algebra 31 (2003), no. 8, 3547--3618

    work page 2003

  2. [2]

    Oprea, and R

    S Canning, D. Oprea, and R. Pandharipande, Tautological and non-tautological cycles on the moduli space of abelian varieties , preprint, arXiv:2408.08718 (2024)

    arXiv 2024

  3. [3]

    Cavalieri and N

    R. Cavalieri and N. Tarasca, Classes of W eierstrass points on genus 2 curves , Trans. Amer. Math. Soc. 372 (2019), no. 4, 2467--2492. 3988583

    work page 2019

  4. [4]

    Delecroix, J

    V. Delecroix, J. Schmitt, and J. van Zelm, admcycles---a S age package for calculations in the tautological ring of the moduli space of stable curves , J. Softw. Algebra Geom. 11 (2021), no. 1, 89--112. 4387186

    work page 2021

  5. [5]

    Edidin and Z

    D. Edidin and Z. Hu, Chow classes of divisors on stacks of pointed hyperelliptic curves, Ann. Sc. Norm. Super. Pisa Cl. Sci. (5) 25 (2024), no. 1, 217--240. 4732638

    work page 2024

  6. [6]

    Ekedahl, S

    T. Ekedahl, S. Lando, M. Shapiro, and A. Vainshtein, Hurwitz numbers and intersections on moduli spaces of curves., Invent. Math. 146 (2001), no. 2, 297--327

    work page 2001

  7. [7]

    William Fulton and Robert MacPherson, A compactification of configuration spaces, Ann. Math. (2) 139 (1994), no. 1, 183--225

    1994

  8. [8]

    Fantechi and R

    B. Fantechi and R. Pandharipande, Stable maps and branch divisors, Compos. Math. 130 (2002), no. 3, 345--364

    work page 2002

Show all 24 references
  1. [9]

    Faber and R

    C. Faber and R. Pandharipande, Relative maps and tautological classes, J. Eur. Math. Soc. 7 (2005), no. 1, 13--49

    2005

  2. [10]

    Graber and R

    T. Graber and R. Pandharipande, Localization of virtual classes, Invent. Math. 135 (1999), no. 2, 487--518

    1999

  3. [11]

    Graber and R

    T. Graber and R. Vakil, Hodge integrals and Hurwitz numbers via virtual localization , Compos. Math. 135 (2003), no. 1, 25--36

    2003

  4. [12]

    Harris and D

    J. Harris and D. Mumford, On the Kodaira dimension of the moduli space of curves , Invent. Math. 67 (1982), 23--86

    1982

  5. [13]

    Janda, R

    F. Janda, R. Pandharipande, A. Pixton, and D. Zvonkine, Double ramification cycles on the moduli spaces of curves, Publ. Math. Inst. Hautes \'Etudes Sci. 125 (2017), 221--266. 3668650

    2017

  6. [14]

    Kim, Logarithmic stable maps, New developments in algebraic geometry, integrable systems and mirror symmetry

    B. Kim, Logarithmic stable maps, New developments in algebraic geometry, integrable systems and mirror symmetry. Papers based on the conference ``New developments in algebraic geometry, integrable systems and mirror symmetry'', Kyoto, Japan, January 7--11, 2008, and the worksh...

    2008

  7. [15]

    Kontsevich, Enumeration of rational curves via torus actions, The moduli space of curves

    M. Kontsevich, Enumeration of rational curves via torus actions, The moduli space of curves. Proceedings of the conference held on Texel Island, Netherlands during the last week of April 1994, 1995, pp. 335--368

    1994

  8. [16]

    Li, Stable morphisms to singular schemes and relative stable morphisms, J

    J. Li, Stable morphisms to singular schemes and relative stable morphisms, J. Differ. Geom. 57 (2001), no. 3, 509--578

    2001

  9. [17]

    Nesterov, Gromov--Witten/Hurwitz wall-crossing , preprint, arXiv:2208.00889 (2022)

    D. Nesterov, Gromov--Witten/Hurwitz wall-crossing , preprint, arXiv:2208.00889 (2022)

    2022 arXiv

  10. [18]

    , U nramified Gromov--Witten and Gopakumar--Vafa invariants , preprint, arXiv :2405.18398 (2024)

    2024 arXiv

  11. [19]

    , Hilbert schemes of points and Fulton--MacPherson compactifications , preprint, arXiv:2501.08269 (2025)

    2025 arXiv

  12. [20]

    Okounkov, Toda equations for Hurwitz numbers , Math

    A. Okounkov, Toda equations for Hurwitz numbers , Math. Res. Lett. 7 (2000), no. 4, 447--453

    2000

  13. [21]

    Okounkov and R

    A. Okounkov and R. Pandharipande, The equivariant Gromov - Witten theory of \( P^1\) , Ann. Math. (2) 163 (2006), no. 2, 561--605

    2006

  14. [22]

    , Gromov- Witten theory, Hurwitz theory, and completed cycles , Ann. Math. (2) 163 (2006), no. 2, 517--560

    2006

  15. [23]

    Schmitt and J

    J. Schmitt and J. van Zelm, Intersections of loci of admissible covers with tautological classes, Selecta Math. (N.S.) 26 (2020), no. 5, Paper No. 79, 69. 4177576

    2020

  16. [24]

    Zhou , Quasimap wall-crossing for GIT quotients , Invent

    Y. Zhou , Quasimap wall-crossing for GIT quotients , Invent. Math. 227 (2022), no. 2, 581--660

    2022

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.