Pith. sign in

REVIEW 3 major objections 6 minor 1 cited by

Pruned shadow Hamiltonians let one track lattice spin dynamics on a register that stays constant as system size grows.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · grok-4.5

2026-07-14 02:26 UTC pith:RSAAGUDX

load-bearing objection Solid methods paper that makes shadow Hamiltonian simulation usable for interacting lattices by controlled algebra pruning; claims hold inside a stated small-parameter regime, with loose but honest error bounds. the 3 major comments →

arxiv 2607.11882 v1 pith:RSAAGUDX submitted 2026-07-13 quant-ph

An efficient algorithm for approximate shadow Hamiltonian simulation

Abhijit Chakraborty , Bharath Sambasivam , Karunya Shirali , Hunter Nelson , Mafalda Ram\^oa , Sophia E. Economou , Edwin Barnes This is my paper
classification quant-ph
keywords shadow Hamiltonian simulationoperator pruningKrylov basislattice spin modelsmagnetization dynamicsout-of-time-order correlatorsquantum simulation
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

Shadow Hamiltonian simulation tracks observables by encoding a vector of their expectation values and evolving it with a derived shadow Hamiltonian. For interacting systems that vector normally grows exponentially, so the encoding needs as many qubits as ordinary state simulation. This paper shows that most operators generated by repeated commutators with the Hamiltonian contribute little to a chosen root observable. By systematically discarding the low-weight ones—via a graph on a fixed Pauli basis, a truncated Krylov (Lanczos) basis, or a hybrid of both—one obtains a far smaller algebra whose shadow register size saturates at a constant number of qubits once the physical system is large enough. Numerical tests on one- and two-dimensional mixed-field Ising models and the longitudinal-field XXZ chain show that magnetization, spin-current autocorrelators and out-of-time-order correlators can be followed with controllable error using registers of roughly 7–14 qubits even when the physical system has dozens or hundreds of spins. The method therefore turns shadow simulation into a practical tool for interacting many-body dynamics whenever a modest hierarchy of operator weights exists.

Core claim

For time-independent interacting Hamiltonians that possess a moderate small parameter, a controlled pruning of the operator algebra generated from a root observable yields an approximate shadow Hamiltonian whose qubit register saturates at a size independent of system size, while still reproducing the real-time dynamics of one- and higher-point correlators to a chosen accuracy.

What carries the argument

Operator pruning: three classical procedures (graph construction on a predefined orthonormal basis, multiplicative-weight operator Lanczos, and their hybrid) that retain only those nested commutators whose accumulated weight exceeds a threshold ε, producing a truncated orthonormal set S_O-bar whose associated shadow Hamiltonian is Hermitian and low-dimensional.

Load-bearing premise

A clear hierarchy of operator weights, induced by a small Hamiltonian parameter or by the multiplicative Krylov weight, must remain strong enough that discarding everything below a fixed numerical threshold keeps the approximation error under control out to the desired evolution time.

What would settle it

For the 1D mixed-field Ising model with transverse field hz of order 1 (instead of 0.3), measure whether the shadow-register size still saturates below the physical system size while the magnetization error stays below a fixed δ out to moderate times; if the register grows to ~N qubits or the error blows up, the pruning claim fails.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Magnetization dynamics of arbitrarily large 1D or 2D lattices can be tracked with a fixed handful of qubits once the pruning threshold and evolution time are fixed.
  • Unequal-time two- and four-point correlators (spin currents, OTOCs) become accessible on the same reduced register by enlarging the root set or by a product-algebra construction.
  • When the pruned shadow Hamiltonian remains sparse or low-rank, the same classical pre-processing can be paired with tensor networks or Pauli propagation for classical simulation beyond exact diagonalization.
  • Quantum hardware only needs to implement a Hermitian matrix of dimension equal to the pruned set, not a full N-qubit unitary.

Where Pith is reading between the lines

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

  • The same weight hierarchy that makes pruning work may also explain why light-cone truncations succeed for certain local observables, suggesting a unified operator-importance criterion.
  • If the classical pre-processing can be replaced by a quantum routine that samples high-weight commutators, the method could become fully scalable even when no small Hamiltonian parameter is known a priori.
  • Chemistry or high-energy Hamiltonians that possess a clear interaction-strength hierarchy are natural next targets once suitable fermionic or Majorana bases are substituted for Pauli strings.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 6 minor

Summary. The paper develops an approximate form of shadow Hamiltonian simulation for time-independent Hamiltonians. Exact shadow simulation encodes the vector of expectation values of an algebra closed under commutation with H; for interacting systems that algebra grows exponentially. The authors systematically prune that algebra to the operators most relevant to a chosen root observable, using three classical schemes: (i) a weighted layered graph on a predefined orthonormal basis (Pauli strings in the examples), (ii) a multiplicative-weight truncated operator Lanczos/Krylov construction, and (iii) a hybrid that builds a Krylov basis inside a graph-pruned subspace. They supply pseudocode (Algorithms 1–2), a priori error bounds via a Dyson series (predefined basis) and an Arnoldi residual (Krylov), and numerical benchmarks on 1D/2D mixed-field Ising and longitudinal-field XXZ models for magnetization, spin-current autocorrelators, and OTOCs. In regimes with a moderate small parameter they report shadow-register sizes that saturate well below the physical system size (e.g., ~10 qubits for large 1D MFIM chains; hybrid ~7 qubits for N=12).

Significance. Shadow Hamiltonian simulation was previously efficient mainly for free models; extending it controllably to interacting systems is a genuine algorithmic gap. The pruning framework, especially the hybrid scheme that keeps Krylov vectors sparse while reducing register size further, is a concrete contribution. Strengths include explicit algorithms, derived (if loose) error bounds with numerical checks, multi-model and multi-observable benchmarks against exact diagonalization, and an honest discussion of regimes where the weight hierarchy collapses (Supp. S.VI). If the saturation of register size persists at large N for fixed accuracy and time, the method offers a practical route to smaller encodings for observable dynamics and can also serve as a classical pre-processing or simulation tool depending on the pruned dimension. The work is therefore of clear interest to quantum algorithms and many-body dynamics.

major comments (3)
  1. [Methods D–E; Figs. 8–9] Methods D–E and Figs. 8–9: The a priori bounds on M and ϵ are substantially loose relative to the numerically required thresholds. While the paper acknowledges this, the claim of a “controlled approximation” still rests on the practitioner’s ability to choose ϵ so that the Heisenberg-series error stays below a target δ out to time t. Without a tighter bound or a validated empirical selection protocol (e.g., a cheap residual estimator that does not require full ED of the physical system), it remains unclear how to set the free parameters ϵ, ϵ1, ϵ2 for a new model/observable at a prescribed (δ,t). This is load-bearing for the “controlled” claim and should be strengthened or the language tempered with an explicit practical protocol.
  2. [Results; Figs. 2–3; Tab. I; Supp. S.III–S.IV] Results (Figs. 2–3, Tab. I) and the claim of “arbitrarily large” systems: Register-size saturation is demonstrated for moderate N, and Supp. S.III shows that once saturation is reached, further increasing N does not degrade accuracy for the cases tested. The extrapolation to arbitrarily large N at fixed n nevertheless assumes that no new high-weight operators become important as N grows. For 2D lattices and the power-law Ising model the crossover to saturation occurs later; the manuscript should state more carefully the conditions under which the plateau is guaranteed (or at least empirically stable) and should separate the quantum-register claim from the classical-memory bottleneck |S_O| ≲ O(2^25) that already limits preprocessing.
  3. [Discussion] Discussion: The paper positions the method as a quantum algorithm whose main gain is a smaller qubit register, yet it provides almost no characterization of the structure of the resulting shadow Hamiltonians (locality, sparsity, gate complexity of time evolution, cost of preparing the initial shadow state beyond counting Pauli strings). Without even a schematic resource estimate for one of the benchmark models, the comparison to light-cone methods and to standard product-formula simulation remains incomplete. A short analysis for the 1D MFIM hybrid case (or an explicit statement that full quantum-resource analysis is out of scope) is needed to support the efficiency claim at the level of a quantum algorithm rather than only at the level of encoding dimension.
minor comments (6)
  1. [Introduction] Introduction, paragraph on shadow encoding: residual editing artifact “This encoding blue, which can be viewed as mapping a sparse spin vector onto qubits” — remove “blue”.
  2. [Results; Methods C] Eq. (1) and surrounding text: the normalization A = ∑ |⟨Oj⟩|² and the identification of |j⟩ with the binary representation of j−1 should be stated once with consistent indexing; later two-point correlator encoding (Methods C) uses a different register layout that could be cross-referenced more clearly.
  3. [Results; Fig. 2] Fig. 2 caption and main text: the hybrid threshold is written as a pair ϵ ≡ (ϵ1, ϵ2) but the figure legend sometimes uses a single ϵ; unify notation across panels and the hybrid description in Sec. 3.
  4. [Methods B; Fig. 7] Methods B / Fig. 7: the OTOC scaling plot is useful; a one-sentence reminder that the physical TFD register is 2N qubits (so advantage is relative to 2N, not N) would avoid misreading the black dotted line.
  5. [Discussion; Supp. S.VI] Supplementary S.VI and the main-text discussion of failure modes are valuable; consider promoting a short “when pruning fails” paragraph into the main Discussion so that the hierarchy premise is visible without opening the SI.
  6. [Throughout] Minor typography: “RESUL TS”, “ST A TEMENT”, “T wo-point”, “DERIV A TION”, “OPERA TORS”, “P ARAMETERS”, “OBSER V ABLES”, “MAGNETIZA TION” — fix spacing in section headers throughout the PDF.

Circularity Check

0 steps flagged

No circularity: pruning criteria, Dyson/Arnoldi error bounds, and ED benchmarks are independent of one another; thresholds are free knobs, not fitted predictions.

full rationale

The paper's derivation chain is self-contained. Shadow encoding (Eq. 1) and exact invariance (Eq. 2) are taken from the external reference [41] (Somma et al., non-overlapping authors). Approximate pruning is introduced via explicit constructions: graph accumulated weight Ω(vk) (Eq. 9) for the predefined basis and multiplicative product of orthogonal norms (Eq. 13) for Krylov/hybrid. Error bounds are derived from the Dyson series residual (Methods D, Eq. 30) and the Arnoldi residual estimate (Methods E, Eq. 41), both parameter-free once ε and t are fixed; the paper itself shows the bounds are loose (Figs. 8–9) rather than tautological. Numerical claims (register-size saturation, magnetization/OTOC/SCACF accuracy) are validated exclusively against exact diagonalization of the physical Hamiltonian, not against quantities defined from the pruning thresholds. Thresholds ε are free simulation parameters analogous to a Trotter step; when the weight hierarchy collapses (hz o O(1) or high-weight observables) the paper reports that pruning ceases to help (Results; Supp. S.VI). No self-definitional loop, no fitted-input-as-prediction, no load-bearing self-citation, and no uniqueness theorem imported from the present authors appear.

Axiom & Free-Parameter Ledger

4 free parameters · 4 axioms · 2 invented entities

The work sits on standard quantum mechanics and the prior shadow-Hamiltonian encoding. The only free knobs that control the central accuracy–resource tradeoff are the pruning thresholds. No new physical entities are postulated; the ‘pruned shadow set’ and hybrid Krylov space are algorithmic constructions, not ontological claims.

free parameters (4)
  • pruning threshold ϵ (predefined basis)
    Chosen by hand (often ϵ=h_z^M or J_z^M) to trade accuracy vs |S_Ō|; central resource claims depend on this choice.
  • Krylov multiplicative threshold ϵ (or ϵ2 in hybrid)
    Cutoff on product of orthogonal-component norms; determines Krylov dimension and reported register sizes.
  • hybrid first-stage threshold ϵ1
    Pauli-graph cutoff that defines the subspace in which Krylov vectors are built; jointly with ϵ2 sets hybrid resource claims.
  • L_max / layer cutoffs
    Maximum graph or Krylov depth used in algorithms; practical truncation beyond the weight threshold.
axioms (4)
  • domain assumption Shadow state encoding and invariance under commutation with H (Eqs. 1–3) as introduced by Somma et al.
    Entire algorithm is an approximation of that framework; used throughout Results and Methods.
  • standard math Heisenberg evolution as nested commutators / Dyson series in the interaction picture (Eq. 4, Methods D).
    Standard operator evolution; basis for predefined-basis error bound.
  • standard math Arnoldi residual estimate for Krylov subspace time evolution (Methods E, Supp. S.I).
    Used to bound Krylov pruning error; cited from numerical linear algebra literature.
  • ad hoc to paper Hamiltonians of interest admit a small parameter that induces a usable operator-weight hierarchy for the chosen observables.
    Not true for generic interacting H; paper’s efficiency claims and SI S.VI inefficient cases rest on this modeling assumption.
invented entities (2)
  • Accumulated path-weight pruning graph on a predefined operator basis no independent evidence
    purpose: Select a truncated orthonormal set S_Ō with controllable relevance to the root observable.
    Algorithmic construction; independent_evidence false because it is a method, not a physical object, though its accuracy is falsifiable via ED comparisons.
  • Multiplicative-weight truncated operator Krylov / hybrid Krylov basis no independent evidence
    purpose: Further compress the shadow register while bounding classical storage of Krylov vectors.
    Algorithmic; same status as the graph pruning entity.

pith-pipeline@v1.1.0-grok45 · 30110 in / 3313 out tokens · 40783 ms · 2026-07-14T02:26:47.437844+00:00 · methodology

0 comments
read the original abstract

We propose an efficient algorithm based on shadow Hamiltonian simulation to approximately simulate the real-time dynamics of observables under time-independent Hamiltonians. Shadow Hamiltonian simulation works at the level of the operator algebra generated by the observables through commutators with the Hamiltonian. Exactly encoding the quantum state in this picture is generally inefficient for interacting systems due to the exponential growth of the operator algebra. Our algorithm overcomes this bottleneck by systematically identifying the elements of the algebra most relevant to the target observables. This targeted approach is a controlled approximation that yields a highly efficient quantum state encoding that substantially reduces the size of the qubit register required to perform the time evolution using the shadow Hamiltonian. We propose two main pruning schemes, one based on a predefined operator basis and another on a constructed Krylov basis. We also present a hybrid scheme that builds a Krylov basis within a pruned algebra in the predefined basis. We benchmark our algorithm using lattice spin systems in one and two dimensions, for both one- and higher-point correlators as observables.

Figures

Figures reproduced from arXiv: 2607.11882 by Abhijit Chakraborty, Bharath Sambasivam, Edwin Barnes, Hunter Nelson, Karunya Shirali, Mafalda Ram\^oa, Sophia E. Economou.

Figure 1
Figure 1. Figure 1: FIG. 1: Schematic of the pruned shadow Hamiltonian simulation algorithm to approximately simulate the evolution of the [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2: Pruned shadow Hamiltonian simulation of the 1D MFIM with [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3: Pruning methods applied to the 2D MFIM with [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4: Study of a two-point correlator using pruned shadow [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5: Study of a four-point correlator using the pruned [PITH_FULL_IMAGE:figures/full_fig_p009_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6: Protocol to measure the OTOC introduced in [PITH_FULL_IMAGE:figures/full_fig_p011_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7: Scaling of the number of qubits required for pruned [PITH_FULL_IMAGE:figures/full_fig_p012_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8: Threshold required to achieve a fixed accuracy [PITH_FULL_IMAGE:figures/full_fig_p013_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: FIG. 9: Theoretical and numerical threshold as a function of [PITH_FULL_IMAGE:figures/full_fig_p014_9.png] view at source ↗

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. Efficient computation of real-time correlators using Pauli Propagation

    quant-ph 2026-07 conditional novelty 6.0

    Short-time truncated Pauli propagation combined with low-rank positive semi-definite time extension recovers dynamical structure factors for 1D/2D Heisenberg models beyond the direct Pauli window.

Reference graph

Works this paper leans on

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

  1. [1]

    For example, for an N -qubit system, one could consider the basis of all traceless Pauli strings spanning su(2N )

    Predefined basis Consider an orthonormal operator basis B that spans the relevant algebra corresponding to the target Hamil- tonian. For example, for an N -qubit system, one could consider the basis of all traceless Pauli strings spanning su(2N ). For a fermionic system, one could choose a Ma- jorana basis for the Clifford algebra. Without loss of gen- eral...

  2. [2]

    Here, we introduce a truncated operator Lanczos algorithm that constructs the Krylov basis, S ¯O

    Krylov basis An alternative approach is to construct a Krylov oper- ator basis iteratively informed by the orbit of orthogonal operators stemming from the root observable [ 44]. Here, we introduce a truncated operator Lanczos algorithm that constructs the Krylov basis, S ¯O. For simplicity, we will again assume that there is a single root observable, ¯O a...

  3. [3]

    This ensures that the resulting Krylov vectors can- not have more than |S ¯O| elements of the predefined ba- sis, which scales polynomially in system size for a given threshold

    Hybrid approach In this hybrid method, instead of building the Krylov basis on the complete algebra generated by the nested commutators with the Hamiltonian, whose dimension scales exponentially for interacting systems, the basis is built on a subspace defined by the set S ¯O determined by the predefined pruning method for a given thresh- old. This ensures ...

  4. [4]

    S. P. Jordan, H. Krovi, K. S. M. Lee, and J. Preskill, BQP-completeness of scattering in scalar quantum field theory, Quantum 2, 44 (2018)

    2018

  5. [5]

    M. C. Bañuls, R. Blatt, J. Catani, A. Celi, J. I. Cirac, M. Dalmonte, L. Fallani, K. Jansen, M. Lewenstein, S. Montangero, C. A. Muschik, B. Reznik, E. Rico, L. Tagliacozzo, K. Van Acoleyen, F. Verstraete, U.-J. Wiese, M. Wingate, J. Zakrzewski, and P. Zoller, Simu- lating lattice gauge theories within quantum technologies, The European Physical Journal D...

    2020

  6. [6]

    Davoudi, C.-C

    Z. Davoudi, C.-C. Hsieh, and S. V. Kadam, Scattering wave packets of hadrons in gauge theories: Preparation on a quantum computer, Quantum 8, 1520 (2024)

    2024

  7. [7]

    E. R. Bennewitz, B. Ware, A. Schuckert, A. Lerose, F. M. Surace, R. Belyansky, W. Morong, D. Luo, A. De, K. S. Collins, O. Katz, C. Monroe, Z. Davoudi, and A. V. Gorshkov, Simulating Meson Scattering on Spin Quantum Simulators, Quantum 9, 1773 (2025)

    2025

  8. [8]

    Karrasch and J

    C. Karrasch and J. E. Moore, Luttinger liquid physics from the infinite-system density matrix renormalization group, Phys. Rev. B 86, 155156 (2012)

    2012

  9. [9]

    Gustafson, Y

    E. Gustafson, Y. Meurice, and J. Unmuth-Yockey, Quan- tum simulation of scattering in the quantum ising model, Phys. Rev. D 99, 094503 (2019)

    2019

  10. [10]

    V. B. Bulchandani, C. Karrasch, and J. E. Moore, Superdiffusive transport of energy in one-dimensional metals, Proceedings of the Na- tional Academy of Sciences 117, 12713 (2020) , https://www.pnas.org/doi/pdf/10.1073/pnas.1916213117

    work page doi:10.1073/pnas.1916213117 2020

  11. [11]

    Eckstein, R

    T. Eckstein, R. Mansuroglu, P. Czarnik, J.-X. Zhu, M. J. Hartmann, L. Cincio, A. T. Sornborger, and Z. Holmes, Large-scale simulations of floquet physics on near-term quantum computers, npj Quantum Information 10, 84 (2024)

    2024

  12. [12]

    Y.-T. Lee, B. Pokharel, J. Cohn, A. Schleife, and A. Banerjee, Digital quantum simulation of spin transport, Phys. Rev. Lett. 136, 050603 (2026)

    2026

  13. [13]

    J. Li, R. Fan, H. Wang, B. Ye, B. Zeng, H. Zhai, X. Peng, and J. Du, Measuring out-of-time-order correlators on a nuclear magnetic resonance quantum simulator, Phys. Rev. X 7, 031011 (2017)

    2017

  14. [14]

    Y. Yang, A. Christianen, S. Coll-Vinent, V. Smelyan- skiy, M. C. Bañuls, T. E. O’Brien, D. S. Wild, and J. I. Cirac, Simulating prethermalization using near-term quantum computers, PRX Quantum 4, 030320 (2023)

    2023

  15. [15]

    D. Hahn, D. J. Luitz, and J. T. Chalker, Eigenstate cor- relations, the eigenstate thermalization hypothesis, and quantum information dynamics in chaotic many-body quantum systems, Phys. Rev. X 14, 031029 (2024)

    2024

  16. [16]

    Anand, S

    A. Anand, S. Srivastava, S. Gangopadhyay, and S. Ghose, Simulating quantum chaos on a quantum com- puter, Scientific Reports 14, 26890 (2024)

    2024

  17. [17]

    Asaduzzaman, R

    M. Asaduzzaman, R. G. Jha, and B. Sambasivam, Sachdev-ye-kitaev model on a noisy quantum computer, Phys. Rev. D 109, 105002 (2024)

    2024

  18. [18]

    Perrin, T

    H. Perrin, T. Scoquart, A. I. Pavlov, and N. V. Gnezdilov, Dynamic thermalization on noisy quantum hardware, Communications Physics 8, 95 (2025)

    2025

  19. [19]

    Kassal, S

    I. Kassal, S. P. Jordan, P. J. Love, M. Mohseni, and A. Aspuru-Guzik, Polynomial-time quantum algorithm for the simulation of chemical dynamics, Proceedings of the National Academy of Sciences 105, 18681 (2008) , https://www.pnas.org/doi/pdf/10.1073/pnas.0808245105

    doi:10.1073/pnas.0808245105 2008

  20. [20]

    H. H. S. Chan, R. Meister, T. Jones, D. P. Tew, and S. C. Benjamin, Grid-based meth- ods for chemistry simulations on a quantum computer, Science Advances 9, eabo7484 (2023) , https://www.science.org/doi/pdf/10.1126/sciadv.abo7484

    doi:10.1126/sciadv.abo7484 2023

  21. [21]

    S. S. Kale and S. Kais, Simulation of chemical reactions on a quantum computer, The Journal of Physical Chem- istry Letters 15, 5633 (2024)

    2024

  22. [22]

    Orús, Tensor networks for complex quantum systems, Nature Reviews Physics 1, 538 (2019)

    R. Orús, Tensor networks for complex quantum systems, Nature Reviews Physics 1, 538 (2019)

    2019

  23. [23]

    M. S. Rudolph, T. Jones, Y. Teng, A. Angrisani, 16 and Z. Holmes, Pauli propagation: A computational framework for simulating quantum systems, (2025), arXiv:2505.21606 [quant-ph]

    Pith/arXiv arXiv 2025

  24. [24]

    Frías-Pérez and M

    M. Frías-Pérez and M. C. Bañuls, Light cone tensor network and time evolution, Phys. Rev. B 106, 115117 (2022)

    2022

  25. [25]

    Y. Kim, A. Eddins, S. Anand, K. X. Wei, E. van den Berg, S. Rosenblatt, H. Nayfeh, Y. Wu, M. Zaletel, K. Temme, and A. Kandala, Evidence for the utility of quantum computing before fault tolerance, Nature 618, 500505 (2023)

    2023

  26. [26]

    M. Suzuki, General theory of fractal path inte- grals with applications to manybody theories and statistical physics, Journal of Mathematical Physics 32, 400 (1991) , https://pubs.aip.org/aip/jmp/article- pdf/32/2/400/19166143/400_1_online.pdf

    1991

  27. [27]

    D. W. Berry, G. Ahokas, R. Cleve, and B. C. Sanders, Efficient quantum algorithms for simulating sparse hamil- tonians, Communications in Mathematical Physics 270, 359 (2007)

    2007

  28. [28]

    Poulin, A

    D. Poulin, A. Qarry, R. Somma, and F. Verstraete, Quantum simulation of time-dependent hamiltonians and the convenient illusion of hilbert space, Physical Review Letters 106 (2011), 10.1103/physrevlett.106.170501

    doi:10.1103/physrevlett.106.170501 2011

  29. [29]

    Wiebe, D

    N. Wiebe, D. W. Berry, P. Høyer, and B. C. Sanders, Simulating quantum dynamics on a quantum computer, Journal of Physics A: Mathematical and Theoretical 44, 445308 (2011)

    2011

  30. [30]

    A. M. Childs, A. Ostrander, and Y. Su, Faster quantum simulation by randomization, Quantum 3, 182 (2019)

    2019

  31. [31]

    A. M. Childs, Y. Su, M. C. Tran, N. Wiebe, and S. Zhu, Theory of trotter error with commutator scaling, Phys. Rev. X 11, 011020 (2021)

    2021

  32. [32]

    Campbell, Random compiler for fast hamilto- nian simulation, Physical Review Letters 123 (2019), 10.1103/physrevlett.123.070503

    E. Campbell, Random compiler for fast hamilto- nian simulation, Physical Review Letters 123 (2019), 10.1103/physrevlett.123.070503

    doi:10.1103/physrevlett.123.070503 2019

  33. [33]

    Nakaji, M

    K. Nakaji, M. Bagherimehrab, and A. Aspuru-Guzik, High-order randomized compiler for hamiltonian simula- tion, PRX Quantum 5, 020330 (2024)

    2024

  34. [34]

    A. M. Childs and N. Wiebe, Hamiltonian simulation us- ing linear combinations of unitary operations, Quantum Info. Comput. 12, 901924 (2012)

    2012

  35. [35]

    Chakraborty, Implementing any Linear Combination of Unitaries on Intermediate-term Quantum Computers, Quantum 8, 1496 (2024)

    S. Chakraborty, Implementing any Linear Combination of Unitaries on Intermediate-term Quantum Computers, Quantum 8, 1496 (2024)

    2024

  36. [36]

    G. H. Low and I. L. Chuang, Optimal hamiltonian sim- ulation by quantum signal processing, Phys. Rev. Lett. 118, 010501 (2017)

    2017

  37. [37]

    Alexis, G

    M. Alexis, G. Mnatsakanyan, and C. Thiele, Quantum signal processing and nonlinear fourier analysis, Revista Matemática Complutense 37, 655694 (2024)

    2024

  38. [38]

    X. Yuan, S. Endo, Q. Zhao, Y. Li, and S. C. Benjamin, Theory of variational quantum simulation, Quantum 3, 191 (2019)

    2019

  39. [39]

    Cîrstoiu, Z

    C. Cîrstoiu, Z. Holmes, J. Iosue, L. Cincio, P. J. Coles, and A. Sornborger, Variational fast forwarding for quan- tum simulation beyond the coherence time, npj Quantum Information 6, 82 (2020)

    2020

  40. [40]

    Y.-X. Yao, N. Gomes, F. Zhang, C.-Z. Wang, K.-M. Ho, T. Iadecola, and P. P. Orth, Adaptive variational quantum dynamics simulations, PRX Quantum 2, 030307 (2021)

    2021

  41. [41]

    Kökcü, T

    E. Kökcü, T. Steckmann, Y. Wang, J. K. Freericks, E. F. Dumitrescu, and A. F. Kemper, Fixed depth hamiltonian simulation via cartan decomposition, Phys. Rev. Lett. 129, 070501 (2022)

    2022

  42. [42]

    Alsheikh, E

    O. Alsheikh, E. Kökcü, B. N. Bakalov, and A. F. Kem- per, Redcard: A quantum assisted algorithm for fixed- depth unitary synthesis via cartan decomposition, (2025), arXiv:2512.06070 [quant-ph]

    arXiv 2025

  43. [43]

    Sambasivam, K

    B. Sambasivam, K. Sherbert, K. Shirali, N. J. May- hall, E. Barnes, and S. E. Economou, Times-adapt: A quantum algorithm for real-time evolution in low- energy subspaces using fixed-depth circuits, (2026), arXiv:2603.02305 [quant-ph]

    arXiv 2026

  44. [44]

    R. D. Somma, R. King, R. Kothari, T. E. OBrien, and R. Babbush, Shadow hamiltonian simulation, Nature Communications 16 (2025), 10.1038/s41467-025-57451-z

    work page doi:10.1038/s41467-025-57451-z 2025

  45. [45]

    Aguilar, S

    G. Aguilar, S. Cichy, J. Eisert, and L. Bittel, Full clas- sification of pauli lie algebras, (2024), arXiv:2408.00081 [quant-ph]

    Pith/arXiv arXiv 2024

  46. [46]

    Wiersema, E

    R. Wiersema, E. Kökcü, A. F. Kemper, and B. N. Bakalov, Classification of dynamical lie algebras of 2- local spin systems on linear, circular and fully con- nected topologies, npj Quantum Information 10 (2024), 10.1038/s41534-024-00900-2

    doi:10.1038/s41534-024-00900-2 2024

  47. [47]

    Nandy, A

    P. Nandy, A. S. Matsoukas-Roubeas, P. Martínez- Azcona, A. Dymarsky, and A. del Campo, Quantum dy- namics in krylov space: Methods and applications, Physics Reports 1125, 1 (2025)

    2025

  48. [48]

    P. N. Jepsen, J. Amato-Grill, I. Dimitrova, W. W. Ho, E. Demler, and W. Ketterle, Spin transport in a tunable heisenberg model realized with ultracold atoms, Nature 588, 403 (2020)

    2020

  49. [49]

    Sundar, A

    B. Sundar, A. Elben, L. K. Joshi, and T. V. Zache, Proposal for measuring out-of-time-ordered correlators at finite temperature with coupled spin chains, New Journal of Physics 24, 023037 (2022)

    2022

  50. [50]

    E. H. Lieb and D. W. Robinson, The finite group velocity of quantum spin systems, Communications in mathemat- ical physics 28, 251 (1972)

    1972

  51. [51]

    Wang and K

    Z. Wang and K. R. Hazzard, Tightening the lieb-robinson bound in locally interacting systems, PRX Quantum 1, 010303 (2020)

    2020

  52. [52]

    Wang and Q

    H. Wang and Q. Ye, Error bounds for the krylov sub- space methods for computations of matrix exponentials, SIAM Journal on Matrix Analysis and Applications 38, 155 (2017)

    2017

  53. [53]

    Michel and S

    M. Michel and S. Zell, Timeevolver: A program for time evolution with improved error bound, Computer Physics Communications 277, 108374 (2022)

    2022

  54. [54]

    Saad, Numerical methods for large eigenvalue prob- lems: revised edition (SIAM, 2011)

    Y. Saad, Numerical methods for large eigenvalue prob- lems: revised edition (SIAM, 2011). 17 Supplementary Information S.I. DERIV A TION OF ARNOLDI RESIDUE Rm(t) The Krylov approximation for the time evolution of the operator q0 inside the Krylov subspace is given by q(m) 0 (t) = VmetAm e1 [S1]. We now define the quantity rm(t) as rm(t) = d dt q(m) 0 (t) −...

    2011