Pith. sign in

REVIEW 3 major objections 5 minor 48 references

A 97-qubit superconducting processor measures magnon spectra, lifetimes, and scattering channels in a 2D quantum magnet with precision that classical tensor-network simulations cannot match.

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 · deepseek-v4-flash

2026-08-02 05:34 UTC pith:N4ITM673

load-bearing objection A convincing experimental tour de force with one untested assumption in the 97-qubit mode identification; deserves serious review with requests for an overlap check and tighter fits. the 3 major comments →

arxiv 2607.13301 v1 pith:N4ITM673 submitted 2026-07-14 quant-ph cond-mat.mes-hallcond-mat.mtrl-sci

Precision quantum simulation of magnon spectra and interactions

Trond I. Andersen , Nikita Astrakhantsev , Jeronimo Martinez , Will Morong , Johannes Motruk , Dario Rossi , Brayden Ware , Bryce Kobrin
show 324 more authors
Weijie Wu Elizabeth Bennewitz Manuel Rudolph Tom Westerhout Amira Abbas Rajeev Acharya Laleh Aghababaie Beni Ross Alcaraz Sayra Alcaraz Markus Ansmann Frank Arute Kunal Arya Walt Askew Juan Atalaya Christopher Ayala Ryan Babbush Brian Ballard Joseph Bardin Hector Bates Andreas Bengtsson Majid Bigdeli Karimi Alexander Bilmes Simon Bilodeau Felix Borjans Alexandre Bourassa Jenna Bovaird Dylan Bowers Leon Brill Peter Brooks Michael Broughton David Browne Brett Buchea Bob Buckley Tim Burger Brian Burkett Jamal Busnaina Nicholas Bushnell Jonas Bylander Anthony Cabrera Juan Campero Hung-Shen Chang Silas Chen Zijun Chen Ben Chiaro Liang-Ying Chih Agnetta Cleland Bryan Cochrane Matt Cockrell Josh Cogan Paul Conner Tom Connolly Harold Cook Rodrigo Cortinas William Courtney Alexander Crook Benjamin Curtin Adam Dally Sayan Das Martin Damyanov Dripto Debroy Himadri Dey Stijn de Graaf Laura De Lorenzo Sean Demura Lucia De Rose Agustin Di Paolo Leon Ding Howard Dobbs Paul Donohoe Ebru Dogan Ilya Drozdov Andrew Dunsworth Rasmus Edholm Valerie Ehimhen Alec Eickbusch Aviv Elbag Lior Ella Mahmoud Elzouka David Enriquez Catherine Erickson Lara Faoro Vinicius Ferreira Marcos Flores Leslie Flores Burgos Sam Fontes Ebrahim Forati Jeremiah Ford Brooks Foxen Masaya Fukami Alan Fung Lenny Fuste Suhas Ganjam Gonzalo Garcia Christopher Garrick Robert Gasca Helge Gehring Robert Geiger \'Elie Genois William Giang Dar Gilboa James Goeders Edward Gonzales Raja Gosula Alejandro Grajales Dau Dietrich Graumann Joel Grebel Alex Greene Jonathan Gross Jose Guerrero Tan Ha. Steve Habegger Tanner Hadick Ali Hadjikhani Michael Hamilton Monica Hansen Matthew Harrigan Sean Harrington Jeanne Hartshorn Stephen Heslin Paula Heu Oscar Higgott Reno Hiltermann Jeremy Hilton Hsin-Yuan Huang Michael Hucka Christopher Hudspeth Ashley Huff William Huggins Aditya Jayaraman Evan Jeffrey Shaun Jevons Zhang Jiang Xiaoxuan Jin Cody Jones Chaitali Joshi Kelsey Josund Pavol Juhas Andreas Kabel Bharath Kannan Hui Kang Kiseo Kang Amir Karamlou Ryan Kaufman Tanuj Khattar Mostafa Khezri Seon Kim Paul Klimov Can Knaut Alexander Korotkov Fedor Kostritsa John Mark Kreikebaum Ryuho Kudo Arun Kumar Ben Kueffler Vladislav Kurilovich Vitali Kutsko Nathan Lacroix Tiano Lange-Dei Brandon Langley Pavel Laptev Kim-Ming Lau Emma Leavell Lo\"ick Le Guevel Justin Ledford Joy Lee Kenny Lee Brian Lester Wendy Leung Matthew Lloyd Lily Li Wing Li Ming Li Alexander Lill Mike Lindmark William Livingston Aditya Locharla Erik Lucero Daniel Lundahl Aaron Lunt Sid Madhuk Donald Macaskill Aniket Maiti Ashley Maloney Salvatore Mandra Leigh Martin Orion Martin Eric Mascot Paul Masih Das Anselme Massimino Melvin Mathews Cameron Maxfield Jarrod McClean Matthew McEwen Seneca Meeks Anthony Megrant Tim Menke Kevin Miao Zlatko Minev Reza Molavi Sebastian Molina Shirin Montazeri Akira Nakamura Charles Neill Konstantin Nesterov Michael Newman Roselle Ngaloy Anthony Nguyen Murray Nguyen Chia-Hung Ni Murphy Niu Nicholas Noll Sergey Novikov Logan Oas Ricky Oliver Will Oliver Raymond Orosco Kristoffer Ottosson Alice Pagano Manan Patel Sherman Peek David Peterson Alex Pizzuto Thomas Plumb-Reyes Elias Portoles Rebecca Potter Orion Pritchard Sam Probst Michael Qian Chris Quintana Matthew Rakher Arpit Ranadive Ganesh Ramachandran Armin Razavi Matthew Reagor Rachel Resnick David Rhodes Daniel Riley Gabrielle Roberts Roberto Rodriguez Emma Ropes Eliott Rosenberg Emma Rosenfeld Dario Rosenstock Elizabeth Rossi Pedram Roushan David Rower Robert Salazar Kannan Sankaragomathi Murat Sarihan Kevin Satzinger Max Schaefer Sebastian Schroeder Henry Schurkus Aria Shahingohar Michael Shearn Aaron Shorter Shvarts Vladimir Vlad Sivak Spencer Small W. Clarke Smith David Sobel Barrett Spells Sofia Springer George Sterling Scott Stonemeyer John Su Jordan Suchard Youngkyu Sung Aaron Szasz Alex Sztein Tomoyuki Tanaka Madeline Taylor Jothi Thiruraman Douglas Thor Brandur Thorgrimsson Dogan Timucin Eifu Tomita Alfredo Torres M. Mert Torunbalci Hao Tran Abeer Vaishnav Justin Vargas Gopinath Vasalamarri Sergey Vdovichev Guifre Vidal Benjamin Villalonga Meghan Voorhees Steven Waltman Jonathan Waltz Shannon Wang Danni Wang James Watson Yonghua Wei Travis Weidel Theodore White Kristi Wong Bryan Woo Christopher Wood Maddy Woodson Cheng Xing Z. Jamie Yao Ping Yeh Bicheng Ying Juhwan Yoo Noureldin Yosri Elliot Young Grayson Young Adam Zalcman Ran Zhang Yaxing Zhang Yiqing Zhou Ningfeng Zhu Nicholas Zobrist Zhenjie Zou Sergio Boixo Yu Chen Julian Kelly Hartmut Neven Vadim Smelyanskiy Dvir Kafri L. B. Ioffe Kostyantyn Kechedzhi Xiao Mi Dmitry Abanin
This is my paper
classification quant-ph cond-mat.mes-hallcond-mat.mtrl-sci
keywords quantum simulationmagnonsspin wavesanalog-digital processorHamiltonian learningresponse functionssuperconducting qubitsXY model
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.

The paper reports a high-precision quantum simulation of a two-dimensional XY spin-1/2 magnet on a 97-qubit superconducting processor, using an interleaved analog-digital protocol with a Hamiltonian learned from cross-entropy benchmarking. It claims to measure the full temperature-dependent magnon response—frequencies, decay rates, and nonlinear two-magnon scattering—resolving individual quasi-particle modes rather than bulk averages. The key observations are that decay rates are enhanced near van Hove singularities and suppressed for edge- and corner-localized modes; that the high-energy decay exponent is sublinear, Γ∝(ε−ε0)^η with η<1, explained by self-consistent line broadening; and that corner modes scatter almost exclusively into one another. This matters because it is a concrete demonstration of precise, mode-resolved simulation of interacting quantum-magnet dynamics on a processor, and a benchmark for classical methods that fail away from small sizes or low temperatures.

Core claim

The central claim is that a 97-qubit analog-digital processor can perform high-precision spectroscopy of individual magnon modes in a 2D XY magnet, extracting linear response (temperature-dependent spectra and lifetimes) and nonlinear response (self-scattering, pump-probe, and coherent two-mode scattering). The measurements reveal that magnon decay rates vary strongly across the Brillouin zone: enhancement near van Hove singularities, and suppression for corner-localized modes; that the decay rate follows a mode-dependent power law Γ∝(ε−ε0)^η whose exponent falls below unity at high mode indices, which the authors attribute to self-consistent broadening of the scattering partners; and that c

What carries the argument

The load-bearing technique is the interleaving of analog Hamiltonian evolution with digital single-qubit Z-gates on a processor whose effective Hamiltonian (including small multi-qubit terms) has been learned to about 0.1% of the dominant coupling by optimizing cross-entropy benchmarking in 24-qubit patches and patching the result to the full device. Magnon modes are defined through the spin-wave (Holstein-Primakoff) Bogoliubov transformation of the mean-field Hamiltonian, giving mode shapes used both to excite (via phase-imprinted Z-rotations) and to read out (via weighted measurement) the retarded Green's function of each mode. Decay rates are modeled with a self-consistent Born approximat

Load-bearing premise

The 97-qubit claims rest on the assumption that the Hamiltonian learned at infinite temperature in 24-qubit patches, then patched to the full device, remains accurate at the low temperatures and 97-qubit scale used for the magnon measurements; if small long-range or temperature-dependent terms are missed, the extracted mode shapes and decay rates would mix several excitations.

What would settle it

Measure the same mode-resolved decay rates at 97 qubits using a Hamiltonian calibrated directly at the operating temperature (e.g., by learning from bitstrings of a cooled, tilted initial state close to the magnon operating point) and compare with the patched learned-Hamiltonian predictions; a mismatch in the mode-resolved decay rates beyond the reported median error, or a difference in the corner-mode pairing pattern, would falsify the largest-system claims. Alternatively, an exact state-vector simulation at 36-40 qubits with the learned Hamiltonian that fails to reproduce the measured 24-31

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

If this is right

  • If the learned Hamiltonian is accurate at the operating temperature, the same interleaved protocol can map the full mode-resolved response function of other 2D spin models on the processor, including anisotropic and frustrated couplings.
  • The observed sublinear exponent η<1 for high-energy modes establishes self-consistent broadening as a measurable feedback effect, giving a concrete target for analytic and numerical many-body theory beyond the Born approximation.
  • The finding that corner modes form a nearly closed scattering manifold implies that boundary-localized magnons can act as long-lived, weakly coupled information carriers, a feature that could matter for spintronic or magnonic device design.
  • The paper's evidence that MPS and PEPS cannot reach experimental precision at 97 qubits and intermediate temperatures, if correct, supplies a concrete quantum-advantage benchmark: a classically hard observable (the magnon decay rate) that a quantum processor can deliver.

Where Pith is reading between the lines

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

  • Because the paper validates the learned Hamiltonian only at infinite temperature and in 24-qubit patches, an immediate extension is to test the same Hamiltonian at the operating temperatures via low-temperature cross-entropy benchmarking or density-matrix reconstruction; success would strengthen confidence in the 97-qubit mode-resolved results.
  • The corner-mode manifold suggests a design principle the paper leaves implicit: in finite open-boundary spin lattices, modes with low effective coordination have strongly reduced scattering phase space, so geometric engineering of boundaries (e.g., notches or holes) could selectively protect or destabilize particular magnon modes.
  • The power-law exponent η, extracted from an energy-density ramp, is a new spectral observable that could be measured in other platforms, such as Rydberg arrays or trapped ions, as a test of whether the self-consistent broadening mechanism is universal or specific to the XY model.
  • If the learned Hamiltonian is ported to another device with the same architecture, the mode-resolved decay maps (e.g., the cross pattern in pump-probe data) could serve as a sensitive fingerprint for diagnosing small parameter drift in the effective Hamiltonian.

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 / 5 minor

Summary. The paper reports a hybrid analog-digital quantum simulation of magnon excitations in a 2D XY spin-1/2 model on a superconducting processor with up to 97 qubits. The authors combine Hamiltonian learning via XEB (with per-qubit-per-cycle errors near 5e-4) with interleaved digital Z-gates to excite and measure individual Holstein-Primakoff magnon modes. They present linear-response spectra and decay rates vs temperature and mode index, reporting van Hove singularity enhancement, suppression for corner-localized modes, and a mode-dependent power-law Γ ∝ (ε - ε0)^η with η < 1 for high-energy modes. They also perform nonlinear amplitude-dependent decay measurements and two-mode pump-probe / collider experiments, including a claim that corner modes scatter almost exclusively into each other. The main experimental results are benchmarked against exact simulations up to 31 qubits (median relative errors ~0.7% in frequency and ~7.8% in decay rate) and against MPS simulations for larger systems, where MPS is reported to become inaccurate at elevated temperatures.

Significance. If the central claims hold, this is a substantial advance in quantum simulation of magnetic response functions: it demonstrates mode-resolved control of individual quasiparticles in a 2D interacting spin system, direct measurement of nonlinear magnon scattering channels, and a concrete regime where classical tensor-network methods are claimed to be insufficient. The paper is unusually thorough in its validation: machine-checked small-system comparisons, XEB fidelity analysis, detailed error modeling, and a large supplementary study of classical complexity. The pump-probe and coherent two-mode spectroscopy protocols are novel and could become broadly useful. The main risk is that the 97-qubit mode-resolved conclusions rely on identifying the measured modes with Holstein-Primakoff modes of the pure XY model, whereas the actual device Hamiltonian contains many additional terms; this untested assumption is load-bearing for the corner-mode and mode-resolved decay claims.

major comments (3)
  1. [§S3 A, Eq. (S25) with Eq. (S5); Figs. 2b,d,f, 3, 5] The mode shapes φ^HP_μ used both to excite and to analyze magnon modes are obtained from the pure XY Hamiltonian (Eq. S25), but the actual device is described by the much richer learned Hamiltonian of Eq. (S5), which includes density-assisted hopping, longer-range XY couplings, and density-density terms, patched from 24-qubit XEB learning. Small-system benchmarks (Nq ≤ 31) validate the full pipeline against exact simulation, but they do not test whether these HP modes remain the true quasiparticle modes of the 97-qubit patched Hamiltonian. If the quadratic part of the full learned Hamiltonian is not diagonalized by the φ^HP modes, the measured χ_μ(t) mixes several true modes. The corner-mode manifold contains only four modes, so even a small overlap leakage could substantially alter the extracted decay rates and the conclusion that corner modes scatter only into each other. I request a d
  2. [§S1 F, Fig. 3b,c] The central claim of sublinear power-law behavior, η < 1 for high-energy modes, is extracted from a two-parameter fit Γ ∝ (ε - ε0)^η in which ε0 is itself estimated from the same decay-rate data (Sec. S1F, ε0 = -0.57 ± 0.01). The exponent η and ε0 are strongly correlated, so the reported mode-dependent η values and the distinction between η > 1 and η < 1 need a sensitivity analysis: fix ε0 at the MPS ground-state value (-0.554) and at the Hartree-Fock value, and show how η changes for the modes displayed in Fig. 3c. Without this, the comparison to the self-consistent broadening model (which requires γ_ijkl = Γ_i + Γ_j + Γ_k) is not fully convincing, since both the data fit and the model are adjusted. Please also report confidence intervals for η that include the uncertainty in ε0.
  3. [§S5 A and main-text MPS comparisons (Fig. 2)] The abstract's statement that matrix-product state simulations 'become inaccurate away from these limits' is supported partly by the experimental comparisons in Fig. 2, but the quantitative computational-complexity analysis in Sec. S5 uses a simplified nearest-neighbor XY model (Eq. S45), omits thermalization, and generally adopts assumptions that the authors themselves state are favorable to classical simulation. The resource estimates (e.g., χ ~ 2^25-2^30 for 97 qubits) therefore do not directly apply to the actual learned Hamiltonian used in the experiment. I request that the authors specify whether the MPS curves in Fig. 2 were computed with the full learned Hamiltonian, and, if so, provide convergence data with bond dimension for at least one elevated-temperature case. If only the simplified model was used, the classical-complexity claims should be tempered accordingly.
minor comments (5)
  1. [Eq. (1)] The text below Eq. (1) defines the energy-conservation delta as δ((ω_k+ω_l-ω_i-ω_j)/γ_ijkl), but Eq. (1) writes δ(Δω_ijkl / γ_ijkl). Please make the notation consistent with the Lorentzian δ_γ(x) defined in Sec. S3E.
  2. [Fig. 3c] The legend labels for the experimental data, the constant-linewidth theory, and the self-consistent theory are difficult to distinguish in grayscale; please use distinct markers/colors and include error bars on the experimental η values.
  3. [Fig. 3e] The caption says 'Solid lines are guides to the eye,' but several curves appear to be fits. Please specify exactly which lines are fits and which are just connecting points, and add error bars for the frequency points.
  4. [Abstract] The phrase 'matrix-product state simulations capture the dynamics well in either small systems or at low temperatures' would benefit from specifying which Hamiltonian was used in those MPS simulations (learned vs nearest-neighbor XY).
  5. [§S1 G, Eq. (S1)-(S2)] The nonlinear decay model contains a large number of free parameters (a0, f, α1, α2, c, ν, t0, y0) fitted simultaneously. The authors mention that 11 parameters are fitted to 204 points; it would be helpful to report the reduced χ² or a similar goodness-of-fit statistic for representative modes to support the claim that the model is not overfitting.

Circularity Check

0 steps flagged

No significant circularity: the magnon measurements are direct experimental observables, and the theoretical models are clearly labeled approximate guides rather than fits recycled as predictions.

full rationale

The paper's central results are measured response functions (frequencies, decay rates, nonlinear decay channels) read directly from the device, not derived from the same data by construction. The Holstein-Primakoff mode shapes in Eq. (S25) define the excitation and projection basis, but they are computed from the model Hamiltonian, not fitted to the measured spectra; the measured χμ(t) is a well-defined spin correlation function even if the HP identification is imperfect. The self-consistent broadening model in Eq. (1), with γ_ijkl = Γ_j + Γ_k + Γ_l, is a parameter-free fixed-point calculation: it does not reduce to the observed η by construction, and the paper explicitly reports quantitative disagreements with experiment (Fig. S17), which is the opposite of a fitted input being renamed a prediction. The Hamiltonian learning from Ref. [37] is a methodological self-citation, but it is independently benchmarked here against exact simulation on systems up to 31 qubits (median relative error 0.007 in frequency and 0.078 in decay rate), so the load-bearing validation is not a bare self-citation. The 97-qubit extrapolation rests on an untested assumption that the HP modes remain valid for the patched learned Hamiltonian; this is a validation gap and correctness risk, not a circular equation-level reduction. No quoted step exhibits Eq. X = Eq. Y by construction or a fitted parameter presented as a prediction.

Axiom & Free-Parameter Ledger

4 free parameters · 5 axioms · 0 invented entities

No new physical entities are introduced (the GGP is a pulse protocol; corner/edge modes are emergent classifications). The free parameters listed are all fit to or inferred from the experimental data, and the axioms are the standard mean-field and device-modeling assumptions that the 97-qubit conclusions rest on.

free parameters (4)
  • ε0 (ground-state energy density in Γ∝(ε−ε0)^η fit) = -0.57 ± 0.01
    Chosen by minimizing RMS fit error of the decay-rate vs energy-density data (Sec. S1F); MPS predicts -0.554. The exponent η is therefore a fit over a data-estimated reference point.
  • α1, α2, c, ν, t0, y0 (nonlinear decay model) = mode-dependent values (Figs. 4c,d and S8)
    Eleven parameters fit simultaneously to 204 data points per mode (4 amplitudes × 51 times) using Eq. (S2); the paper's κi conclusions compare these fitted rates.
  • Hartree–Fock rescaling factor for Δlog ω0 = 1.2
    Applied empirically to HF theory in Fig. 3f to match bulk high-μ behavior; the paper states it 'requires rescaling with empirical factor of 1.2'.
  • Lorentzian linewidth γ_ijkl in Eq. (1) = γ = Σ_i Γ_i (self-consistent)
    The broadening is set to the sum of the decay rates being computed; this self-consistent choice is what turns η>1 into η<1 (Fig. 3c,d). It is a modeling choice made to match the data.
axioms (5)
  • standard math Holstein-Primakoff expansion and bosonic Bogoliubov diagonalization define the magnon modes and mode shapes φμ_i (Eq. S25).
    Standard many-body tool used to define the excitation operators; not proved in the paper.
  • domain assumption The effective spin Hamiltonian Eq. (S5) with truncation of terms ≲0.001g captures all relevant device dynamics.
    Assumes the perturbative expansion in g/Δ and g/η converges and that XEB-learned terms are sufficient up to the experimental evolution times.
  • domain assumption Hamiltonian parameters learned at infinite temperature (XEB) transfer unchanged to the lower temperatures used for magnon measurements.
    Learning is done at T=∞ (Sec. S2), while magnon experiments run at ε≈-0.4 to -0.53; temperature independence is assumed.
  • domain assumption Measured energy density ε maps to a thermal magnon temperature via the mean-field relation Eq. (S41).
    The temperature used in the scattering-rate model is inferred from ε through a mean-field Bose occupation formula; the paper notes only rough cross-checks from ramp-recovery entropy.
  • domain assumption The qubit array is an isolated spin-1/2 system; coupler states and higher qubit levels are fully projected out.
    Relies on large anharmonicity η~20g and detuning Δ~100g; the effective Hamiltonian includes perturbative corrections but no leakage dynamics.

pith-pipeline@v1.3.0-alltime-deepseek · 54615 in / 13268 out tokens · 136521 ms · 2026-08-02T05:34:49.010462+00:00 · methodology

0 comments
read the original abstract

Quantum simulation promises to advance materials discovery by accurately simulating complex states of matter, their microscopic excitations, and macroscopic response functions. The central challenge in resolving the underlying interacting dynamics is to combine high-fidelity evolution with the sophisticated control necessary to manipulate individual quasi-particles in quantum many-body states. Here, we report on high-precision simulation of both linear and non-linear response functions in a 2D XY spin-1/2 magnet using an analog-digital superconducting processor of up to 97 qubits. By interleaving digital gates with analog evolution precisely characterized via Hamiltonian learning, we selectively excite magnons at tunable energy densities. Measuring first the linear magnon response -- a central probe in neutron-scattering experiments -- we extract temperature-dependent spectra and lifetimes. Our results reveal stark variations in magnon decay rates across the Brillouin zone, with enhancement near van Hove singularities and suppression for edge-localized modes. Next, we perform a suite of nonlinear measurements, including the study of self-scattering mechanisms, as well as pump-probe spectroscopy to directly characterize the magnon interactions. While matrix-product state simulations capture the dynamics well in either small systems or at low temperatures, their predictions become inaccurate away from these limits. This work demonstrates precise simulation of the interacting dynamics in quantum magnets, and provides key insights into quasi-particles and their microscopic scattering mechanisms.

Figures

Figures reproduced from arXiv: 2607.13301 by Aaron Lunt, Aaron Shorter, Aaron Szasz, Abeer Vaishnav, Adam Dally, Adam Zalcman, Aditya Jayaraman, Aditya Locharla, Agnetta Cleland, Agustin Di Paolo, Akira Nakamura, Alan Fung, Alec Eickbusch, Alejandro Grajales Dau, Alexander Bilmes, Alexander Crook, Alexander Korotkov, Alexander Lill, Alexandre Bourassa, Alex Greene, Alex Pizzuto, Alex Sztein, Alfredo Torres, Alice Pagano, Ali Hadjikhani, Amira Abbas, Amir Karamlou, Andreas Bengtsson, Andreas Kabel, Andrew Dunsworth, Aniket Maiti, Anselme Massimino, Anthony Cabrera, Anthony Megrant, Anthony Nguyen, Aria Shahingohar, Armin Razavi, Arpit Ranadive, Arun Kumar, Ashley Huff, Ashley Maloney, Aviv Elbag, Barrett Spells, Ben Chiaro, Benjamin Curtin, Benjamin Villalonga, Ben Kueffler, Bharath Kannan, Bicheng Ying, Bob Buckley, Brandon Langley, Brandur Thorgrimsson, Brayden Ware, Brett Buchea, Brian Ballard, Brian Burkett, Brian Lester, Brooks Foxen, Bryan Cochrane, Bryan Woo, Bryce Kobrin, Cameron Maxfield, Can Knaut, Catherine Erickson, Chaitali Joshi, Charles Neill, Cheng Xing, Chia-Hung Ni, Chris Quintana, Christopher Ayala, Christopher Garrick, Christopher Hudspeth, Christopher Wood, Cody Jones, Daniel Lundahl, Daniel Riley, Danni Wang, Dar Gilboa, Dario Rosenstock, Dario Rossi, David Browne, David Enriquez, David Peterson, David Rhodes, David Rower, David Sobel, Dietrich Graumann, Dmitry Abanin, Dogan Timucin, Donald Macaskill, Douglas Thor, Dripto Debroy, Dvir Kafri, Dylan Bowers, Ebrahim Forati, Ebru Dogan, Edward Gonzales, Eifu Tomita, Elias Portoles, \'Elie Genois, Eliott Rosenberg, Elizabeth Bennewitz, Elizabeth Rossi, Elliot Young, Emma Leavell, Emma Ropes, Emma Rosenfeld, Eric Mascot, Erik Lucero, Evan Jeffrey, Fedor Kostritsa, Felix Borjans, Frank Arute, Gabrielle Roberts, Ganesh Ramachandran, George Sterling, Gonzalo Garcia, Gopinath Vasalamarri, Grayson Young, Guifre Vidal, Hao Tran, Harold Cook, Hartmut Neven, Hector Bates, Helge Gehring, Henry Schurkus, Himadri Dey, Howard Dobbs, Hsin-Yuan Huang, Hui Kang, Hung-Shen Chang, Ilya Drozdov, Jamal Busnaina, James Goeders, James Watson, Jarrod McClean, Jeanne Hartshorn, Jenna Bovaird, Jeremiah Ford, Jeremy Hilton, Jeronimo Martinez, Joel Grebel, Johannes Motruk, John Mark Kreikebaum, John Su, Jonas Bylander, Jonathan Gross, Jonathan Waltz, Jordan Suchard, Jose Guerrero, Joseph Bardin, Josh Cogan, Jothi Thiruraman, Joy Lee, Juan Atalaya, Juan Campero, Juhwan Yoo, Julian Kelly, Justin Ledford, Justin Vargas, Kannan Sankaragomathi, Kelsey Josund, Kenny Lee, Kevin Miao, Kevin Satzinger, Kim-Ming Lau, Kiseo Kang, Konstantin Nesterov, Kostyantyn Kechedzhi, Kristi Wong, Kristoffer Ottosson, Kunal Arya, Laleh Aghababaie Beni, Lara Faoro, Laura De Lorenzo, L. B. Ioffe, Leigh Martin, Lenny Fuste, Leon Brill, Leon Ding, Leslie Flores Burgos, Liang-Ying Chih, Lily Li, Lior Ella, Logan Oas, Lo\"ick Le Guevel, Lucia De Rose, Maddy Woodson, Madeline Taylor, Mahmoud Elzouka, Majid Bigdeli Karimi, Manan Patel, Manuel Rudolph, Marcos Flores, Markus Ansmann, Martin Damyanov, Masaya Fukami, Matt Cockrell, Matthew Harrigan, Matthew Lloyd, Matthew McEwen, Matthew Rakher, Matthew Reagor, Max Schaefer, Meghan Voorhees, Melvin Mathews, Michael Broughton, Michael Hamilton, Michael Hucka, Michael Newman, Michael Qian, Michael Shearn, Mike Lindmark, Ming Li, M. Mert Torunbalci, Monica Hansen, Mostafa Khezri, Murat Sarihan, Murphy Niu, Murray Nguyen, Nathan Lacroix, Nicholas Bushnell, Nicholas Noll, Nicholas Zobrist, Nikita Astrakhantsev, Ningfeng Zhu, Noureldin Yosri, Orion Martin, Orion Pritchard, Oscar Higgott, Paula Heu, Paul Conner, Paul Donohoe, Paul Klimov, Paul Masih Das, Pavel Laptev, Pavol Juhas, Pedram Roushan, Peter Brooks, Ping Yeh, Rachel Resnick, Raja Gosula, Rajeev Acharya, Ran Zhang, Rasmus Edholm, Raymond Orosco, Rebecca Potter, Reno Hiltermann, Reza Molavi, Ricky Oliver, Robert Gasca, Robert Geiger, Roberto Rodriguez, Robert Salazar, Rodrigo Cortinas, Roselle Ngaloy, Ross Alcaraz, Ryan Babbush, Ryan Kaufman, Ryuho Kudo, Salvatore Mandra, Sam Fontes, Sam Probst, Sayan Das, Sayra Alcaraz, Scott Stonemeyer, Sean Demura, Sean Harrington, Sebastian Molina, Sebastian Schroeder, Seneca Meeks, Seon Kim, Sergey Novikov, Sergey Vdovichev, Sergio Boixo, Shannon Wang, Shaun Jevons, Sherman Peek, Shirin Montazeri, Shvarts Vladimir, Sid Madhuk, Silas Chen, Simon Bilodeau, Sofia Springer, Spencer Small, Stephen Heslin, Steve Habegger, Steven Waltman, Stijn de Graaf, Suhas Ganjam, Tan Ha., Tanner Hadick, Tanuj Khattar, Theodore White, Thomas Plumb-Reyes, Tiano Lange-Dei, Tim Burger, Tim Menke, Tom Connolly, Tomoyuki Tanaka, Tom Westerhout, Travis Weidel, Trond I. Andersen, Vadim Smelyanskiy, Valerie Ehimhen, Vinicius Ferreira, Vitali Kutsko, Vladislav Kurilovich, Vlad Sivak, Walt Askew, W. Clarke Smith, Weijie Wu, Wendy Leung, William Courtney, William Giang, William Huggins, William Livingston, Will Morong, Will Oliver, Wing Li, Xiao Mi, Xiaoxuan Jin, Yaxing Zhang, Yiqing Zhou, Yonghua Wei, Youngkyu Sung, Yu Chen, Zhang Jiang, Zhenjie Zou, Zijun Chen, Z. Jamie Yao, Zlatko Minev.

Figure 1
Figure 1. Figure 1: FIG. 1 [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2 [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3 [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4 [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: b, which displays the decay of mode 50 both in the absence of pumping (purple), and with separate pump￾ing of a corner mode (mode index 29; blue) and a bulk mode (80; green). Interestingly, we find that pumping mode 80 induces a large increase in the decay rate, while pumping the corner mode has almost no effect. To under￾stand this better, we next plot the pump-induced change in decay rate for all pairs o… view at source ↗

discussion (0)

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

Reference graph

Works this paper leans on

48 extracted references · 4 linked inside Pith

  1. [1]

    the PEPS bond dimensionχmust be unfeasibly large to capture the evolved states to sufficient fi- delity,

  2. [2]

    Let us first consider the fidelity

    expectation values from low-fidelity PEPS are in- deed inaccurate. Let us first consider the fidelity. Ref. [92] describes a method to approximate the running fidelity of a PEPS during simulation of a quantum circuit as a product of the per-gate fidelities. These fidelities are computed from the magnitude of singular values thrown away, which is precisely...

  3. [47]

    & Lindner, N

    Bairey, E., Arad, I. & Lindner, N. H. Learning a Local Hamiltonian from Local Measurements.Phys. Rev. Lett. 122, 020504 (2019)

    2019

  4. [48]

    & Preskill, J

    Huang, H.-Y., Kueng, R. & Preskill, J. Predicting many properties of a quantum system from very few measure- ments.Nat. Phys.16, 1050–1057 (2020)

    2020

  5. [49]

    K., Choi, J., Shaw, A

    Mark, D. K., Choi, J., Shaw, A. L., Endres, M. & Choi, S. Benchmarking quantum simulators using ergodic quan- tum dynamics.Phys. Rev. Lett.131, 110601 (2023)

    2023

  6. [50]

    & Cory, D

    Wiebe, N., Granade, C., Ferrie, C. & Cory, D. Quantum hamiltonian learning using imperfect quantum resources. Phys. Rev. A89, 042314 (2014)

    2014

  7. [51]

    E., Ferrie, C., Wiebe, N

    Granade, C. E., Ferrie, C., Wiebe, N. & Cory, D. G. Robust online Hamiltonian learning.New J. Phys.14, 103013 (2012)

    2012

  8. [52]

    Broughton, M.et al.Tensorflow quantum: A software framework for quantum machine learning.arXiv preprint arXiv:2003.02989(2021)

    Pith/arXiv arXiv 2003

  9. [53]

    & Gacon, J

    Jones, T. & Gacon, J. Efficient calculation of gradients in classical simulations of variational quantum algorithms. arXiv preprint arXiv:2009.02823(2020)

    Pith/arXiv arXiv 2009

  10. [54]

    & Schuster, D

    Leung, N., Abdelhafez, M., Koch, J. & Schuster, D. Speedup for quantum optimal control from automatic differentiation based on graphics processing units.Phys. Rev. A95, 042318 (2017)

    2017

  11. [55]

    & Held, K

    Wallerberger, M. & Held, K. Trie-based ranking of quan- tum many-body states.Phys. Rev. Res.4, 033238 (2022)

    2022

  12. [56]

    Lin, H. Q. Exact diagonalization of quantum-spin mod- els.Phys. Rev. B42, 6561–6567 (1990)

    1990

  13. [57]

    Dormand, J. R. & Prince, P. J. A family of embedded Runge-Kutta formulae.J. Comput. Appl. Math.6, 19–26 (1980)

    1980

  14. [58]

    Kingma, D. P. & Ba, J. Adam: A method for stochastic optimization. InInternational Conference on Learning Representations (ICLR)(2015)

    2015

  15. [59]

    Manole, T.et al.How much can we learn from quantum random circuit sampling?arXiv preprint arXiv:2510.09919(2025)

    arXiv 2025

  16. [60]

    Phys.14, 595–600 (2018)

    Boixo, S.et al.Characterizing quantum supremacy in near-term devices.Nat. Phys.14, 595–600 (2018)

    2018

  17. [61]

    & Primakoff, H

    Holstein, T. & Primakoff, H. Field Dependence of the Intrinsic Domain Magnetization of a Ferromagnet.Phys. Rev.58, 1098–1113 (1940)

    1940

  18. [62]

    Anderson, P. W. An approximate quantum theory of the antiferromagnetic ground state.Phys. Rev.86, 694–701 (1952)

    1952

  19. [63]

    & Joannopoulos, J

    Gomez-Santos, G. & Joannopoulos, J. D. Application of spin-wave theory to the ground state of XY quantum Hamiltonians.Phys. Rev. B36, 8707–8711 (1987)

    1987

  20. [64]

    Colpa, J. H. P. Diagonalization of the quadratic boson hamiltonian.Phys. A: Stat. Mech. Appl.93, 327–353 (1978)

    1978

  21. [65]

    Jonay, C., Huse, D. A. & Nahum, A. Coarse-grained dynamics of operator and state entanglement.arXiv preprint arXiv:1803.00089(2018)

    Pith/arXiv arXiv 2018

  22. [66]

    Rakovszky, T., von Keyserlingk, C. W. & Pollmann, F. Entanglement growth after inhomogenous quenches. Phys. Rev. B.100, 125139 (2019)

    2019

  23. [67]

    M., Verstraete, F

    Schuch, N., Wolf, M. M., Verstraete, F. & Cirac, J. I. En- tropy scaling and simulability by matrix product states. Phys. Rev. Lett.100, 030504 (2008)

    2008

  24. [68]

    & White, S

    Barthel, T., Schollw¨ ock, U. & White, S. R. Spectral func- tions in one-dimensional quantum systems at finite tem- perature using the density matrix renormalization group. Phys. Rev. B79, 245101 (2009)

    2009

  25. [69]

    Yi-Thomas, S., Ware, B., Sau, J. D. & White, C. D. Comparing numerical methods for hydrodynamics in a one-dimensional lattice spin model.Phys. Rev. B110, 134308 (2024)

    2024

  26. [70]

    White, C. D. Effective dissipation rate in a Liouvillian- graph picture of high-temperature quantum hydrody- namics.Phys. Rev. B.107, 094311 (2023)

    2023

  27. [71]

    & Rakovszky, T

    von Keyserlingk, C., Pollmann, F. & Rakovszky, T. Op- erator backflow and the classical simulation of quantum transport.Phys. Rev. B.105, 245101 (2022)

    2022

  28. [72]

    Rakovszky, T., von Keyserlingk, C. W. & Pollmann, F. Dissipation-assisted operator evolution method for capturing hydrodynamic transport.Phys. Rev. B.105, 075131 (2022)

    2022

  29. [73]

    E., Cao, X., Avdoshkin, A., Scaffidi, T

    Parker, D. E., Cao, X., Avdoshkin, A., Scaffidi, T. & Alt- man, E. A universal operator growth hypothesis.Phys. Rev. X.9, 041017 (2019)

    2019

  30. [74]

    Artiaco, C., Fleckenstein, C., Aceituno Ch´ avez, D., Kvorning, T. K. & Bardarson, J. H. Efficient large-scale many-body quantum dynamics via local-information time evolution.PRX Quantum5, 020352 (2024)

    2024

  31. [75]

    P., Trauzettel, B., Klein Kvorning, T., Bar- darson, J

    Bauer, N. P., Trauzettel, B., Klein Kvorning, T., Bar- darson, J. H. & Artiaco, C. Local information flow in quantum quench dynamics.Phys. Rev. A112, 022221 (2025)

    2025

  32. [76]

    qsim (2025)

    Quantum AI team and collaborators. qsim (2025)

    2025

  33. [77]

    Haegeman, J.et al.Time-Dependent Variational Princi- ple for Quantum Lattices.Phys. Rev. Lett.107, 070601 (2011)

    2011

  34. [78]

    Paeckel, S.et al.Time-evolution methods for matrix- product states.Ann. Phys. (N. Y.)411, 167998 (2019)

    2019

  35. [79]

    Cataldi, G.et al.Hilbert curve vs Hilbert space: ex- ploiting fractal 2D covering to increase tensor network efficiency.Quantum5, 556 (2021)

    2021

  36. [80]

    & Bar Lev, Y

    Kloss, B., Reichman, D. & Bar Lev, Y. Studying dynam- ics in two-dimensional quantum lattices using tree tensor network states.SciPost Phys.9, 070 (2020)

    2020

  37. [81]

    Vovrosh, J.et al.Simulating dynamics of the two- dimensional transverse-field ising model: A comparative study of large-scale classical numerics.Phys. Rev. Res. 8, 023311 (2026)

    2026

  38. [82]

    & White, S

    Yang, M. & White, S. R. Time-dependent variational principle with ancillary krylov subspace.Phys. Rev. B 102, 094315 (2020)

    2020

  39. [83]

    Singh, S., Pfeifer, R. N. C. & Vidal, G. Tensor network states and algorithms in the presence of a global U(1) symmetry.Phys. Rev. B83, 115125 (2011)

    2011

  40. [84]

    Hubig, C., McCulloch, I. P. & Schollw¨ ock, U. Generic construction of efficient matrix product operators.Phys. Rev. B95, 035129 (2017)

    2017

  41. [85]

    P., Soeteman, A., Cade, C

    Thompson, A. P., Soeteman, A., Cade, C. & Niesen, I. Non-zero noise extrapolation: accurately simulat- ing noisy quantum circuits with tensor networks.arXiv preprint arXiv:2501.13237(2025)

    arXiv 2025

  42. [86]

    Zhou, Y., Stoudenmire, E. M. & Waintal, X. What limits the simulation of quantum computers?Phys. Rev. X.10, 041038 (2020). 42

    2020

  43. [87]

    Ayral, T.et al.Density-matrix renormalization group algorithm for simulating quantum circuits with a finite fidelity.PRX Quantum4, 020304 (2023)

    2023

  44. [88]

    Mandr` a, S.et al.A heuristic for matrix product state simulation of out-of-equilibrium dynamics of two- dimensional transverse-field ising models.arXiv preprint arXiv:2511.23438(2025)

    arXiv 2025

  45. [89]

    & Fishman, M

    Tindall, J. & Fishman, M. Gauging tensor networks with belief propagation.SciPost Physics15, 222 (2023)

    2023

  46. [90]

    & Chan, G

    Evenbly, G., Pancotti, N., Milsted, A., Gray, J. & Chan, G. K.-L. Loop series expansions for tensor networks. Phys. Rev. Res.8, 013245 (2026)

    2026

  47. [91]

    Gray, J.et al.Tensor network loop cluster expan- sions for quantum many-body problems.arXiv preprint arXiv:2510.05647(2025)

    Pith/arXiv arXiv 2025

  48. [92]

    Rudolph, M. S. & Tindall, J. Simulating and sampling from quantum circuits with 2D tensor networks.arXiv preprint arXiv:2507.11424(2025)

    arXiv 2025