Quantum Master Equation and Open Gromov-Witten theory

Vito Iacovino

REVIEW 4 major objections 5 minor 1 cited by

Quantum Master Equation and Open Gromov-Witten theory

T0 review · 4 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read The paper constructs the higher-genus Open-Closed Gromov-Witten potential as a solution of the quantum master equation, defined up to quantum master isotopy.

desk verdict Plausible and important, but the factorization argument is a sketch and the factorization map has a domain error that needs fixing. read the letter →

arxiv 2412.04230 v1 pith:NHJ4UASO submitted 2024-12-05 math.SG

Vito Iacovino This is my paper · ORCID

classification math.SG MSC 53D45

keywords quantummasterequationOpenGromov-Wittentheorymulti-curvecyclesBatalin-VilkoviskyformalismChern-SimonspropagatorKuranishistructuresfactorizationpropertyMaslovindexzeroLagrangian

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

This paper establishes that the higher-genus Open-Closed Gromov-Witten potential exists as a solution of the quantum master equation, defined up to the natural notion of quantum master isotopy. The setting is a Calabi-Yau six-manifold $X$ with a Maslov-index-zero Lagrangian $L$, the same pair physicists use for the open topological string. The construction packages the counts of pseudo-holomorphic multi-curves into multi-curve cycles, couples them with the abelian Chern-Simons propagator, and shows that the connected graph sum $W$ satisfies $\frac{1}{2}\{W,W\}+g_s\Delta W=0$ whenever the cycles satisfy the factorization property. This matters because open curve counts have historically depended on perturbation choices due to the boundary of the moduli space; the quantum master equation makes the perturbation dependence an isotopy freedom rather than an obstruction. If the construction is right, it supplies a rigorous higher-genus open Gromov-Witten theory with numerical invariants obtained from the effective action at critical points.

What carries the argument

The central object is the multi-curve cycle (MC-cycle): a formal power series in $g_s$ whose generators are collections of disjoint closed one-dimensional chains on a compact oriented three-manifold (the Lagrangian), together with degree labels $\beta_v\in H_2(X,L)$. These cycles arise from the moduli spaces of pseudo-holomorphic multi-curves, whose corner faces model the splitting of curves and whose perturbations define the cycles. The factorization property is the additional identity that makes the sum over all classes behave like an exponential: the cycle for $\beta_1+\beta_2$ is the product of the cycles for $\beta_1$ and $\beta_2$. Coupling MC-cycles with the abelian Chern-Simons propagator $P\in\Omega^2(\operatorname{Conf}_2(M))$ turns chains into functions on the odd symplectic vector space $H^*(M)[1]$; the propagator identities are exactly what converts the geometric boundary relations into the quantum master equation.

What would settle it

Compute the two sides of the factorization identity $fact_{\beta_1,\beta_2}(Z_{\beta_1+\beta_2})=Z_{\beta_1}\boxtimes Z_{\beta_2}$ for a concrete low-genus multi-curve in a simple pair such as $X=T^*L$ with the zero-section Lagrangian; if, for some choice of perturbations, the two sides are not connected by the allowed crossing-and-gluing moves of the MC-chain complex, the potential $W(Z)$ is not defined and the main theorem fails.

Watch

Extended reading notes

Core claim

The paper constructs, for each class $\beta\in H_2(X,L)$ with $L$ a Maslov-index-zero Lagrangian in a Calabi-Yau six-manifold $X$, a multi-curve cycle $Z_\beta$ attached to the moduli space of pseudo-holomorphic multi-curves in class $\beta$, and it proves that the collection can be chosen to satisfy the factorization property $fact_{\beta_1,\beta_2}(Z_{\beta_1+\beta_2})=Z_{\beta_1}\boxtimes Z_{\beta_2}$. With this property, the connected graph sum $W(Z)$ is well defined and the full partition function satisfies $P(Z)=\exp(g_s^{-1}W(Z))$. The quantum master equation $\frac{1}{2}\{W,W\}+g_s\Delta W=0$ then holds up to quantum master isotopy, meaning the potential is a well-defined object in the Batalin-Vilkovisky formalism rather than a fixed function. The paper also shows the effective action at a nondegenerate critical point is isotopy invariant and extends the whole construction to bulk deformations by four-chains.

Load-bearing premise

The load-bearing premise is that the moduli spaces of pseudo-holomorphic multi-curves admit a consistent way to define virtual counts (Kuranishi structures and compatible perturbations) that also respects the factorization property; this is imported from the author's earlier preprint [7] and adapted here, but the adaptation is asserted rather than proved.

Editorial extensions

If this is right

The open-closed partition function $P(Z)=\sum_\beta P(Z_\beta)T^{\omega(\beta)}$ equals $\exp(g_s^{-1}W(Z))$, so all-genus information is encoded in the connected potential $W$.
The potential $W$ satisfies the quantum master equation up to quantum master isotopy, and its solution class is independent of the choices of almost complex structure, perturbations, and propagator.
The effective action $W_{\mathrm{eff}}$ evaluated at a nondegenerate critical point of the classical potential is invariant under quantum master isotopy, yielding numerical open Gromov-Witten invariants.
Bulk deformations by any finite collection of four-chains $A_1,\dots,A_l$ can be included; shifting the bounding four-chain $K$ by $\sum r_i A_i$ is equivalent to shifting the bulk variables $b_i$ by $r_i g_s$.

Reading between the lines

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

If the factorization-compatible virtual-count theory can be made fully explicit, the same chain-level construction should extend the potential to families of Lagrangians or to relative settings with several Lagrangian components, since the MC-cycle formalism is not tied to a single pair.
The 'up to quantum master isotopy' statement suggests that the underlying invariant is a chain-level object, and that genuinely numerical invariants arise only after evaluating the effective action at a critical point; this matches the physical idea that the open-string background must be fixed before numbers appear.
A concrete consistency check would be to compare the genus-zero part of $W$ with the usual closed Gromov-Witten potential when the boundary chains are turned off; the quantum master equation should then reduce to the associativity or WDVV equations of the closed theory.
The role of the abelian Chern-Simons propagator suggests that the MC-cycle partition function should agree with a point-splitting regularization of abelian Wilson-loop expectations, so computing the same invariant in both formalisms would independently test the factorization property.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Short version: this is a real step in a serious program, but the central new ingredient—factorization-compatible MC-cycles—is asserted rather than proved, and one definition in the paper is wrong as written. The paper deserves a referee, but it needs real work.

What is new: the construction of W(Z) as connected graph contribution and the claim that factorization gives P = exp(W/gs) and the QME. That is the natural missing piece in the author's series, and the formal argument in §3.4.3 is correct if factorization holds. The propagator section is clean and self-contained. The bulk deformation extension is a useful addition.

Soft spots:

1. The factorization map (17) is not well-defined. In a generator (5) each h is attached to a specific vertex v(h). The sum over all partitions H1⊔H2 of H independently of V produces terms where h∈H1 but v(h)∈V2; for those terms (w_{v,h}) is undefined. The map should split H according to the vertex sets: H_i = {h : v(h) ∈ V_i}. As written, the domain is not Zβ. This looks like a fixable typo, but it matters because the whole factorization formalism hangs on this map.

2. The proof of Prop 3 is a sketch of a sketch. The new hypothesis that the obstruction bundle of M_{G,m} contains the obstruction bundle of M_{cut E(G)} for every E is stated with no existence argument; the text says the inductive argument of [7] can be adapted, but that is the load-bearing claim. Likewise the restriction to product perturbations and C^0-closeness to s† is asserted, and Lemma 5 is adapted without checking the isotopy stays in the restricted class. If this fails, (18) fails and the QME for W is unjustified.

3. Prop 4 is outlined with 'standard graph techniques'. I believe it, but it is not a proof.

The paper is not self-contained: the core input is Theorem 1 from [7], an unpublished preprint. That is not by itself a flaw, but it makes independent verification hard. The reader's circularity worry is fair but I would not call it circularity: the factorization property is constructed, not predicted. That is legitimate if the construction is actually carried out.

Bottom line: If the author fixes (17) and supplies the missing perturbation theory, this is a significant result. As it stands, it is an important conjecture-plus-framework with a plausible but unverified central step. Send it to a knowledgeable referee, and ask that referee to focus on §5.3 and the definition of fact.

Referee Report

4 major / 5 minor

Summary. The paper claims to construct the higher genus open-closed Gromov-Witten potential as a solution of the quantum master equation, defined up to quantum master isotopy. The construction reviews the multi-curve cycles Zβ from the author's preprint [7], defines a partition function P by pairing MC-cycles with an abelian Chern-Simons propagator, and shows that isotopies of MC-cycles give master homotopies. It then introduces a factorization property for the collection (Zβ) and defines the potential W from connected graphs, asserting P=exp(W/gs). Section 4 extends the construction to bulk deformations, Section 5 attempts to prove the factorization property by adapting Kuranishi perturbation theory from [7], and an appendix constructs the propagator. The main theorem therefore rests on two load-bearing inputs: the existence of factorization-compatible MC-cycles and the exact equality (18).

Significance. If the construction were fully established, the paper would be a significant contribution: it would supply a concrete framework in which an open-closed Gromov-Witten potential satisfies the quantum master equation, and it would connect that potential to the abelian Chern-Simons propagator and to multi-curve moduli spaces. The paper contains some genuinely useful concrete steps: Proposition 2 gives an explicit local computation showing that jumps in P are compensated by the propagator pole, the appendix gives a self-contained construction of the propagator and its one-parameter version, and the cancellation mechanism outlined in Proposition 4 is plausible. However, the central claim is not proven to the standard required by the journal. The factorization property is not proved but imposed as an unverified assumption on the perturbation theory imported from [7], and the factorization map itself has a well-definedness defect. The advertised result is therefore conditional on substantial missing technical work.

major comments (4)

[§3.4.2, Eq. (17)] The factorization map fact_{β1,β2} is not well-defined on the generators (5). Formula (17) sums over arbitrary partitions H1⊔H2=H and V1⊔V2=V, but in a generator each half-edge h∈H is attached to a unique vertex v(h)∈V through the data (w_{v,h}). A term with h∈H1 and v(h)∈V2 is not a generator of Z_{β1}⊗Z_{β2}, since the chain w_{h,v} does not belong to the first factor in that case. The map should be restricted to partitions satisfying H_i = v^{-1}(V_i), with a separate convention for the annulus labels w^{ann}_h. As written, Eq. (18) is not a statement about the intended MC-cycles.
[§5.3, Proposition 3] The proof of the factorization property is a sketch and does not establish the existence of the required Kuranishi structures and perturbations. The key new hypothesis—that the obstruction bundle of M_{G,m} contains the obstruction bundle of M_{cut E(G)} for every E⊂E(G)—is introduced with 'We assume', and no argument is given that such Kuranishi structures exist or that they can be produced by the inductive method of [7]. The text then says 'We can adapt the inductive argument of [7]' and 'Adapting Lemma 5 to these class of perturbations', but the adaptation is not carried out. In particular, it is not shown that the isotopy arising from (44) and (47) stays inside the restricted class of product perturbations that are C^0-close to s†. Since the exact equality (18) is the only bridge between the partition function and the potential W in §3.4.3, this gap is load-bearing.
[§5.3, Lemma 5] The claimed isotopy between fact_{β1,β2}(Zβ) and Zβ1⊠Zβ2 is not proved. After constructing the two MC-cycles via (42) and (46), the proof only asserts that the corresponding Kuranishi structures and perturbations are isotopic, referring to 'general argument, as in [7]'. But the two sides are built from different moduli spaces, and the isotopy must respect the corner identifications (43) as well as the factorization-compatible perturbation class used for the nice MC-cycles. None of these compatibility conditions is verified. Lemma 5 can therefore not serve as a black-box substitute for the missing proof of Proposition 3.
[§3.4.3 and §5.3] The advertised well-definedness of W 'up to quantum master isotopy' is not established. Section 3.4.3 shows that if ilde Z is a factorization-compatible isotopy, then W satisfies the quantum master equation, but it does not prove that any two choices entering the construction of (Zβ)β are related by such an isotopy. The last paragraph of Section 5.3 only states that 'The same argument applies to construct isotopies of nice MC-cycles satisfying the factorization property', without proof. Thus the paper does not currently show that different choices lead to isotopic W, which is part of the main claim in the abstract.

minor comments (5)

[Throughout] There are numerous typographical errors and misspellings, including 'simplectic', 'pseudoholomophic', 'mutlti-curves', 'balk deformation', and 'pertubative'. These should be corrected throughout.
[§2.1, Eq. (5); §3.2, Eq. (14)] The notation w_{s(h),v_h} in Eq. (14) is ambiguous: v_h should be v_{s(h)}, and the incidence relation between H and V should be stated explicitly before the partition-function formula is applied.
[§4.4] The displayed relation P(β,K+rA,A)(gs,b) = P(β,K,A)(gs,b+rgs) has inconsistent notation on the right-hand side; if the left side includes A in the list of bulk deformations, the right side should also specify the same deformation variable consistently.
[§6, Lemma 6] The statement says P∈Ω²(Conf₂(L)) while the surrounding text and the proof are about Conf₂(M); this is presumably a typo and should be corrected.
[References] Reference [9] is cited without an arXiv identifier or publication data, and the paper relies heavily on preprints [3]-[9]; the author should state precisely which results are assumed from each preprint.

Circularity Check

1 steps flagged · score 4.0 of 10

The factorization property needed for the quantum master equation is imported from the author's prior preprint [7] via an asserted adaptation; the rest of the construction is self-contained.

self citation load bearing [Section 5.3, proof of Proposition 3 (factorization property)]
"We can adapt the inductive argument of [7] used to construct the perturbation of the moduli space imposing the above conditions. ... In this way we obtain the nice MC-cycle Z≤κ ... Adapting Lemma 5 to these class of perturbations, we can prove that Z≤κ satisfies the factorization property."

Proposition 3 is the only asserted support for the factorization property (18), and (18) is the sole bridge from the partition function P(Z) to the potential W(Z) and to the quantum master equation in Section 3.4.3. The proof of Proposition 3 does not construct the required Kuranishi structures or the restricted perturbations; it assumes the obstruction-bundle containment and product/C0-closeness conditions and then appeals to the same author's preprint [7] both for the inductive perturbation construction and for the isotopy statement ('as in [7]'). Since [7] is not independently verified in the present paper and the adaptation is asserted rather than proved, the central existence claim on which the QME result depends is justified by a load-bearing self-citation.

full rationale

The paper is transparent that the exponential relation P(Z) = exp(W(Z)/gs) is not an independent prediction but a consequence of imposing the factorization property: it says 'In order for this to be true Z needs to satisfies what we call factorization property.' The combinatorial step from factorization to the QME for W is a formal implication of equation (16), not a circular reduction. The BV formalism, the Chern-Simons propagator, and the isotopy argument for P(Z) are developed in the paper itself. The main circularity concern is localized to Proposition 3 and Section 5.3, where the existence of MC-cycles satisfying strict factorization is delegated to an unproved adaptation of the author's earlier preprint [7] under new assumptions. That is a load-bearing self-citation and a gap, but it is not a case where a fitted parameter is renamed a prediction or where an equation reduces to its own input. A separate well-definedness issue with the factorization map (17) is a correctness risk rather than a circularity. Overall score 4: some self-citation is load-bearing, but the central construction retains substantial independent content.

Assumptions & free parameters 0 free parameters · 4 assumptions · 3 invented entities

The paper introduces no numerically fitted constants. Formal variables (gs, T, b_i) are bookkeeping for the expansion. The central claim rests on domain assumptions about the existence of Kuranishi structures and perturbations satisfying compatibility and factorization conditions, imported from the author's prior preprints, plus a new axiom-like restriction to factorization-compatible perturbations.

assumptions (4)

domain assumption The moduli spaces of pseudo-holomorphic multi-curves admit Kuranishi structures and compatible perturbations with the properties listed in Sections 4.1 and 5.3, in particular obstruction bundle inclusion and factorization compatibility.
Theorem 1 is cited to [7]; the factorization version is asserted in Prop. 3 and argued in Section 5.3 by adapting [7]. No complete proof is given in this paper.
domain assumption The constructed MC-cycles Zβ are independent of choices (almost complex structure, perturbations, propagator) up to isotopy.
Stated as part of Theorem 1 from [7] and used throughout, e.g. Sections 3.3 and 4.4, without proof here.
standard math The abelian Chern-Simons propagator exists with properties (1)-(4) of Lemma 6 and its one-parameter version Lemma 9.
Proved in the appendix within the paper using standard de Rham theory; it is a non-trivial analytic tool but the argument is sketched.
ad hoc to paper The factorization property can be imposed without changing the isotopy class of the MC-cycles.
The paper restricts the class of allowed perturbations specifically to force fact(Z)=Z⊗Z (Section 5.3). This is a structural choice rather than a consequence of geometry alone.

invented entities (3)

Nice MC-cycles and multi-curve cycles
purpose: Replace the naive count of open pseudo-holomorphic curves by algebraic cycles on a three-manifold, to solve the boundary problem.
Introduced in author's prior work [4],[5],[7]; carried forward without new falsifiable handle outside the paper.
Extended nice MC-cycles with finite limit support
purpose: Used in Section 3.4.1 to state the factorization property for the full MC-cycle.
A new technical object defined in this paper; no independent evidence beyond the construction.
MC-cycles with bulk deformations
purpose: Incorporate bulk deformations via four-chains A into the partition function and QME.
Defined in Section 4.3; depends on the same chain-level framework.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Quantum Master Equation and Open Gromov-Witten theory." pith.science (2026). https://pith.science/paper/NHJ4UASO

@misc{pith2026241204230,
  author       = {Pith},
  title        = {Pith review of: Quantum Master Equation and Open Gromov-Witten theory},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NHJ4UASO}},
  note         = {Machine review of arXiv:2412.04230}
}

read the original abstract

We construct the higher genus Open-Closed Gromov-Witten potential as a solution of the quantum master equation defined up to quantum master isotopy.

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

Quantum Master Equation and Open Gromov-Witten Theory 2
math.SG 2024-12 reject novelty 6.0 of 10

The paper defines the non-abelian open Gromov-Witten potential and derives a quantum master equation for it, conditional on auxiliary constructions from the author's companion papers.

Reference graph

Works this paper leans on

23 extracted references · 2 canonical work pages · cited by 1 Pith paper

[7]

Gaiotto, G

D. Gaiotto, G. Moore, A. Neitzke, Framed BPS States , arXiv:1006.0146

arXiv
[1]

Bullimore, T

M. Bullimore, T. Dimofte, D. Gaiotto, and J. Hilburn, Boundaries, Mirror Symmetry, and Symplectic Duality in 3d N = 4 Gauge Theory, arXiv:1603.0838

arXiv
[2]

Cecotti and C

S. Cecotti and C. Vafa, BPS Wall Crossing and Topological Strings , arXiv: 0910.2615

arXiv
[3]

Cecotti, C

S. Cecotti, C. Cordova, and C. Vafa, Braids, Walls, and Mirrors , arXiv:1110.2115

arXiv
[4]

Cordova, S

C. Cordova, S. Espahbodi, B. Haghighat, A. Rastogi, and C. Vafa, Tangles, generalized Reidemeister moves, and thre-dimensional mirror symmetry, arXiv:1211.3730

arXiv
[5]

Gaiotto, G

D. Gaiotto, G. W. Moore and A. Neitzke, Four-dimensional wall-crossing via threedimensional field theory, arXiv:0807.4723

arXiv
[6]

Gaiotto, G

D. Gaiotto, G. W. Moore and A. Neitzke, Wall-crossing, Hitchin Systems, and the WKB Approximation, arXiv:0907.3987

arXiv
[8]

Gaiotto, G

D. Gaiotto, G. Moore, A. Neitzke, Wall-Crossing in Coupled 2d-4d Systems, arXiv:1103.2598

arXiv

Show all 23 references

[9]

Fukaya; K

K. Fukaya; K. Ono, Arnold conjecture and Gromov-Witten invariant, Topology 38 (1999), no. 5, 933--1048

1999
[10]

Fukaya, Y.-G

K. Fukaya, Y.-G. OH, H. Ohta, K. Ono, Lagrangian intersection Floer theory - anomaly and obstruction
[11]

Kontsevich and Y

M. Kontsevich and Y. Soibelman, Stability Structures, Motivic Donaldson-Thomas Invariants and Cluster Transformations , arXiv:0811.2435

arXiv
[12]

Iacovino, Master Equation and Perturbative Chern-Simons theory, arXiv:0811.2181

V. Iacovino, Master Equation and Perturbative Chern-Simons theory, arXiv:0811.2181

arXiv
[13]

Iacovino, Open Gromov-Witten theory on Calabi-Yau three-fold I, arXiv:0907.5225

V. Iacovino, Open Gromov-Witten theory on Calabi-Yau three-fold I, arXiv:0907.5225

arXiv
[14]

Iacovino, Open Gromov-Witten theory on Calabi-Yau three-folds II, arXiv:0908.0393

V. Iacovino, Open Gromov-Witten theory on Calabi-Yau three-folds II, arXiv:0908.0393

arXiv
[15]

Iacovino, Frame ambiguity in Open Gromov-Witten invariants, arXiv:1003.4684

V. Iacovino, Frame ambiguity in Open Gromov-Witten invariants, arXiv:1003.4684

arXiv
[17]

Iacovino, Open Gromov-Witten theory without Obstruction, arXiv:1711.05302

V. Iacovino, Open Gromov-Witten theory without Obstruction, arXiv:1711.05302

arXiv
[18]

Iacovino, Kontsevich-Soibelman Wall Crossing Formula and Holomorphic Disks, arXiv:1711.05306

V. Iacovino, Kontsevich-Soibelman Wall Crossing Formula and Holomorphic Disks, arXiv:1711.05306

arXiv
[19]

Iacovino, Open Gromov-Witten invariants and Bounrdary States,

V. Iacovino, Open Gromov-Witten invariants and Bounrdary States,
[20]

Iacovino, Holomorphic Lagrangian Floer Homology, in preparation

V. Iacovino, Holomorphic Lagrangian Floer Homology, in preparation
[21]

Iacovino, Refined Wall Crossing and Open Holomorphic Curves, in preparation

V. Iacovino, Refined Wall Crossing and Open Holomorphic Curves, in preparation
[22]

Ooguri, C

H. Ooguri, C. Vafa, Knot invariants and topological strings, Nuclear Phys. B 577 (2000) 419â 438 MR1765411

2000
[23]

Witten, Chern-Simons Gauge Theory As A String Theory, arXiv:hep-th/9207094

E. Witten, Chern-Simons Gauge Theory As A String Theory, arXiv:hep-th/9207094

arXiv
[24]

Zaslow, Wavefunctions for a Class of Branes in Three-space, arXiv:1803.02462

E. Zaslow, Wavefunctions for a Class of Branes in Three-space, arXiv:1803.02462

arXiv

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.

[1] [7]

Gaiotto, G

D. Gaiotto, G. Moore, A. Neitzke, Framed BPS States , arXiv:1006.0146

arXiv

[2] [1]

Bullimore, T

M. Bullimore, T. Dimofte, D. Gaiotto, and J. Hilburn, Boundaries, Mirror Symmetry, and Symplectic Duality in 3d N = 4 Gauge Theory, arXiv:1603.0838

arXiv

[3] [2]

Cecotti and C

S. Cecotti and C. Vafa, BPS Wall Crossing and Topological Strings , arXiv: 0910.2615

arXiv

[4] [3]

Cecotti, C

S. Cecotti, C. Cordova, and C. Vafa, Braids, Walls, and Mirrors , arXiv:1110.2115

arXiv

[5] [4]

Cordova, S

C. Cordova, S. Espahbodi, B. Haghighat, A. Rastogi, and C. Vafa, Tangles, generalized Reidemeister moves, and thre-dimensional mirror symmetry, arXiv:1211.3730

arXiv

[6] [5]

Gaiotto, G

D. Gaiotto, G. W. Moore and A. Neitzke, Four-dimensional wall-crossing via threedimensional field theory, arXiv:0807.4723

arXiv

[7] [6]

Gaiotto, G

D. Gaiotto, G. W. Moore and A. Neitzke, Wall-crossing, Hitchin Systems, and the WKB Approximation, arXiv:0907.3987

arXiv

[8] [8]

Gaiotto, G

D. Gaiotto, G. Moore, A. Neitzke, Wall-Crossing in Coupled 2d-4d Systems, arXiv:1103.2598

arXiv