Pith. sign in

REVIEW 4 major objections 6 minor 2 cited by

Is attention all you need to solve the correlated electron problem?

T0 review · 4 major / 6 minor · reviewed 2026-08-08 · deepseek-v4-flash

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

Pith's one-line read A neural wavefunction built from self-attention solves the correlated electron problem in a moiré solid with a parameter cost that grows only quadratically with electron number.

desk verdict A solid NN-VMC benchmark on a moiré model with a credible energy study; the N^2 scaling law is too weakly supported to carry the paper's efficiency claim. read the letter →

arxiv 2502.05383 v3 pith:YJPOXVYO submitted 2025-02-07 cond-mat.str-el cond-mat.mes-hallcs.AI

Max Geier , Khachatur Nazaryan , Timothy Zaklama , Liang Fu This is my paper · ORCID
classification cond-mat.str-elcond-mat.mes-hallcs.AI
keywords neuralnetworkwavefunctionsself-attentionvariationalMonteCarlocorrelatedelectronsmoirématerialsWSe2/WS2heterobilayersgeneralizedWignercrystalFermiliquid
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 proposes that a wavefunction built from a self-attention neural network, with no hand-designed correlation terms, can serve as an accurate and efficient variational ansatz for interacting electrons in periodic solids. On a WSe2/WS2 moiré heterobilayer at 2/3 filling, the ansatz yields ground-state energies below band-projected exact diagonalization, even when the diagonalization includes five bands, and below Hartree-Fock from the same network. The central efficiency claim is a numerically observed scaling law: the number of variational parameters needed for convergence saturation grows roughly as $N^{2}$, specifically N*_par = (400 ± 49) × N^(2.01 ± 0.05). If this scaling holds at larger sizes, the method points toward practical simulations of strongly correlated solids beyond the reach of exact methods.

What carries the argument

The central object is the self-attention layer acting on per-electron feature streams inside a neural-network wavefunction. Each electron stream is embedded through periodic coordinates (sines and cosines of reciprocal supercell vectors), then transformed by multi-head attention: learned keys, queries, and values compute weighted sums exp(q_j · k_i) v_j over all electrons, so each electron's orbital becomes a permutation-equivariant function of the full configuration. These correlated orbitals are assembled into a small number of generalized Slater determinants, and the whole network is optimized with variational Monte Carlo using natural gradient descent. This construction replaces hand-built backflow and Jastrow terms with a learned, parameter-rich but symmetry-respecting correlation mechanism; the paper's scaling law counts the parameters of this network at the saturation point.

What would settle it

Re-extract N*_par on the same 9-, 12-, 27-, and 36-site systems using a stricter definition of saturation, such as requiring the energy to lie within a small fraction of the statistical error or extrapolating the energy-versus-parameter curve to zero slope, and check whether the exponent remains near 2. A decisive test is to run the ansatz on a larger system, such as 48 or 64 electrons at the same filling, and see whether the parameter count needed for a fixed energy accuracy still follows the fitted curve; a significantly larger exponent or a failure to lower the energy would refute the scaling claim.

Watch

Extended reading notes

Core claim

The authors claim that the self-attention mechanism alone—letting every electron's orbital depend on the positions of all other electrons—is enough to capture electron correlation in a periodic solid without pretraining, envelope functions, or Jastrow factors. The resulting generalized Slater determinant wavefunction is optimized by variational Monte Carlo. For the moiré system studied, it produces energies lower than band-projected exact diagonalization for all benchmarked system sizes and interaction strengths, with the improvement over single-band exact diagonalization reaching about 2.5% for 18 electrons; it also reproduces both the Fermi liquid phase at weak interaction and the generalized Wigner crystal at strong interaction. The paper's quantitative efficiency claim is the empirical scaling law, Eq. (32), for the parameters required to reach convergence saturation.

Load-bearing premise

The quadratic scaling law rests on the paper's operational definition of convergence saturation—the parameter count beyond which converged energies stay within one standard deviation of the lowest observed energy; if that threshold is too generous, the fitted $N^{2}$ law could reflect the criterion rather than an intrinsic property of the ansatz.

Editorial extensions

If this is right

  • Energies below band-projected exact diagonalization are obtained even when the diagonalization includes five bands, indicating that band-truncation error is substantial at realistic moiré parameters.
  • The parameter count for convergence saturation grows as N^2, giving a practical guideline for choosing network size and suggesting better scaling than tensor-network approaches whose parameter count grows roughly as e^{sqrt(N)}.
  • Without pretraining, envelope functions, or a Jastrow factor, the ansatz captures both the Fermi liquid at epsilon=10 and the generalized Wigner crystal at epsilon=5, including the density and correlation signatures of each phase.
  • Band-projected exact diagonalization shows a first-order metal-insulator transition near epsilon^{-1} = 0.11, and the self-attention results are consistent with that transition.
  • A similar self-attention ansatz has already been shown to describe fractional quantum Hall ground states, supporting the hope of a unified variational wavefunction for distinct correlated phases.

Reading between the lines

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

  • If the N^2 scaling persists at larger electron numbers, self-attention variational Monte Carlo could reach systems far beyond exact diagonalization; the prefactor of about 400 parameters per electron squared still implies substantial compute, so the practical bottleneck will be the constant rather than the exponent.
  • Because the ansatz works without a Jastrow factor, it implicitly suggests that attention layers learn the short-range cusp behavior; a direct test would be measuring the local-energy variance at small electron-electron separation and comparing it with a cusp-corrected Jastrow variant.
  • A sharp test of generality is to apply the same randomly initialized self-attention ansatz to doped fillings, spinful systems, or frustrated lattices; if the quadratic parameter scaling and accuracy survive those changes, the architecture would move closer to a unifying fermionic solver.
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.

Referee Report

4 major / 6 minor

Summary. The paper presents a neural-network variational Monte Carlo method for periodic correlated electrons in moiré TMD heterobilayers, using a self-attention wavefunction ansatz adapted from Psiformer. It benchmarks ground-state energies against SlaterNet and band-projected exact diagonalization for 6- and 18-electron systems at ν=2/3 filling and ε=10 and ε=5, finding lower energies than both references. It also introduces a saturation analysis to estimate the minimal number of variational parameters N* required for energy convergence at four system sizes (9, 12, 27, and 36 sites) and fits a power law N* ≈ (400±49)·N^(2.01±0.05). The paper additionally reports density and pair-correlation signatures of a Fermi liquid and a generalized Wigner crystal, with a supporting BP-ED finite-size gap study.

Significance. If the main claims are established, the paper would provide useful evidence that a self-attention ansatz can describe a realistic strongly correlated solid-state model with low human bias and a favorable parameter-count scaling. The energy benchmark is the strongest part: the self-attention ansatz systematically lowers the energy relative to SlaterNet and BP-ED for the studied moiré systems, and the Fermi-liquid-to-Wigner-crystal density and correlation signatures are consistent with the BP-ED gap analysis. The scaling-law claim is a promising hypothesis but is not yet supported by the presented data, and the paper would be substantially strengthened by additional system sizes and a well-defined saturation protocol. The manuscript is transparent about some limitations, including the caveat that Eq. (32) may not directly generalize and that a minimum number of attention layers and heads is required.

major comments (4)
  1. [Sec. V.A, Eq. (32)] The central efficiency claim is not established by the presented data. The saturation point N* is defined by a one-standard-deviation threshold on the batch-mean local energy, which measures Monte Carlo sampling noise of a single optimization run rather than systematic convergence in architecture or optimization; the 'lowest observed energy' is a single sample. Since architectures are scanned on a discrete grid (Fig. 8), each N* is effectively an upper bound at a discrete point with no propagated uncertainty, yet the fit in Fig. 4 treats the four N* values as exact. With four points at one filling (ν=2/3) and one dielectric constant (ε=10), the reported exponent error ±0.05 is a fit residual, not a statistical statement about the scaling. A stricter or multi-seed saturation criterion, additional system sizes, and out-of-sample tests at other fillings and dielectric constants are needed before Eq. (32) can support the abstract's scaling claim.
  2. [Sec. V.A and text after Eq. (32)] The manuscript itself notes that a minimum number of attention heads and layers (roughly 3) is required and that increasing the parameter count without meeting these minima does not improve performance. This means the raw parameter count is not by itself the relevant complexity measure; the scaling law is conditional on an architecture search in (nlayers, nheads, d_attn, d_perc). The main text also states that architectural parameters must be adjusted for each system, and Appendix C confirms that the listed hyperparameters are not fixed across the study. The extracted N* values therefore conflate ansatz capability with the authors' convergence and architecture choices, and a sensitivity analysis with respect to the architecture grid and the saturation tolerance is required.
  3. [Sec. V.B and Fig. 5] The accuracy claim is supported only by comparison with band-truncated BP-ED, which is itself not converged in the number of bands; the text says the BP-ED energy approaches but remains higher than the self-attention NN as bands are increased. For the 18-electron system, BP-ED is limited to a single band, so the fact that the NN energy is lower is expected if band mixing is strong. To substantiate 'accurate' quantitatively, the authors should benchmark against a converged reference for at least the 6-electron system, for example ED with enough bands or fixed-node diffusion Monte Carlo, or provide evidence of convergence of the NN energy with respect to architecture and optimization seeds that is independent of the BP-ED comparison.
  4. [Abstract, Sec. V.A, Sec. VII] The efficiency claim is stated in terms of the number of variational parameters, but the practical computational cost of the method also includes the O(N^2 d) self-attention evaluation per layer per Monte Carlo sample, the cost of the KFAC optimizer, and Monte Carlo sampling autocorrelation. The demonstrated N_par ≈ N^2 scaling does not by itself establish that large-scale simulations are efficient in wall-clock or memory cost. The authors should either report wall-clock or floating-point-operation scalings, or explicitly restrict the claim to parameter count rather than overall efficiency.
minor comments (6)
  1. [Sec. II.B, Eq. (4)] The Madelung term in Eq. (4) is typeset confusingly as NX_i ξ_M; this should be written as (N/2)ξ_M or equivalent, with ξ_M defined consistently with Appendix A.
  2. [Sec. III.C, Eq. (18)] The normalization factor N in Eq. (18) conflicts with the use of N for the number of electrons elsewhere in the paper; please rename the normalization constant, for example to N_norm.
  3. [Fig. 3(b) caption] The caption contains a typo, '8edata', which should read '8e data'; additionally, the shifted datasets should be labeled with both the number of electrons and the corresponding supercell size (6e/9-site, 8e/12-site, 18e/27-site, 24e/36-site) for clarity.
  4. [Sec. V.A, Fig. 3(a)] The text says learning curves are shown for the '9- and 27-site systems' while the caption says '6e' and '18e'; this site/electron nomenclature should be made consistent.
  5. [Appendix C, Table II] The entry 'Training iterations 15e4' is ambiguous and should be written as 1.5×10^5; the learning-rate schedule η_0(1+t/t_0)^{-1} would also benefit from a definition of t in units of optimization steps.
  6. [References] Reference [63] contains raw LaTeX macro text in the bibliographic entry and should be cleaned; several other references would benefit from consistent formatting of preprint identifiers.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the variational energy benchmark and the N^2 scaling law rest on independent numerical evidence rather than on the paper's own assumptions.

full rationale

The paper's central numerical claims are self-contained. The self-attention wavefunction is a variational ansatz whose energy is evaluated by Monte Carlo and minimized by natural-gradient descent, and the benchmarks are band-projected exact diagonalization and SlaterNet, both computed in the paper from the same Hamiltonian; the NN energies are compared as variational upper bounds. This is an external benchmark, not a re-import of the conclusion. The scaling law in Eq. (32) is an empirical power-law fit to saturation points N*_par extracted in Sec. V.A from convergence curves at four system sizes. The definition of N*_par as the threshold "beyond which the converged energies consistently fall within one standard deviation of the lowest observed energy" is a methodological choice; it is neither defined in terms of N^2 nor fitted to the final exponent. A different threshold would change the fitted values, but this is a robustness concern, not a circular reduction. The paper itself cautions that the scaling "may not directly generalize," and it leaves detailed dimension-wise scaling for future work. Self-citations to prior Psiformer work, Ref. [27] on fractional quantum Hall states, and the authors' earlier moiré parameter papers are used for architectural inspiration or supplementary validation; none is used as a load-bearing uniqueness theorem or as the evidence for the present energy benchmarks. No equation in the manuscript reduces to an input by construction, and no fitted parameter is renamed as a prediction. I therefore find no significant circularity.

Assumptions & free parameters 4 free parameters · 6 assumptions · 0 invented entities

No new physical entities are introduced; self-attention is an architecture, not a postulated physical object. The central claim depends on the fitted scaling law parameters, the chosen saturation criterion, and several domain assumptions about the moiré model and optimization.

free parameters (4)
  • Scaling exponent alpha = 2.01 ± 0.05
    Power-law fit N*_par = C·N^alpha to saturation points in Fig. 4; the exponent is extracted from the paper's own in-sample data, not from theory or external benchmarks.
  • Scaling prefactor C = (4.00 ± 0.49) × 10^2
    Same fit, Eq. (32); determines the parameter-count estimate and therefore the efficiency claim.
  • Convergence saturation tolerance = 1 standard deviation of the lowest observed energy
    N*_par is defined as the threshold beyond which converged energies stay within one standard deviation of the lowest observed energy (Sec. V.A). This threshold is chosen post hoc and directly controls the extracted N* values and the scaling law.
  • Per-system network architecture dimensions = varied per system, e.g. (3,6,32,64), (6,6,16,128)
    Appendix C notes architectural parameters must be adjusted per system size, and Sec. V.A requires a minimum of roughly 3 attention heads and layers. These choices constrain the regime in which the scaling law applies.
assumptions (6)
  • standard math Generalized Slater determinants built from permutation-equivariant orbitals can represent fermionic ground states
    Sec. III.C, Eq. (23); relies on the known result that any fermionic wavefunction can be written as a generalized Slater determinant if discontinuities are allowed, and assumes the continuous neural network ansatz is sufficiently expressive.
  • domain assumption Continuum moiré model parameters from first-principles calculations are accurate for WSe2/WS2
    Sec. V: m* = 0.35 m_e, V0 = 15 meV, phi = pi/4, aM = 8.031 nm, epsilon ~= 5; imported from Ref. [55]. The numerical results are only as good as this model.
  • domain assumption KFAC optimization with random initialization converges to the ground state rather than a local minimum
    Sec. IV.B approximates the inverse quantum geometric tensor with KFAC and neglects the phase; Sec. VI claims random initialization suffices. There is no proof that the nonconvex optimization reaches the global minimum.
  • ad hoc to paper The empirical scaling law extracted at nu=2/3, epsilon=10, spin-polarized systems with up to 24 electrons generalizes to other fillings and interaction strengths
    Sec. V.A states the law is derived for the present system and architecture and may not directly generalize, yet the abstract presents N^2 scaling as a path to efficient large-scale simulations.
  • ad hoc to paper The saturation criterion based on one standard deviation of batch-mean energies is a valid measure of convergence
    Sec. V.A and Fig. 3(b); the criterion is not an independent convergence test and can be affected by batch noise and local minima.
  • standard math Ewald summation with a neutralizing background correctly evaluates the periodic Coulomb interaction
    Appendix A; standard technique, but the comparison between real-space VMC and reciprocal-space ED requires consistent handling of the q=0 term and the Madelung energy.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Is attention all you need to solve the correlated electron problem?." pith.science (2026). https://pith.science/paper/YJPOXVYO

@misc{pith2026250205383,
  author       = {Pith},
  title        = {Pith review of: Is attention all you need to solve the correlated electron problem?},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YJPOXVYO}},
  note         = {Machine review of arXiv:2502.05383}
}
abstract

The attention mechanism has transformed artificial intelligence research by its ability to learn relations between objects. In this work, we explore how a many-body wavefunction ansatz constructed from a large-parameter self-attention neural network can be used to solve the interacting electron problem in solids. By a systematic neural-network variational Monte Carlo study on a moir\'e quantum material, we demonstrate that the self-attention ansatz provides an accurate and efficient solution without human bias. Moreover, our numerical study finds that the required number of variational parameters scales roughly as $N^2$ with the number of electrons, which opens a path towards efficient large-scale simulations.

Figures

Figures reproduced from arXiv: 2502.05383 by the authors.

Figure 1
Figure 1. shows schematically and compares the network architectures representing general single-Slater determi￾nant wavefunctions, and its extension by self-attention capturing correlations. A. Deep neural network for one-body orbitals A single particle orbital describes the motion of a sin￾gle electron in an external potential. Mathematically, it [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2 [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 4
Figure 4. FIG. 4. Scaling law for the number of variational parame [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figures from the paper (9 more)
Figure 3
Figure 3. Figure 3: FIG. 3. (a) Raw (unsmoothed) learning curves for the energy [PITH_FULL_IMAGE:figures/full_fig_p008_3.png]
Figure 5
Figure 5. Figure 5: FIG. 5. Comparison of ground state energies obtained from [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]
Figure 6
Figure 6. Figure 6: FIG. 6. Converged ground state charge density [(a), (b)] and [PITH_FULL_IMAGE:figures/full_fig_p010_6.png]
Figure 7
Figure 7. Figure 7: FIG. 7. Comparison of training curves for 27 sites with 2/3 [PITH_FULL_IMAGE:figures/full_fig_p013_7.png]
Figure 8
Figure 8. Figure 8: FIG. 8. Converged energy of the moir´e system at [PITH_FULL_IMAGE:figures/full_fig_p014_8.png]
Figure 9
Figure 9. Figure 9: FIG. 9. Correlation energy per electron, [PITH_FULL_IMAGE:figures/full_fig_p015_9.png]
Figure 10
Figure 10. Figure 10: FIG. 10. Many Body Spectrum (MBS) [(a),(b)] and ground [PITH_FULL_IMAGE:figures/full_fig_p015_10.png]
Figure 11
Figure 11. Figure 11: FIG. 11. Many Body Energy Gap ( [PITH_FULL_IMAGE:figures/full_fig_p015_11.png]
Figure 12
Figure 12. Figure 12: FIG. 12 [PITH_FULL_IMAGE:figures/full_fig_p016_12.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

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

  1. Accurate Self-Attention Wavefunctions at Large Scale

    cond-mat.str-el 2026-07 conditional novelty 6.0 of 10

    Self-attention variational wavefunctions for the 2D homogeneous electron gas up to N=169 yield energies below DMC and a converged collective-mode dispersion including a roton-like minimum.

  2. Comparing Symmetrized Determinant Neural Quantum States for the Hubbard Model

    cond-mat.str-el 2025-10 conditional novelty 6.0 of 10

    For the doped square-lattice Hubbard model, hidden-fermion and backflow neural quantum states with a Vision Transformer backbone reach nearly equal variational energies; translation-equivariant attention is outperform...

Reference graph

Works this paper leans on

75 extracted references · 41 canonical work pages · cited by 2 Pith papers

  1. [11]

    von Glehn, J

    I. von Glehn, J. S. Spencer, and D. Pfau, A Self- Attention Ansatz for Ab-initio Quantum Chemistry, arXiv 10.48550/arXiv.2211.13672 (2022), 2211.13672

    arXiv doi:10.48550/arxiv.2211.13672 2022

  2. [1]

    D. R. Hartree, The Wave Mechanics of an Atom with a Non-Coulomb Central Field. Part II. Some Results and Discussion, Math. Proc. Cambridge Philos. Soc.24, 111 (1928)

    work page 1928

  3. [2]

    Fock, N¨ aherungsmethode zur L¨ osung des quanten- mechanischen Mehrk¨ orperproblems, Z

    V. Fock, N¨ aherungsmethode zur L¨ osung des quanten- mechanischen Mehrk¨ orperproblems, Z. Phys.61, 126 (1930)

    work page 1930

  4. [3]

    W. M. C. Foulkes, L. Mitas, R. J. Needs, and G. Ra- jagopal, Quantum Monte Carlo simulations of solids, Rev. Mod. Phys.73, 33 (2001)

    2001

  5. [4]

    Carleo and M

    G. Carleo and M. Troyer, Solving the quantum many- body problem with artificial neural networks, Science 355, 602 (2017)

    2017

  6. [5]

    K. T. Sch¨ utt, P.-J. Kindermans, H. E. Sauceda, S. Chmiela, A. Tkatchenko, and K.-R. M¨ uller, SchNet: a continuous-filter convolutional neural network for model- ing quantum interactions, inGuide Proceedings(Curran Associates Inc., 2017) pp. 992–1002

    work page 2017

  7. [6]

    Carleo, K

    G. Carleo, K. Choo, D. Hofmann, J. E. T. Smith, T. Westerhout, F. Alet, E. J. Davis, S. Efthymiou, I. Glasser, S.-H. Lin, M. Mauri, G. Mazzola, C. B. Mendl, E. van Nieuwenburg, O. O’Reilly, H. Th´ eveniaut, G. Tor- lai, F. Vicentini, and A. Wietek, NetKet: A machine learning toolkit for many-body quantum systems, Soft- wareX10, 100311 (2019)

    work page 2019

  8. [8]

    K. Choo, A. Mezzacapo, and G. Carleo, Fermionic neural-network states for ab-initio electronic structure, Nat. Commun.11, 1 (2020)

    work page 2020

Show all 75 references
  1. [9]

    Hermann, Z

    J. Hermann, Z. Sch¨ atzle, and F. No´ e, Deep-neural- network solution of the electronic Schr¨ odinger equation, Nat. Chem.12, 891 (2020)

    2020

  2. [10]

    D. Pfau, J. S. Spencer, A. G. D. G. Matthews, and W. M. C. Foulkes, Ab initio solution of the many-electron Schr\”odinger equation with deep neural networks, Phys. Rev. Res.2, 033429 (2020)

    2020

  3. [12]

    L. L. Viteritti, R. Rende, and F. Becca, Transformer vari- ational wave functions for frustrated quantum spin sys- tems, Phys. Rev. Lett.130, 236401 (2023)

    2023

  4. [13]

    R. Li, H. Ye, D. Jiang, X. Wen, C. Wang, Z. Li, X. Li, D. He, J. Chen, W. Ren, and L. Wang, For- ward Laplacian: A New Computational Framework for Neural Network-based Variational Monte Carlo, arXiv 10.48550/arXiv.2307.08214 (2023), 2307.08214

    2023 doi arXiv

  5. [14]

    Gao and S

    N. Gao and S. G¨ unnemann, Generalizing Neural Wave Functions, inInternational Conference on Machine Learning(PMLR, 2023) pp. 10708–10726

    2023

  6. [15]

    J. R. Moreno, G. Carleo, A. Georges, and J. Stokes, Fermionic wave functions from neural-network con- strained hidden states, Proceedings of the National Academy of Sciences119, e2122059119 (2022), https://www.pnas.org/doi/pdf/10.1073/pnas.2122059119. 17

    2022 doi

  7. [16]

    D. Luo, Z. Chen, K. Hu, Z. Zhao, V. M. Hur, and B. K. Clark, Gauge-invariant and anyonic-symmetric au- toregressive neural network for quantum lattice models, Phys. Rev. Res.5, 013216 (2023)

    2023

  8. [17]

    Chen and M

    A. Chen and M. Heyl, Empowering deep neural quantum states through efficient optimization, Nature Physics20, 1476 (2024)

    2024

  9. [18]

    Cassella, H

    G. Cassella, H. Sutterud, S. Azadi, N. D. Drummond, D. Pfau, J. S. Spencer, and W. M. C. Foulkes, Discov- ering quantum phase transitions with fermionic neural networks, Phys. Rev. Lett.130, 036401 (2023)

    2023

  10. [19]

    Wilson, S

    M. Wilson, S. Moroni, M. Holzmann, N. Gao, F. Wu- darski, T. Vegge, and A. Bhowmik, Neural network ansatz for periodic wave functions and the homogeneous electron gas, Phys. Rev. B107, 235139 (2023)

    2023

  11. [20]

    D. Luo, D. D. Dai, and L. Fu, Pairing-based graph neu- ral network for simulating quantum materials (2023), arXiv:2311.02143 [cond-mat.str-el]

    2023 arXiv

  12. [21]

    J. Kim, G. Pescia, B. Fore, J. Nys, G. Carleo, S. Gandolfi, M. Hjorth-Jensen, and A. Lovato, Neural-network quan- tum states for ultra-cold Fermi gases, Commun. Phys.7, 1 (2024)

    2024

  13. [22]

    Smith, Y

    C. Smith, Y. Chen, R. Levy, Y. Yang, M. A. Morales, and S. Zhang, Unified variational approach description of ground-state phases of the two-dimensional electron gas, Phys. Rev. Lett.133, 266504 (2024)

    2024

  14. [23]

    Pescia, J

    G. Pescia, J. Nys, J. Kim, A. Lovato, and G. Carleo, Message-passing neural quantum states for the homoge- neous electron gas, Phys. Rev. B110, 035108 (2024)

    2024

  15. [24]

    J. A. Sobral, M. Perle, and M. S. Scheurer, Physics- informed transformers for electronic quantum states (2024), arXiv:2412.12248 [cond-mat.str-el]

    2024

  16. [25]

    X. Li, Y. Qian, W. Ren, Y. Xu, and J. Chen, Emergent wigner phases in moir´ e superlattice from deep learning (2024), arXiv:2406.11134 [physics.comp-ph]

    2024 arXiv

  17. [26]

    D. Luo, D. D. Dai, and L. Fu, Simulating moir´ e quantum matter with neural network (2024), arXiv:2406.17645 [cond-mat.str-el]

    2024 arXiv

  18. [27]

    Y. Teng, D. D. Dai, and L. Fu, Solving and visual- izing fractional quantum Hall wavefunctions with neu- ral network, arXiv 10.48550/arXiv.2412.00618 (2024), 2412.00618

    2024 doi arXiv

  19. [28]

    Y. Qian, T. Zhao, J. Zhang, T. Xiang, X. Li, and J. Chen, Taming landau level mixing in fractional quantum hall states with deep learning (2024), arXiv:2412.14795 [cond- mat.str-el]

    2024 arXiv

  20. [29]

    Vaswani, N

    A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. Kaiser, and I. Polosukhin, At- tention is all you need, Advances in neural information processing systems30, 5998 (2017)

    2017

  21. [30]

    E. C. Regan, D. Wang, C. Jin, M. I. Bakti Utama, B. Gao, X. Wei, S. Zhao, W. Zhao, Z. Zhang, K. Yu- migeta,et al., Mott and generalized wigner crystal states in wse2/ws2 moir´ e superlattices, Nature579, 359 (2020)

    2020

  22. [31]

    Y. Tang, L. Li, T. Li, Y. Xu, S. Liu, K. Barmak, K. Watanabe, T. Taniguchi, A. H. MacDonald, J. Shan, and K. F. Mak, Simulation of Hubbard model physics in WSe2/WS2 moir´ e superlattices, Nature579, 353 (2020)

    2020

  23. [32]

    E. A. Arsenault, Y. Li, B. Yang, X. Wang, H. Park, E. Mosconi, E. Ronca, T. Taniguchi, K. Watanabe, D. Gamelin, A. Millis, C. R. Dean, F. de Angelis, X. Xu, and X. Y. Zhu, Two-Dimensional Moir\’e Polaronic Elec- tron Crystals, Phys. Rev. Lett.132, 126501 (2024)

    2024

  24. [33]

    H. Li, S. Li, E. C. Regan, D. Wang, W. Zhao, S. Kahn, K. Yumigeta, M. Blei, T. Taniguchi, K. Watanabe, et al., Imaging two-dimensional generalized wigner crys- tals, Nature597, 650 (2021)

    2021

  25. [34]

    W. Zhao, B. Shen, Z. Tao, Z. Han, K. Kang, K. Watan- abe, T. Taniguchi, K. F. Mak, and J. Shan, Gate-tunable heavy fermions in a moir´ e kondo lattice, Nature616, 61 (2023)

    2023

  26. [35]

    F. Wu, T. Lovorn, E. Tutuc, and A. H. MacDonald, Hub- bard model physics in transition metal dichalcogenide moir´ e bands, Phys. Rev. Lett.121, 026402 (2018)

    2018

  27. [36]

    Zhang, H

    Y. Zhang, H. Isobe, and L. Fu, Density functional ap- proach to correlated moire states: itinerant magnetism, arXiv preprint arXiv:2005.04238 (2020)

    2020 arXiv

  28. [37]

    A. P. Reddy, T. Devakul, and L. Fu, Artificial atoms, wigner molecules, and an emergent kagome lattice in semiconductor moir´ e superlattices, Physical Review Let- ters131, 246501 (2023)

    2023

  29. [38]

    Hornik, M

    K. Hornik, M. Stinchcombe, and H. White, Multilayer feedforward networks are universal approximators, Neu- ral Networks2, 359 (1989)

    1989

  30. [39]

    H. Kim, Y. Zhou, Y. Xu, K. Varma, A. H. Karam- lou, I. T. Rosen, J. C. Hoke, C. Wan, J. P. Zhou, W. D. Oliver, Y. D. Lensky, K. Q. Weinberger, and E.-A. Kim, Attention to Quantum Complexity, arXiv 10.48550/arXiv.2405.11632 (2024), 2405.11632

    2024 doi

  31. [40]

    Pescia, J

    G. Pescia, J. Han, A. Lovato, J. Lu, and G. Carleo, Neural-network quantum states for periodic systems in continuous space, Phys. Rev. Res.4, 023138 (2022)

    2022

  32. [41]

    Cassella, H

    G. Cassella, H. Sutterud, S. Azadi, N. D. Drummond, D. Pfau, J. S. Spencer, and W. M. C. Foulkes, Discover- ing Quantum Phase Transitions with Fermionic Neural Networks, Phys. Rev. Lett.130, 036401 (2023)

    2023

  33. [42]

    R. P. Feynman and M. Cohen, Energy Spectrum of the Excitations in Liquid Helium, Phys. Rev.102, 1189 (1956)

    1956

  34. [43]

    Y. Kwon, D. M. Ceperley, and R. M. Martin, Effects of three-body and backflow correlations in the two- dimensional electron gas, Phys. Rev. B48, 12037 (1993)

    1993

  35. [44]

    Luo and B

    D. Luo and B. K. Clark, Backflow Transformations via Neural Networks for Quantum Many-Body Wave Func- tions, Phys. Rev. Lett.122, 226401 (2019)

    2019

  36. [45]

    X. Li, Z. Li, and J. Chen, Ab initio calculation of real solids via neural network ansatz, Nat. Commun.13, 1 (2022)

    2022

  37. [46]

    M. J. Buehler, Graph-Aware Isomorphic Atten- tion for Adaptive Dynamics in Transformers, arXiv 10.48550/arXiv.2501.02393 (2025), 2501.02393

    2025 doi arXiv

  38. [47]

    Anandan and Y

    J. Anandan and Y. Aharonov, Geometry of quantum evo- lution, Phys. Rev. Lett.65, 1697 (1990)

    1990

  39. [48]

    Metropolis and S

    N. Metropolis and S. Ulam, The Monte Carlo Method, J. Am. Stat. Assoc. (1949)

    1949

  40. [49]

    W. K. Hastings, Monte Carlo sampling methods using Markov chains and their applications, Biometrika57, 97 (1970)

    1970

  41. [50]

    Amari, Natural Gradient Works Efficiently in Learn- ing, Neural Comput.10, 251 (1998)

    S.-i. Amari, Natural Gradient Works Efficiently in Learn- ing, Neural Comput.10, 251 (1998)

    1998

  42. [51]

    Sorella, Green Function Monte Carlo with Stochastic Reconfiguration, Phys

    S. Sorella, Green Function Monte Carlo with Stochastic Reconfiguration, Phys. Rev. Lett.80, 4558 (1998)

    1998

  43. [52]

    Martens and R

    J. Martens and R. Grosse, Optimizing Neural Networks with Kronecker-factored Approximate Curvature, inIn- ternational Conference on Machine Learning(PMLR,

  44. [53]

    Campos Venuti and P

    L. Campos Venuti and P. Zanardi, Quantum Critical Scaling of the Geometric Tensors, Phys. Rev. Lett.99, 18 095701 (2007)

    2007

  45. [54]

    Gu, FIDELITY APPROACH TO QUANTUM PHASE TRANSITIONS, Int

    S.-J. Gu, FIDELITY APPROACH TO QUANTUM PHASE TRANSITIONS, Int. J. Mod. Phys B24, 4371 (2010)

    2010

  46. [55]

    Zhang, N

    Y. Zhang, N. F. Yuan, and L. Fu, Moir´ e quantum chem- istry: Charge transfer in transition metal dichalcogenide superlattices, Physical Review B102, 10.1103/phys- revb.102.201115 (2020)

    2020 doi

  47. [56]

    Fuchs, C

    D. Fuchs, C. W. Schneider, R. Schneider, and H. Ri- etschel, High dielectric constant and tunability of epitax- ial SrTiO3 thin film capacitors, J. Appl. Phys.85, 7362 (1999)

    1999

  48. [57]

    Ganahl, J

    M. Ganahl, J. Beall, M. Hauru, A. G. M. Lewis, T. Wo- jno, J. H. Yoo, Y. Zou, and G. Vidal, Density Matrix Renormalization Group with Tensor Processing Units, PRX Quantum4, 010317 (2023)

    2023

  49. [58]

    Morales-Dur´ an, A

    N. Morales-Dur´ an, A. H. MacDonald, and P. Potasz, Metal-insulator transition in transition metal dichalco- genide heterobilayer moir´ e superlattices, Physical Review B103, L241110 (2021)

    2021

  50. [59]

    Y. Yang, M. A. Morales, S. Zhang,et al., Metal-insulator transition in a semiconductor heterobilayer model, Phys- ical Review Letters132, 076503 (2024)

    2024

  51. [60]

    D. Pfau, S. Axelrod, H. Sutterud, I. von Glehn, and J. S. Spencer, Accurate computation of quantum excited states with neural networks, Science385, 10.1126/sci- ence.adn0137 (2024)

    2024 doi

  52. [61]

    Hendry, H

    D. Hendry, H. Chen, P. Weinberg, and A. E. Feiguin, Chebyshev expansion of spectral functions using re- stricted Boltzmann machines, Phys. Rev. B104, 205130 (2021)

    2021

  53. [62]

    Irikura and H

    N. Irikura and H. Saito, Neural-network quantum states at finite temperature, Phys. Rev. Res.2, 013284 (2020)

    2020

  54. [63]

    S. Lu, G. Giu@articleLu2024Jan, author = Lu, Sirui and Giudice, Giacomo and Cirac, J. Ignacio, title = Variational Neural and Tensor Network Approximations of Thermal States, journal = arXiv, year = 2024, month = jan, eprint = 2401.14243, doi = 10.1103/Phys- RevB.111.075102 di...

    2024 arXiv

  55. [64]

    Jackiw and A

    R. Jackiw and A. Kerman, Time-dependent variational principle and the effective action, Phys. Lett. A71, 158 (1979)

    1979

  56. [65]

    M. J. Hartmann and G. Carleo, Neural-Network Ap- proach to Dissipative Quantum Many-Body Dynamics, Phys. Rev. Lett.122, 250502 (2019)

    2019

  57. [66]

    Nagy and V

    A. Nagy and V. Savona, Variational Quantum Monte Carlo Method with a Neural-Network Ansatz for Open Quantum Systems, Phys. Rev. Lett.122, 250501 (2019)

    2019

  58. [67]

    Vicentini, A

    F. Vicentini, A. Biella, N. Regnault, and C. Ciuti, Varia- tional Neural-Network Ansatz for Steady States in Open Quantum Systems, Phys. Rev. Lett.122, 250503 (2019)

    2019

  59. [68]

    Yoshioka and R

    N. Yoshioka and R. Hamazaki, Constructing neural sta- tionary states for open quantum many-body systems, Phys. Rev. B99, 214306 (2019)

    2019

  60. [69]

    M. Reh, M. Schmitt, and M. G¨ arttner, Time-Dependent Variational Principle for Open Quantum Systems with Artificial Neural Networks, Phys. Rev. Lett.127, 230501 (2021)

    2021

  61. [70]

    Mellak, E

    J. Mellak, E. Arrigoni, and W. von der Linden, Deep neu- ral networks as variational solutions for correlated open quantum systems, Commun. Phys.7, 1 (2024)

    2024

  62. [71]

    Reuther, J

    A. Reuther, J. Kepner, C. Byun, S. Samsi, W. Ar- cand, D. Bestor, B. Bergeron, V. Gadepally, M. Houle, M. Hubbell, M. Jones, A. Klein, L. Milechin, J. Mullen, A. Prout, A. Rosa, C. Yee, and P. Michaleas, Interactive supercomputing on 40,000 cores for machine learning and data ...

    2018

  63. [72]

    D. P. James S. Spencer and F. Contributors, FermiNet (2020)

    2020

  64. [73]

    Chen, Poor Man’s Forward Laplacian (2024)

    Y. Chen, Poor Man’s Forward Laplacian (2024)

    2024

  65. [74]

    L. M. Fraser, W. M. C. Foulkes, G. Rajagopal, R. J. Needs, S. D. Kenny, and A. J. Williamson, Finite-size effects and Coulomb interactions in quantum Monte Carlo calculations for homogeneous systems with peri- odic boundary conditions, Phys. Rev. B53, 1814 (1996)

    1996

  66. [75]

    Jastrow, Many-Body Problem with Strong Forces, Phys

    R. Jastrow, Many-Body Problem with Strong Forces, Phys. Rev.98, 1479 (1955)

    1955

  67. [76]

    Wilhelm, T

    P. Wilhelm, T. C. Lang, and A. M. L¨ auchli, Interplay of fractional chern insulator and charge density wave phases in twisted bilayer graphene, Physical Review B 103, 125406 (2021)

    2021

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.