pith. machine review for the scientific record. sign in

arxiv: 2604.25792 · v1 · submitted 2026-04-28 · ⚛️ nucl-th · hep-ph

Recognition: unknown

Exact emulation of few-body systems at low cost

Sven Heihoff , Arseniy A. Filin , Evgeny Epelbaum
Authors on Pith no claims yet

Pith reviewed 2026-05-07 14:11 UTC · model grok-4.3

classification ⚛️ nucl-th hep-ph
keywords few-body systemslow-rank updatesHamiltonian emulationnuclear effective field theoryscattering statesbound statesemulators
0
0 comments X

The pith

A parametric low-rank update to any Hamiltonian reduces the A-body problem at fixed energy to a small matrix equation.

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

The paper proves that when a Hamiltonian receives an update that is both low-rank and parametric, the full A-body Schrödinger problem at one fixed energy becomes exactly equivalent to solving a matrix equation whose dimension equals only the rank of that update. This holds no matter how large the underlying Hilbert space, so the reduction is exact rather than approximate. The authors demonstrate the result by constructing emulators for few-body scattering and bound states that remain accurate for any parameter value, even far from the points used to build them. The motivation is the high cost of solving A-body problems repeatedly when fitting low-energy constants in nuclear effective field theories, especially those involving three-body forces.

Core claim

For a Hamiltonian of the form H0 plus a parametric low-rank update, the A-body problem at fixed energy exactly reduces to a low-dimensional matrix equation whose size is independent of the Hilbert-space dimension. Snapshot-based emulators built from this reduction solve few-body scattering and bound states exactly and cheaply, remain precise outside the training region, and work for parameter values inaccessible to direct solution methods.

What carries the argument

The parametric low-rank update of the Hamiltonian, which converts the resolvent or scattering equation at fixed energy into an algebraic matrix problem via a finite-dimensional inversion.

If this is right

  • Snapshot-based emulators can be applied at any parameter value without loss of precision.
  • The reduction works for scattering states, bound states, and arbitrary numbers of particles.
  • The method applies independently of how the snapshots are generated and to any interaction type.
  • Computational cost for repeated A-body solves in nuclear physics drops dramatically for three-body forces and similar terms.

Where Pith is reading between the lines

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

  • Parameter scans needed to fit low-energy constants in effective field theories become feasible for forces previously too expensive to treat repeatedly.
  • The same algebraic structure may extend to other observables or to problems in atomic and molecular physics where Hamiltonians admit low-rank corrections.
  • If the low-rank property holds only approximately, controlled error bounds could still be derived from the same matrix equation.

Load-bearing premise

The added piece of the Hamiltonian must be exactly low-rank and depend on parameters while the energy is held fixed during the reduction.

What would settle it

A direct numerical comparison in which a non-low-rank perturbation is introduced and the reduced matrix equation no longer reproduces the exact eigenvalues or phase shifts of the original Hamiltonian to machine precision.

Figures

Figures reproduced from arXiv: 2604.25792 by Arseniy A. Filin, Evgeny Epelbaum, Sven Heihoff.

Figure 1
Figure 1. Figure 1: SVD of solution-vectors FIG1SVD of 11 solutionvect FIG1SVD of 11 solutionvectors FIG. 1. SVD of 11 solution-vector equation (8) at fixed e equation (8) at fixed ener equation (8) at a fixed fi (C FIG1SVD f 11 l view at source ↗
Figure 2
Figure 2. Figure 2: For the emulated results, we observe ↵ect the accuracy of the solution and the emulated the accuracy of the solution and the emulated with rank-1 update VE. Next, we consider two-para with rank-1 update VE. Next, we consider two-parame receives parametric low-rank updates. First, we co ilil f hfV3N V receives parametric low-rank updates. First, we consi ll f hfV3N V T view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. Numerical error Eq. ( Figure 3. Numerical accuracy solutions are obtained by perfo view at source ↗
read the original abstract

Effective field theories have established themselves as key pillars of modern nuclear physics. They enable a quantitative understanding of the strong nuclear force, provided low-energy constants that parametrize short-distance physics can be determined from experimental data. This, however, often becomes prohibitively expensive due to a significant computational cost of solving the A-body problem. The computational challenge is particularly severe for three-body forces, which are at the frontier of nuclear and atomic physics and play an important role in the equation of state of neutron stars. Here we prove that for a parametric low-rank update of a Hamiltonian, the A-body problem at a fixed energy exactly reduces to a low-dimensional matrix equation regardless of the size of the Hilbert space. As a proof-of-principle, we present exact and computationally cheap snapshot-based emulators for few-body scattering and bound states. Unlike alternatives, our emulators can be used far away from the snapshot region without loss of precision and yield accurate results for parameter values not accessible using conventional solution techniques. Our approach is not restricted by the interaction type, number of particles, and methods for generating snapshots and can be applied to mitigate the computational burden of the A-body problem to a broad class of problems in nuclear, atomic, and molecular physics.

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.

Referee Report

2 major / 2 minor

Summary. The manuscript claims to prove that a parametric low-rank update to a Hamiltonian allows the A-body Schrödinger equation at fixed energy to reduce exactly to a low-dimensional matrix equation independent of Hilbert-space dimension. It supports this with a proof-of-principle demonstration of snapshot-based emulators for few-body scattering and bound states that remain exact and accurate far from the training region and for parameter values inaccessible by direct solution.

Significance. If the reduction is rigorously established, the result would substantially lower the cost of A-body calculations in nuclear effective field theory, particularly for three-body forces relevant to neutron-star equations of state. The parameter-free, exact character of the reduction together with its applicability across interaction types and snapshot-generation methods constitutes a genuine technical advance.

major comments (2)
  1. [Bound-state reduction] § on bound states (around the nonlinear eigenvalue problem): the fixed-energy reduction is immediate for scattering, but bound-state energies are eigenvalues. The resulting nonlinear eigenvalue problem in the small matrix requires an explicit theorem showing that its roots coincide exactly with the full-space spectrum (no missing eigenvalues and no spurious roots introduced by the low-rank structure or the resolvent).
  2. [Mathematical proof] Proof of the low-rank reduction (central theorem): while the abstract invokes linear-algebra properties of low-rank updates, the derivation should include a self-contained statement of the assumptions on the update operators and a direct verification that the reduced equation reproduces every solution of the original A-body problem at the prescribed E.
minor comments (2)
  1. [Notation] Notation for the low-rank update operators should be introduced once and used consistently; the current alternation between symbols for the update and the resolvent is occasionally confusing.
  2. [Figures] Figure captions should state the A-body system, the rank of the update, and the range of parameters sampled in each panel.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We are grateful to the referee for the positive assessment and the detailed suggestions for improvement. We will revise the manuscript to address the concerns regarding the bound-state reduction and the mathematical proof, as detailed in our point-by-point responses below.

read point-by-point responses

  1. Referee: [Bound-state reduction] § on bound states (around the nonlinear eigenvalue problem): the fixed-energy reduction is immediate for scattering, but bound-state energies are eigenvalues. The resulting nonlinear eigenvalue problem in the small matrix requires an explicit theorem showing that its roots coincide exactly with the full-space spectrum (no missing eigenvalues and no spurious roots introduced by the low-rank structure or the resolvent).

    Authors: We agree with this observation. While our central theorem establishes the equivalence for fixed-energy problems, the bound-state case leads to a nonlinear eigenvalue problem whose equivalence to the original spectrum requires explicit verification. In the revised manuscript, we will add a dedicated theorem (Theorem X) that proves the roots of the reduced nonlinear eigenvalue problem are exactly the bound-state energies of the full A-body Hamiltonian. The proof will show that (i) every full-space eigenvalue is a root of the reduced equation, and (ii) there are no spurious roots introduced by the low-rank update or the resolvent, using the fact that the resolvent is analytic away from the spectrum and the low-rank perturbation preserves the kernel dimension appropriately. This will be supported by a direct algebraic verification. revision: yes

  2. Referee: [Mathematical proof] Proof of the low-rank reduction (central theorem): while the abstract invokes linear-algebra properties of low-rank updates, the derivation should include a self-contained statement of the assumptions on the update operators and a direct verification that the reduced equation reproduces every solution of the original A-body problem at the prescribed E.

    Authors: We thank the referee for highlighting this point. The current derivation relies on standard properties of low-rank updates (Sherman-Morrison-Woodbury formula and rank-nullity theorem), but we acknowledge that a more self-contained presentation would improve clarity. In the revision, we will expand the central theorem to include: (1) a precise statement of the assumptions, namely that the Hamiltonian update is of the form H(λ) = H0 + V(λ) where V(λ) has rank at most r independent of the Hilbert-space dimension, and (2) a direct, step-by-step verification that if ψ is a solution of the full-space equation at energy E, then the projected coefficients satisfy the reduced matrix equation, and conversely, that any solution of the reduced equation can be lifted to a full-space solution. This will make the proof independent of external references to linear-algebra results. revision: yes

Circularity Check

0 steps flagged

No circularity: direct linear-algebra reduction at fixed energy

full rationale

The central claim is a mathematical proof that a parametric low-rank Hamiltonian update reduces the fixed-energy A-body Schrödinger equation to a low-dimensional matrix equation, independent of Hilbert-space dimension. This follows immediately from the algebraic properties of low-rank operators and the resolvent at fixed E; no parameters are fitted to data, no predictions are made by construction from inputs, and no self-citation chain or uniqueness theorem is invoked to close the argument. The resulting nonlinear eigenvalue problem for bound states is handled inside the same reduction without additional assumptions that loop back to the claim itself. The derivation is therefore self-contained and does not reduce to its own inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on standard linear-algebra facts about low-rank matrix updates applied to the Schrödinger equation at fixed energy; no new free parameters or invented entities are introduced in the abstract.

axioms (1)
  • standard math Low-rank matrix updates admit exact low-dimensional reductions in linear systems
    Invoked implicitly when stating that the A-body problem reduces to a low-dimensional matrix equation.

pith-pipeline@v0.9.0 · 5520 in / 1151 out tokens · 53892 ms · 2026-05-07T14:11:41.471200+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

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

  1. Hadronic parity violation: successes, challenges, and future prospects

    nucl-th 2026-05 unverdicted novelty 2.0

    A review of hadronic parity violation covers observed effects, theoretical and experimental progress, and prospects for extending few-nucleon precision studies to nuclei and new physics searches.

Reference graph

Works this paper leans on

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

  1. [1]

    (9) to write them as a sum of rank-1 matrices,V i =P a Vrank-1 i,a

    Perform (randomized) SVD ofV i’s in Eq. (9) to write them as a sum of rank-1 matrices,V i =P a Vrank-1 i,a

  2. [2]

    For rank-1 updates, suchc i always exist and are unique

    For eachV rank-1 i,a , tune the value ofc i to ensure that (H0 +c iVrank-1 i,a )Ψ ¯E i,a = ¯EΨ ¯E i,a and save all Ψ ¯E i,a. For rank-1 updates, suchc i always exist and are unique

  3. [3]

    Build an orthonormal basis from all eigenvectors Ψ ¯E i,a obtained in step 2

  4. [4]

    (9) onto this reduced basis to obtainH rb

    Project the full HamiltonianHfrom Eq. (9) onto this reduced basis to obtainH rb

  5. [5]

    In the reduced basis, find a constraint on the param- etersc i to keep the eigenvalue ofH rb equal to ¯E

  6. [6]

    In the reduced basis, for any⃗ cthat fulfill the con- straint from step 5, compute the eigenvector ofH rb

  7. [7]

    deuteron

    Convert it to the original basis to obtain Ψ ¯E(⃗ c). The resulting emulator allows one to calculate any de- sired bound-state observable for values of⃗ cconsistent with the fixed binding energy ¯Ewithout computational effort. As a proof-of-principle, we describe in Ref. [21] an emulator for the “deuteron” wave function Ψ ¯E(c1, c2, c3) using the already ...

    2020

  8. [8]

    Evgeny Epelbaum, Hans-Werner Hammer, and Ulf-G. Meißner. Modern Theory of Nuclear Forces.Rev. Mod. Phys., 81:1773–1825, 2009.arXiv:0811.1338,doi:10. 1103/RevModPhys.81.1773

    work page Pith review arXiv 2009

  9. [9]

    Three-body forces and Efimov physics in nuclei and atoms.Eur

    Shimpei Endo, Evgeny Epelbaum, Pascal Naidon, Yusuke Nishida, Kimiko Sekiguchi, and Yoshiro Taka- hashi. Three-body forces and Efimov physics in nuclei and atoms.Eur. Phys. J. A, 61(1):9, 2025.arXiv: 2405.09807,doi:10.1140/epja/s10050-024-01467-4

    work page doi:10.1140/epja/s10050-024-01467-4 2025

  10. [10]

    Girlanda, A

    L. Girlanda, A. Kievsky, and M. Viviani. Subleading contributions to the three-nucleon contact interaction. Phys. Rev. C, 84(1):014001, 2011. [Erratum: Phys.Rev.C 102, 019903 (2020)].arXiv:1102.4799,doi:10.1103/ PhysRevC.84.014001

    work page arXiv 2011

  11. [11]

    Subleading D-like Three-Nucleon Interac- tions

    Henri Paul Huesmann, Hermann Krebs, and Evgeny Epelbaum. Subleading D-like Three-Nucleon Interac- tions. 2 2026.arXiv:2602.12879. 6

    work page arXiv 2026

  12. [12]

    Wesolowski, N

    S. Wesolowski, N. Klco, R. J. Furnstahl, D. R. Phillips, and A. Thapaliya. Bayesian parameter estimation for effective field theories.J. Phys. G, 43(7):074001, 2016.arXiv:1511.03618,doi:10.1088/0954-3899/43/ 7/074001

    work page doi:10.1088/0954-3899/43/ 2016

  13. [13]

    Z. H. Sun, A. Ekstr¨ om, C. Forss´ en, G. Hagen, G. R. Jansen, and T. Papenbrock. Multiscale Physics of Atomic Nuclei from First Principles.Phys. Rev. X, 15(1):011028, 2025.arXiv:2404.00058,doi:10.1103/ PhysRevX.15.011028

    work page arXiv 2025

  14. [14]

    Wita la, J

    H. Wita la, J. Golak, and R. Skibi´ nski. Efficient emulator for solving three-nucleon continuum Faddeev equations with chiral three-nucleon force comprising any number of contact terms.Eur. Phys. J. A, 57(7):241, 2021.arXiv: 2103.13237,doi:10.1140/epja/s10050-021-00555-z

    work page doi:10.1140/epja/s10050-021-00555-z 2021

  15. [15]

    Wesolowski, I

    S. Wesolowski, I. Svensson, A. Ekstr¨ om, C. Forss´ en, R. J. Furnstahl, J. A. Melendez, and D. R. Phillips. Rigorous constraints on three-nucleon forces in chiral effective field theory from fast and accurate calculations of few-body observables.Phys. Rev. C, 104(6):064001, 2021.arXiv: 2104.04441,doi:10.1103/PhysRevC.104.064001

    work page doi:10.1103/physrevc.104.064001 2021

  16. [16]

    Training and projecting: A reduced basis method emulator for many-body physics.Phys

    Edgard Bonilla, Pablo Giuliani, Kyle Godbey, and Dean Lee. Training and projecting: A reduced basis method emulator for many-body physics.Phys. Rev. C, 106(5):054322, 2022.arXiv:2203.05284,doi:10.1103/ PhysRevC.106.054322

    work page arXiv 2022

  17. [17]

    A. J. Garcia, C. Drischler, R. J. Furnstahl, J. A. Me- lendez, and Xilin Zhang. Wave-function-based emula- tion for nucleon-nucleon scattering in momentum space. Phys. Rev. C, 107(5):054001, 2023.arXiv:2301.05093, doi:10.1103/PhysRevC.107.054001

    work page doi:10.1103/physrevc.107.054001 2023

  18. [18]

    Armstrong, Pablo Giuliani, Kyle Godbey, Rahul Somasundaram, and Ingo Tews

    Cassandra L. Armstrong, Pablo Giuliani, Kyle Godbey, Rahul Somasundaram, and Ingo Tews. Emulators for Scarce and Noisy Data: Application to Auxiliary-Field Diffusion Monte Carlo for Neutron Matter.Phys. Rev. Lett., 135(14):142501, 2025.arXiv:2502.03680,doi:10. 1103/9928-wyjm

    work page arXiv 2025

  19. [19]

    Alex Gnech, Xilin Zhang, Christian Drischler, R. J. Furn- stahl, Alessandro Grassi, Alejandro Kievsky, Laura E. Marcucci, and Michele Viviani. Emulation of Proton- Deuteron Scattering via the Reduced Basis Method and Active Learning: Detailed Description. 11 2025.arXiv: 2511.10420

    work page arXiv 2025

  20. [20]

    Aaijet al.[LHCb Collaboration]

    Dillon Frame, Rongzheng He, Ilse Ipsen, Daniel Lee, Dean Lee, and Ermal Rrapaj. Eigenvector continuation with subspace learning.Phys. Rev. Lett., 121(3):032501, 2018.arXiv:1711.07090,doi:10.1103/PhysRevLett. 121.032501

    work page doi:10.1103/physrevlett 2018

  21. [21]

    om , author R. J. \ Furnstahl , author S. K\

    Thomas Duguet, Andreas Ekstr¨ om, Richard J. Furn- stahl, Sebastian K¨ onig, and Dean Lee. Colloquium: Eigenvector continuation and projection-based emula- tors.Rev. Mod. Phys., 96(3):031002, 2024.arXiv: 2310.19419,doi:10.1103/RevModPhys.96.031002

    work page doi:10.1103/revmodphys.96.031002 2024

  22. [22]

    Epelbaum, H

    E. Epelbaum, H. Krebs, and U.-G. Meißner. Precision nucleon-nucleon potential at fifth order in the chiral ex- pansion.Phys. Rev. Lett., 115(12):122301, 2015.arXiv: 1412.4623,doi:10.1103/PhysRevLett.115.122301

    work page doi:10.1103/physrevlett.115.122301 2015

  23. [23]

    Reinert, H

    P. Reinert, H. Krebs, and E. Epelbaum. Semilocal momentum-space regularized chiral two-nucleon poten- tials up to fifth order.Eur. Phys. J. A, 54(5):86, 2018. arXiv:1711.08821,doi:10.1140/epja/i2018-12516-4

    work page doi:10.1140/epja/i2018-12516-4 2018

  24. [24]

    D. R. Entem, R. Machleidt, and Y. Nosyk. High-quality two-nucleon potentials up to fifth order of the chiral expansion.Phys. Rev. C, 96(2):024004, 2017.arXiv: 1703.05454,doi:10.1103/PhysRevC.96.024004

    work page doi:10.1103/physrevc.96.024004 2017

  25. [25]

    Ekstr¨ om et al

    A. Ekstr¨ om et al. Optimized Chiral Nucleon-Nucleon Interaction at Next-to-Next-to-Leading Order.Phys. Rev. Lett., 110(19):192502, 2013.arXiv:1303.4674, doi:10.1103/PhysRevLett.110.192502

    work page doi:10.1103/physrevlett.110.192502 2013

  26. [26]

    Ekstr¨ om, G

    A. Ekstr¨ om, G. R. Jansen, K. A. Wendt, G. Hagen, T. Papenbrock, B. D. Carlsson, C. Forss´ en, M. Hjorth- Jensen, P. Navr´ atil, and W. Nazarewicz. Accurate nu- clear radii and binding energies from a chiral interac- tion.Phys. Rev. C, 91(5):051301, 2015. [Erratum: Phys.Rev.C 109, 059901 (2024)].arXiv:1502.04682, doi:10.1103/PhysRevC.109.059901

    work page doi:10.1103/physrevc.109.059901 2015

  27. [27]

    H. V. Henderson and S. R. Searle. On deriving the inverse of a sum of matrices.SIAM Review, 23(1):53–60, 1981. doi:10.1137/1023004

    work page doi:10.1137/1023004 1981

  28. [28]

    See Supplemental Material

  29. [29]

    Wita la, D

    Walter Gl¨ ockle, H. Wita la, D. H¨ uber, H. Kamada, and J. Golak. The Three nucleon continuum: Achievements, challenges and applications.Phys. Rept., 274:107–285, 1996.doi:10.1016/0370-1573(95)00085-2

    work page doi:10.1016/0370-1573(95)00085-2 1996

  30. [30]

    Maris et al

    P. Maris et al. Light nuclei with semilocal momentum- space regularized chiral interactions up to third order. Phys. Rev. C, 103(5):054001, 2021.arXiv:2012.12396, doi:10.1103/PhysRevC.103.054001

    work page doi:10.1103/physrevc.103.054001 2021

  31. [31]

    Self-learning emula- tors and eigenvector continuation.Phys

    Avik Sarkar and Dean Lee. Self-learning emula- tors and eigenvector continuation.Phys. Rev. Res., 4(2):023214, 2022.arXiv:2107.13449,doi:10.1103/ PhysRevResearch.4.023214

    work page arXiv 2022

  32. [32]

    J. M. Maldonado, C. Drischler, R. J. Furnstahl, and P. Mlinari´ c. Greedy emulators for nuclear two-body scattering.Phys. Rev. C, 112(2):024002, 2025.arXiv: 2504.06092,doi:10.1103/k77q-f82l. 7 Supplemental Material This Supplemental Material is organized as follows. We start with providing in Sec. I a detailed derivation of the Lippmann-Schwinger Woodbury...

    work page doi:10.1103/k77q-f82l 2025