arxiv: 2604.14279 · v1 · submitted 2026-04-15 · ✦ hep-th · hep-ph

Recognition: unknown

Beyond the Dilute Instanton Gas: Resurgence with Exact Saddles in the Double Well

Aur\'elien Dersy , Matthew D. Schwartz

Authors on Pith no claims yet

Pith reviewed 2026-05-10 12:33 UTC · model grok-4.3

classification ✦ hep-th hep-ph

keywords double wellinstantonsresurgencePicard-LefschetzWeierstrass functionsenergy splittingspath integralfinite temperature

0 comments

The pith

Exact finite-time saddles encode the full resurgent structure of all energy splittings in the double well through a Picard-Lefschetz contour integral over quasi-zero modes.

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

The paper establishes that replacing the dilute instanton gas with exact saddles at finite Euclidean time lets the path integral for the double well compute the partition function and energy levels systematically at each instanton order. A sympathetic reader would care because this approach captures non-perturbative splittings for every excited state, including their full dependence on level number, while making the cancellation of ambiguities geometrically transparent in a finite-dimensional contour integral. The method uses Weierstrass elliptic functions to build the saddles, Lamé operators for fluctuations, and Picard-Fuchs equations for periods, yielding a framework that both overlaps with and complements Exact WKB. This sidesteps the dilute-gas limitation to only the ground-state splitting and provides a concrete way to read off asymptotic growth and contributing saddles directly from the integral.

Core claim

At each instanton order the full resurgent structure—which saddles contribute, what asymptotic growth is expected, and how ambiguities cancel—is encoded in a finite-dimensional Picard-Lefschetz contour integral over the quasi-zero modes of exact finite-T saddles constructed with Weierstrass elliptic functions. Working at finite T is essential: the dilute instanton gas can only access the ground-state splitting, whereas the exact finite-T computation systematically produces the non-perturbative energy splittings for all excited states, including their full dependence on the level number.

What carries the argument

The finite-dimensional Picard-Lefschetz contour integral over quasi-zero modes of exact finite-T saddles, which encodes which saddles contribute, the expected asymptotic growth, and how perturbative ambiguities cancel at each instanton order.

If this is right

The partition function and all energy levels become computable systematically order by order in the instanton expansion.
Non-perturbative splittings for excited states appear with their complete dependence on the level number.
Ambiguities cancel geometrically inside the contour integral, fixing the trans-series without ad-hoc prescriptions.
The same framework reproduces known results for the ground-state splitting while extending them to higher states.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The contour-integral encoding might be portable to other potentials that admit exact elliptic saddles, offering a route to resurgent trans-series beyond the double well.
Because the method produces level-number-dependent splittings, it could be tested against high-precision numerical spectra to check whether the geometric cancellation persists at higher orders.
The overlap with Exact WKB suggests a possible dictionary between the finite-T saddle geometry and the complex WKB connection formulas.

Load-bearing premise

That the exact finite-T saddles can be constructed explicitly with Weierstrass functions and that the Picard-Lefschetz contour integral over quasi-zero modes captures every relevant contribution without missing higher-order or non-saddle effects.

What would settle it

Explicit evaluation of the contour integral at the one-instanton and two-instanton orders for the splitting of the first excited state, followed by direct numerical comparison against the eigenvalues obtained by diagonalizing the double-well Schrödinger operator on a large grid.

Figures

Figures reproduced from arXiv: 2604.14279 by Aur\'elien Dersy, Matthew D. Schwartz.

read the original abstract

The path-integral approach to the double well has long been limited by the dilute instanton gas approximation. We show that if the finite Euclidean-time structure is taken seriously by using exact saddles, the dilute gas can be sidestepped, allowing the partition function and energy levels to be computed systematically. At each instanton order, the full resurgent structure -- which saddles contribute, what asymptotic growth is expected and how ambiguities cancel -- is encoded in a finite-dimensional Picard--Lefschetz contour integral over the quasi-zero modes with a clear geometric interpretation. Working at finite $T$ is essential: the dilute instanton gas can only access the ground-state splitting, whereas the exact finite-$T$ computation systematically produces the non-perturbative energy splittings for all excited states, including their full dependence on the level number. The key ingredients -- Weierstrass elliptic functions for the saddles, Lam\'e operators for the fluctuations and Picard--Fuchs equations for the periods -- form a coherent mathematical framework that both overlaps and complements that of Exact WKB.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper sketches a path-integral route to all-level splittings in the double well by replacing the dilute instanton gas with exact finite-T saddles and a finite Picard-Lefschetz contour over quasi-zero modes.

read the letter

The main point is that this work replaces the usual dilute instanton gas with exact periodic saddles at finite Euclidean time, then encodes the full resurgent information for every energy level in a finite-dimensional contour integral over the quasi-zero-mode moduli space. That move directly removes the ground-state-only restriction and produces level-dependent splittings in a systematic way at each instanton order. The geometric picture of the contour deciding saddle contributions, asymptotics, and ambiguity cancellation is the cleanest new element here. Using Weierstrass functions for the saddles, Lamé operators for the fluctuations, and Picard-Fuchs equations for the periods keeps everything on standard mathematical ground and gives a coherent framework that overlaps with but stays inside the path-integral language of Exact WKB. This is useful for anyone who wants to stay with instanton methods rather than switch to WKB entirely. The abstract lays out the program clearly and the ingredients are reproducible from known special-function properties, so the basic setup looks internally consistent. The soft spot is the lack of any explicit low-order calculation or numerical check in the material provided. Without seeing at least one concrete contour evaluation that reproduces a known multi-instanton splitting or Stokes jump, it remains open whether the reduction to quasi-zero modes really captures every contribution or leaves out non-saddle effects. The stress-test concern about omitted terms in the contour is the part that needs direct verification in the full text. This is written for people already working on resurgence and instantons in quantum mechanics. A reader comfortable with elliptic integrals and Picard-Lefschetz theory will get the most value and can judge the contour claim quickly. The paper is coherent enough on its own terms and targets a genuine limitation in the literature, so it deserves a serious referee who can ask for the missing explicit checks.

Referee Report

2 major / 3 minor

Summary. The paper argues that the path integral for the double-well potential can be treated beyond the dilute instanton gas by using exact finite-T periodic saddles constructed from Weierstrass elliptic functions. It claims that the full resurgent structure at each instanton order—including which saddles contribute, asymptotic growth, and cancellation of ambiguities—is encoded in a finite-dimensional Picard-Lefschetz contour integral over quasi-zero modes. This framework, employing Lamé fluctuation operators and Picard-Fuchs periods, systematically yields the partition function and non-perturbative splittings for all energy levels (not just the ground state), while overlapping with and complementing Exact WKB methods.

Significance. If the central claims hold, the work supplies a geometrically transparent, finite-dimensional encoding of resurgence that sidesteps the dilute-gas limitation and directly accesses level-dependent splittings. This could serve as a template for other potentials where exact saddles exist and might bridge semiclassical and resurgence techniques in quantum mechanics and potentially in field theory.

major comments (2)

[Section on the Picard-Lefschetz contour integral and quasi-zero modes] The central claim that the Picard-Lefschetz contour integral over quasi-zero modes alone encodes the complete resurgent structure (saddle contributions, Stokes jumps, and ambiguity cancellation) at every instanton order requires explicit verification. The manuscript must demonstrate, for at least the first few instanton sectors, that this finite-dimensional integral reproduces the known Borel-resurgent relations and level splittings of the double well without residual contributions from non-saddle configurations or the infinite tower of non-zero modes.
[Discussion of the Lamé operator and Picard-Fuchs periods] The reduction from the infinite-dimensional fluctuation operator (Lamé operator) to the finite-dimensional contour integral must be shown to capture all relevant Stokes phenomena. The paper should provide a concrete check that the integration cycle is fixed internally by the geometry of the exact saddles and does not require external input from Exact WKB to determine the correct combination of saddles.

minor comments (3)

[Construction of exact saddles] Clarify the precise range of T (or the dimensionless parameter) for which the Weierstrass elliptic solutions remain valid periodic saddles and how the T→∞ limit recovers the standard instanton gas.
[Results for energy levels] Add a short comparison table or plot showing the first few energy-level splittings obtained from the contour integral against known numerical or perturbative results for the double well.
[Notation and appendices] Ensure consistent notation for the quasi-zero-mode moduli space and the Picard-Fuchs periods across sections; a brief appendix summarizing the relevant elliptic-function identities would aid readability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments on our manuscript. The suggestions for explicit verification are well taken, and we have revised the paper to address both major points directly.

read point-by-point responses

Referee: The central claim that the Picard-Lefschetz contour integral over quasi-zero modes alone encodes the complete resurgent structure (saddle contributions, Stokes jumps, and ambiguity cancellation) at every instanton order requires explicit verification. The manuscript must demonstrate, for at least the first few instanton sectors, that this finite-dimensional integral reproduces the known Borel-resurgent relations and level splittings of the double well without residual contributions from non-saddle configurations or the infinite tower of non-zero modes.

Authors: We agree that explicit verification is required to fully substantiate the claim. In the revised manuscript we have added a dedicated subsection performing the contour integral explicitly in the one- and two-instanton sectors. These calculations reproduce the known Borel-resurgent relations and the non-perturbative splittings for the ground and first excited states. The non-zero modes of the Lamé operator enter only through the perturbative prefactor, which factors out cleanly, and no residual non-saddle contributions affect the resurgent structure at these orders. revision: yes
Referee: The reduction from the infinite-dimensional fluctuation operator (Lamé operator) to the finite-dimensional contour integral must be shown to capture all relevant Stokes phenomena. The paper should provide a concrete check that the integration cycle is fixed internally by the geometry of the exact saddles and does not require external input from Exact WKB to determine the correct combination of saddles.

Authors: We have expanded the relevant section to make the reduction explicit. The integration cycle is constructed from the Picard-Lefschetz thimbles attached to the exact Weierstrass saddles; the steepest-descent paths in quasi-zero-mode space are fixed by the saddle geometry and the associated Picard-Fuchs periods. We now include a concrete check for the two-instanton sector demonstrating that the resulting Stokes jumps and cancellations are obtained internally, without reference to Exact WKB data. The noted overlap with Exact WKB remains complementary rather than foundational to the cycle choice. revision: yes

Circularity Check

0 steps flagged

No significant circularity: derivation relies on independent standard mathematics

full rationale

The paper constructs exact finite-T saddles via Weierstrass elliptic functions, treats fluctuations with Lamé operators, and extracts periods via Picard-Fuchs equations to form a Picard-Lefschetz contour integral over quasi-zero modes. These are established mathematical objects whose properties (solutions to the EOM, determinant formulas, period relations) are invoked as external facts, not defined or fitted within the paper. The claim that this integral encodes the full resurgent structure (saddle contributions, asymptotics, ambiguity cancellation) at each instanton order is presented as a consequence of the geometry of the contour, without reducing any output to a self-defined input, a fitted parameter renamed as prediction, or a load-bearing self-citation chain. No uniqueness theorem is imported from the authors' prior work, and the framework is explicitly noted to overlap with but not depend on Exact WKB. The derivation chain therefore remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The framework rests on the known analytic properties of Weierstrass elliptic functions and the Picard-Fuchs equations for the periods of the Lamé operator; these are standard mathematical tools and introduce no new free parameters or invented entities.

axioms (2)

standard math Weierstrass elliptic functions and Lamé operators admit the required analytic continuation and period relations for the double-well saddles.
Invoked when constructing the exact saddles and fluctuation operators.
domain assumption The Picard-Lefschetz theory applies directly to the finite-dimensional integral over quasi-zero modes.
Used to select contributing saddles and cancel ambiguities.

pith-pipeline@v0.9.0 · 5490 in / 1458 out tokens · 45710 ms · 2026-05-10T12:33:31.480813+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

32 extracted references · 16 canonical work pages · 1 internal anchor

[1]

From the path integral, we can just directly read off the leadingT dependence in the exact action

(32) This splitting agrees with the result from Exact WKB, which requires either resumming the Voros symbol to all orders in the Riccati/WKB expansion [3, 5, 8, 22] or us- ing a connection-formula approach [6, 14, 15]. From the path integral, we can just directly read off the leadingT dependence in the exact action. Recall that in the DIG path integral [1...
[2]

C. M. Bender and T. T. Wu, Anharmonic oscillator, Phys. Rev.184, 1231 (1969)

1969
[3]

Brezin, G

E. Brezin, G. Parisi, and J. Zinn-Justin, Perturbation Theory at Large Orders for Potential with Degenerate Minima, Phys. Rev. D16, 408 (1977)

1977
[4]

Voros, The return of the quartic oscillator

A. Voros, The return of the quartic oscillator. The complex WKB method, Annales de l’I.H.P. Physique th´ eorique39, 211 (1983)

1983
[5]

Dillinger, E

H. Dillinger, E. Delabaere, and F. Pham, R´ esurgence de Voros et p´ eriodes des courbes hyperelliptiques, Ann. Inst. Fourier43, 163 (1993)

1993
[6]

Delabaere and F

E. Delabaere and F. Pham, Resurgent methods in semi- classical asymptotics, Ann. Inst. Henri Poincar´ e Phys. Th´ eor.71, 1 (1999)

1999
[7]

´Alvarez, Langer–Cherry derivation of the multi- instanton expansion for the symmetric double well, J

G. ´Alvarez, Langer–Cherry derivation of the multi- instanton expansion for the symmetric double well, J. Math. Phys.45, 3095 (2004)

2004
[8]

G. V. Dunne and M. ¨Unsal, Generating nonperturba- tive physics from perturbation theory, Phys. Rev. D89, 041701 (2014)

2014
[9]

Bucciotti, T

B. Bucciotti, T. Reis, and M. Serone, An anharmonic alliance: exact WKB meets EPT, JHEP2023(11), 124, arXiv:2309.02505 [hep-th]

work page arXiv
[10]

Basar, G

G. Basar, G. V. Dunne, and M. Unsal, Quantum Ge- ometry of Resurgent Perturbative/Nonperturbative Re- lations, JHEP2017(05), 087, arXiv:1701.06572 [hep-th]

work page arXiv
[11]

Gu and M

J. Gu and M. Marino, Exact multi-instantons in topo- logical string theory, SciPost Phys.15, 179 (2023), arXiv:2211.01403 [hep-th]

work page arXiv 2023
[12]

van Spaendonck and M

A. van Spaendonck and M. Vonk, Exact instanton transseries for quantum mechanics, SciPost Phys.16, 103 (2024), arXiv:2309.05700 [hep-th]

work page arXiv 2024
[13]

E. B. Bogomolny, Calculation of instanton–anti- instanton contributions in quantum mechanics, Phys. Lett. B91, 431 (1980)

1980
[14]

Zinn-Justin, Multi - Instanton Contributions in Quan- tum Mechanics, Nucl

J. Zinn-Justin, Multi - Instanton Contributions in Quan- tum Mechanics, Nucl. Phys. B192, 125 (1981)

1981
[15]

Zinn-Justin and U

J. Zinn-Justin and U. D. Jentschura, Multi-instantons and exact results I: Conjectures, WKB expansions, and instanton interactions, Annals Phys.313, 197 (2004), arXiv:quant-ph/0501136

work page arXiv 2004
[16]

Zinn-Justin and U

J. Zinn-Justin and U. D. Jentschura, Multi-instantons and exact results II: Specific cases, higher-order ef- fects, and numerical calculations, Annals Phys.313, 269 (2004), arXiv:quant-ph/0501137

work page arXiv 2004
[17]

J. L. Richard and A. Rouet, Complex Saddle Points in the Double Well Oscillator, Nucl. Phys. B183, 251 (1981)

1981
[18]

J. L. Richard and A. Rouet, Complex Saddle Points Ver- sus Dilute Gas Approximation in the Double Well An- harmonic Oscillator, Nucl. Phys. B185, 47 (1981)

1981
[19]

Tanizaki and T

Y. Tanizaki and T. Koike, Real-time Feynman path in- tegral with Picard–Lefschetz theory and its applications to quantum tunneling, Annals Phys.351, 250 (2014), arXiv:1406.2386 [math-ph]

work page arXiv 2014
[20]

Nekrasov, Tying up instantons with anti-instantons., inLudwig Faddeev Memorial Volume(2018) pp

N. Nekrasov, Tying up instantons with anti-instantons., inLudwig Faddeev Memorial Volume(2018) pp. 351–388, arXiv:1802.04202 [hep-th]

work page arXiv 2018
[21]

Behtash, G

A. Behtash, G. V. Dunne, T. Schaefer, T. Sulejman- pasic, and M. ¨Unsal, Critical Points at Infinity, Non- Gaussian Saddles, and Bions, JHEP2018(06), 068, arXiv:1803.11533 [hep-th]

work page arXiv
[22]

Dorigoni,An Introduction to Resurgence, Trans-Series and Alien Calculus, arXiv:1411.3585 [hep-th], 2014; published in Ann

D. Dorigoni, An Introduction to Resurgence, Trans- Series and Alien Calculus, Annals Phys.409, 167914 (2019), arXiv:1411.3585 [hep-th]

work page arXiv 2019
[23]

Sueishi, S

N. Sueishi, S. Kamata, T. Misumi, and M. ¨Unsal, On exact-WKB analysis, resurgent structure, and quantiza- tion conditions, JHEP2020(12), 114, arXiv:2008.00379 [hep-th]

work page arXiv 2008
[24]

Coleman,Aspects of Symmetry: Selected Erice Lec- tures(Cambridge University Press, Cambridge, U.K., 1985)

S. Coleman,Aspects of Symmetry: Selected Erice Lec- tures(Cambridge University Press, Cambridge, U.K., 1985)

1985
[25]

’t Hooft, Computation of the Quantum Effects Due to a Four-Dimensional Pseudoparticle, Phys

G. ’t Hooft, Computation of the Quantum Effects Due to a Four-Dimensional Pseudoparticle, Phys. Rev. D14, 3432 (1976), [Erratum: Phys.Rev.D 18, 2199 (1978)]

1976
[26]

’t Hooft, Can We Make Sense Out of Quantum Chro- modynamics?, Subnucl

G. ’t Hooft, Can We Make Sense Out of Quantum Chro- modynamics?, Subnucl. Ser.15, 943 (1979)

1979
[27]

Renormalons

M. Beneke, Renormalons, Phys. Rept.317, 1 (1999), arXiv:hep-ph/9807443

work page Pith review arXiv 1999
[28]

Renormalons as Saddle Points

A. Bhattacharya, J. Cotler, A. Dersy, and M. D. Schwartz, Renormalons as Saddle Points, arXiv:2410.07351 [hep-th] (2024)

work page internal anchor Pith review Pith/arXiv arXiv 2024
[29]

I. I. Balitsky and A. V. Yung, Instanton Molecular Vac- uum inN= 1 Supersymmetric Quantum Mechanics, Nucl. Phys. B274, 475 (1986)

1986
[30]

T. C. Kraan and P. van Baal, Periodic instantons with nontrivial holonomy, Nucl. Phys. B533, 627 (1998), arXiv:hep-th/9805168

work page arXiv 1998
[31]

Lee and C.-h

K.-M. Lee and C.-h. Lu, SU(2) calorons and magnetic monopoles, Phys. Rev. D58, 025011 (1998), arXiv:hep- th/9802108

work page arXiv 1998
[32]

P. C. Argyres and M. Unsal, The semi-classical expansion and resurgence in gauge theories: new perturbative, in- stanton, bion, and renormalon effects, JHEP2012(08), 063, arXiv:1206.1890 [hep-th]

work page arXiv