arxiv: 2508.10997 · v2 · submitted 2025-08-14 · 🪐 quant-ph · cond-mat.str-el

Reliable high-accuracy error mitigation for utility-scale quantum circuits

Dorit Aharonov , Ori Alberton , Itai Arad , Yosi Atia , Eyal Bairey , Matan Ben Dov , Asaf Berkovitch , Zvika Brakerski

show 31 more authors

Itsik Cohen Eran Fuchs Omri Golan Or Golan Barak D. Gur Ilya Gurwich Avieli Haber Rotem Haber Dorri Halbertal Yaron Itkin Barak A. Katzir Oded Kenneth Shlomi Kotler Roei Levi Eyal Leviatan Yotam Y. Lifshitz Adi Ludmer Shlomi Matityahu Ron Aharon Melcer Adiel Meyer Omrie Ovdat Aviad Panahi Gil Ron Ittai Rubinstein Gili Schul Tali Shnaider Maor Shutman Asif Sinay Tasneem Watad Assaf Zubida Netanel H. Lindner

This is my paper

Pith reviewed 2026-05-18 22:33 UTC · model grok-4.3

classification 🪐 quant-ph cond-mat.str-el

keywords quantum error mitigationQESEMquasi-probabilistic mitigationzero-noise extrapolationprobabilistic error cancellationutility-scale circuitsIsing modelvariational quantum eigensolver

0 comments p. Extension

The pith

QESEM resolves the performance-reliability tradeoff in quantum error mitigation by delivering rigorous accuracy guarantees with far lower overhead than probabilistic error cancellation.

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

The paper introduces QESEM, a characterization-based framework that applies quasi-probabilistic mitigation techniques to large quantum circuits while keeping computational cost manageable. This approach matters because current methods either lack accuracy guarantees, as with zero-noise extrapolation, or incur prohibitive overhead, as with probabilistic error cancellation. The authors demonstrate the framework on the largest unbiased error-mitigation experiment to date by simulating a kicked transverse-field Ising model on an IBM Heron processor and by running molecular variational quantum eigensolver circuits on both superconducting and trapped-ion hardware. If the central claim holds, it makes reliable, high-fidelity results from utility-scale circuits feasible on present-day devices without the runtime penalties of prior rigorous methods.

Core claim

QESEM is a rigorously grounded error mitigation and suppression framework that extracts a noise model through characterization and then applies quasi-probabilistic techniques to produce unbiased estimates at dramatically reduced overhead; the method is validated by simulating the kicked transverse field Ising model with far-from-Clifford parameters on an IBM Heron device and by executing molecular VQE circuits on both IBM Heron and IonQ devices, consistently outperforming multiple variants of zero-noise extrapolation while avoiding the cost of full probabilistic error cancellation.

What carries the argument

QESEM, the characterization-based quasi-probabilistic error mitigation framework that learns a noise model once and then suppresses and mitigates errors with accuracy guarantees but low overhead.

If this is right

Enables simulation of the kicked transverse-field Ising model at utility scale with unbiased mitigation on IBM Heron hardware.
Produces higher-accuracy molecular ground-state energies via VQE on both superconducting and trapped-ion processors than zero-noise extrapolation achieves.
Removes the prohibitive runtime cost of full probabilistic error cancellation while retaining its accuracy guarantees.
Supports concrete performance projections for near-term devices pursuing quantum advantage across diverse algorithms.

Where Pith is reading between the lines

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

The reduced overhead may allow hybrid quantum-classical loops to incorporate more shots or deeper circuits than previously practical.
If noise-model stability holds across device families, the same characterization pipeline could be reused for other circuit families beyond Ising and VQE.
Future hardware with faster calibration cycles could further lower the already-reduced overhead of the method.
The framework's emphasis on characterization stability points to a possible need for lightweight online monitoring techniques when circuits exceed current utility-scale sizes.

Load-bearing premise

The noise model learned from initial characterization remains accurate and stable throughout the full execution of the utility-scale circuit without additional recalibration.

What would settle it

If repeated runs of the same kicked Ising or VQE circuit with a fixed initial noise model produce mitigated results whose error grows beyond that of zero-noise extrapolation or requires mid-experiment recalibration to match exact values, the stability premise would be falsified.

Figures

Figures reproduced from arXiv: 2508.10997 by Adiel Meyer, Adi Ludmer, Asaf Berkovitch, Asif Sinay, Assaf Zubida, Aviad Panahi, Avieli Haber, Barak A. Katzir, Barak D. Gur, Dorit Aharonov, Dorri Halbertal, Eran Fuchs, Eyal Bairey, Eyal Leviatan, Gili Schul, Gil Ron, Ilya Gurwich, Itai Arad, Itsik Cohen, Ittai Rubinstein, Maor Shutman, Matan Ben Dov, Netanel H. Lindner, Oded Kenneth, Omrie Ovdat, Omri Golan, Or Golan, Ori Alberton, Roei Levi, Ron Aharon Melcer, Rotem Haber, Shlomi Kotler, Shlomi Matityahu, Tali Shnaider, Tasneem Watad, Yaron Itkin, Yosi Atia, Yotam Y. Lifshitz, Zvika Brakerski.

**Figure 1.** Figure 1: QESEM overview. (a) QESEM workflow and stages. Device characterization maps gate fidelities and identifies coherent errors, providing real-time calibration data. Noise-aware transpilation generates and evaluates alternative qubit mappings, operation sets, and measurement bases, selecting the variant that minimizes estimated QPU runtime, with optional parallelization to accelerate data collection. Error sup… view at source ↗

**Figure 2.** Figure 2: Unbiased mitigation of utility-scale Hamiltonian simulations with QESEM: (a) Kicked Ising circuit – The circuit includes three distinct layers of fractional RZZ two-qubit gates, sandwiched between RX and RZ single-qubit gate layers. (b) Device geometry – The algorithm ran on a 103-qubit graph embedded in ibm_marrakesh. The geometry was selected to have the minimum possible infidelity of the two-qubit gates… view at source ↗

**Figure 3.** Figure 3: The VQE benchmark results: (a) The nine patches used in the VQE benchmark, in different colors, on Marrakesh’s connectivity graph. (b) The variational ground state energy of H2O as obtained from the VQE circuit, for each qubit-patch and the final QESEM estimator given by the inverse-varianceweighted average. QESEM-mitigated values appear in blue, and noisy values in red. The ideal result is shown as a das… view at source ↗

**Figure 4.** Figure 4: The QPU runtime of QESEM in the generic kicked Ising benchmark, compared to various estimations, with active volume determined by Algorithm 4 with ϵLC = 0.03: QESEM’s QPU runtime (blue) is extracted from IBM’s reported workload usage [65] and is rescaled to a precision ϵ = 0.01 in each step via TQPU ∝ ϵ −2 , for ease of comparison to estimations; The empirical time estimate (orange) is based on a small sam… view at source ↗

**Figure 5.** Figure 5: Analytical extrapolation of QPU runtimes of QESEM for larger active volumes and with varying hardware properties, using the phenomenological model (13). In the two plots, we use ⟨O⟩ (ideal) C0 = 1, required accuracy ϵ = 0.01, and r = 1 unless specified otherwise. The tc = 0 lines correspond to QPUs without any controller delays. (a) Projections for superconducting qubits, using timescales close to those… view at source ↗

read the original abstract

Error mitigation is essential for unlocking the full potential of quantum algorithms and accelerating the timeline toward quantum advantage. As quantum hardware progresses to push the boundaries of classical simulation, efficient and robust error mitigation methods are becoming increasingly important for producing accurate and reliable outputs. However, existing error-mitigation approaches face a fundamental tradeoff between practical performance and reliability: heuristic methods such as zero-noise extrapolation (ZNE) enjoy faster runtime but lack accuracy guarantees, while rigorous techniques such as probabilistic error cancellation (PEC) provide unbiased estimates at prohibitive computational cost. We introduce a characterization-based, rigorously-grounded quantum error mitigation and error suppression framework (QESEM) that resolves this tradeoff by leveraging the accuracy guarantees of quasi-probabilistic mitigation with dramatically reduced overhead. We explain the innovative methods underlying QESEM and demonstrate its capabilities in the largest utility-scale error mitigation experiment based on an unbiased method. This experiment simulates the kicked transverse field Ising model with far-from-Clifford parameters on an IBM Heron device. We further validate QESEM's versatility across arbitrary quantum circuits and devices through high-accuracy error-mitigated molecular VQE circuits executed on IBM Heron and IonQ trapped-ion devices. Compared with multiple variants of the widely used zero-noise extrapolation method, QESEM consistently achieves higher accuracy while avoiding the prohibitive runtime overhead associated with PEC. These results mark a significant step forward in accuracy and reliability for running quantum circuits on current devices across diverse applications. Finally, we provide projections of QESEM's performance on near-term devices toward quantum advantage.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note private letter to a colleague

QESEM pairs device characterization with quasi-probabilistic correction to cut overhead while keeping unbiased estimates, and the experiments on Heron and IonQ show measurable accuracy gains over ZNE on large non-Clifford circuits.

read the letter

The core advance is a framework that first characterizes the device to learn a noise model, then derives quasi-probability weights from it for mitigation. This targets the gap between fast but heuristic ZNE and the high-cost unbiased PEC, and they apply it to arbitrary circuits rather than just specific ansatze. The experiments run the kicked Ising model with far-from-Clifford parameters on IBM Heron and also execute molecular VQE on both Heron and IonQ hardware, reporting higher accuracy than several ZNE variants at practical runtimes. That scale of demonstration is the strongest part of the work and gives concrete data points for people who need to run similar workloads now. The method itself builds directly on established quasi-probability ideas but packages the characterization step in a way that appears tailored for utility-scale use. The main soft spot is the stability of the learned noise model across the full duration of these long circuits. The results give final mitigated accuracies, yet there is no reported re-characterization at intervals, no drift monitoring, and no explicit check for state-dependent or time-varying effects that could shift the actual noise away from the model. If that mismatch occurs, the correction becomes biased even though the math assumes a fixed model. This is a practical rather than theoretical issue and matters most for the longest runs they highlight. The paper is aimed at researchers and practitioners who already work with error mitigation on superconducting and trapped-ion hardware and want a middle path between speed and guarantees. It supplies the kind of head-to-head numbers on real devices that are useful for deciding whether to try the approach. I would send it to peer review. The experimental claims are substantial enough to justify referee time, and the stability question is something reviewers can press on with specific requests for additional checks.

Referee Report

2 major / 1 minor

Summary. The paper introduces QESEM, a characterization-based quantum error mitigation and suppression framework that combines quasi-probabilistic methods with device-specific noise models to deliver unbiased estimates at substantially lower overhead than probabilistic error cancellation (PEC). It demonstrates the approach on utility-scale circuits, including kicked transverse-field Ising model simulations with far-from-Clifford parameters on an IBM Heron device and molecular VQE circuits on both IBM Heron and IonQ trapped-ion hardware, reporting higher accuracy than multiple ZNE variants while avoiding PEC-level runtime costs. Projections for near-term devices are also provided.

Significance. If the central claims hold, the work marks a meaningful advance by narrowing the longstanding gap between heuristic speed and rigorous unbiasedness in error mitigation. The scale of the reported experiments—the largest utility-scale demonstrations based on an unbiased method—together with cross-device validation on superconducting and trapped-ion platforms, would constitute a concrete step toward reliable execution of quantum algorithms beyond classical simulability. Explicit credit is due for the experimental scope and the attempt to ground performance in characterization rather than ad-hoc fitting.

major comments (2)

[Experimental validation sections] Experimental validation sections (kicked Ising model on Heron and VQE on Heron/IonQ): the unbiased character of the quasi-probabilistic correction rests on the assumption that the noise model obtained during characterization remains accurate and stable throughout the full-duration execution of the utility-scale circuits. The manuscript reports final mitigated accuracies but contains no time-stamped re-characterization, drift-monitoring data, or controlled comparisons with deliberately varied calibration intervals. If temporal or state-dependent drift exceeds the model’s capture range, the derived quasi-probability weights become mismatched and the mitigation converts from unbiased to systematically biased; this assumption is load-bearing for the reliability claims.
[Abstract and methods] Abstract and overhead discussion: the claim of “dramatically reduced overhead” relative to PEC is central to resolving the performance-reliability tradeoff, yet the text supplies limited quantitative detail on how overhead is defined and measured (e.g., whether characterization cost is amortized, how shot overhead is counted, and the precise baseline PEC implementation). Without these metrics, it is difficult to assess whether the reported runtime advantage is robust or device-specific.

minor comments (1)

[Methods] Clarify in the text or supplementary material whether the characterization step itself is performed once per device or per circuit instance, and how any state-preparation assumptions in the noise model are validated.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the positive assessment of our work and for the constructive comments that help strengthen the presentation of QESEM. We address each major comment point by point below, with a focus on clarifying assumptions, providing additional quantitative detail where possible, and indicating planned revisions.

read point-by-point responses

Referee: Experimental validation sections (kicked Ising model on Heron and VQE on Heron/IonQ): the unbiased character of the quasi-probabilistic correction rests on the assumption that the noise model obtained during characterization remains accurate and stable throughout the full-duration execution of the utility-scale circuits. The manuscript reports final mitigated accuracies but contains no time-stamped re-characterization, drift-monitoring data, or controlled comparisons with deliberately varied calibration intervals. If temporal or state-dependent drift exceeds the model’s capture range, the derived quasi-probability weights become mismatched and the mitigation converts from unbiased to systematically biased; this assumption is load-bearing for the reliability claims.

Authors: We agree that the stability of the noise model is a critical assumption for preserving unbiased estimates in QESEM. In the reported experiments, full characterization was performed immediately prior to circuit execution, and all utility-scale runs on each device were completed within a single calibration window to limit exposure to drift. The manuscript does not include explicit time-stamped re-characterization or controlled drift-monitoring data. We will add a dedicated subsection in the methods describing the characterization protocol, the time scales involved, and any internal consistency checks performed across repeated executions. This revision will also include a brief discussion of the assumption and its practical implications for the reported results. revision: partial
Referee: Abstract and overhead discussion: the claim of “dramatically reduced overhead” relative to PEC is central to resolving the performance-reliability tradeoff, yet the text supplies limited quantitative detail on how overhead is defined and measured (e.g., whether characterization cost is amortized, how shot overhead is counted, and the precise baseline PEC implementation). Without these metrics, it is difficult to assess whether the reported runtime advantage is robust or device-specific.

Authors: We appreciate this observation. Overhead is defined as the total number of shots (circuit executions) needed to reach a target statistical precision, with the one-time characterization cost amortized across all circuits executed under the same noise model. The PEC baseline employs the identical noise model for quasi-probability decomposition but lacks the additional suppression layer introduced in QESEM. We will revise the abstract and methods section to state these definitions explicitly, report the measured overhead factors for the kicked-Ising and VQE experiments, and specify the PEC implementation details used for comparison. revision: yes

Circularity Check

0 steps flagged

No significant circularity in derivation or claims

full rationale

The paper introduces QESEM as a characterization-based framework combining quasi-probabilistic mitigation with reduced overhead, supported by experimental demonstrations on IBM Heron and IonQ devices for kicked Ising and VQE circuits. Central claims rest on device-specific noise model learning and empirical accuracy measurements rather than any self-definitional reduction, fitted input renamed as prediction, or load-bearing self-citation chain. No equations or steps in the abstract or context reduce the mitigation result to its own inputs by construction; validation uses independent circuit executions and comparisons to ZNE variants. The noise stability assumption is a methodological limitation but does not create circularity per the enumerated patterns.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The framework relies on standard quantum channel assumptions and device characterization; no new particles or forces are introduced. Free parameters are limited to those arising from noise model fitting during characterization.

free parameters (1)

noise model parameters from characterization
Fitted to device data to enable the quasi-probabilistic correction; central to overhead reduction claim.

axioms (1)

domain assumption The quantum noise can be represented as a quasi-probability distribution that is learnable via characterization.
Invoked to justify the unbiased mitigation step.

pith-pipeline@v0.9.0 · 5987 in / 1220 out tokens · 47255 ms · 2026-05-18T22:33:25.248040+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

QESEM constructs quasi-probabilistic (QP) decompositions for the noisy operations... based on the data from the circuit characterization stage.
IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

We introduce QESEM... leveraging the accuracy guarantees of quasi-probabilistic mitigation with dramatically reduced overhead.

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Forward citations

Cited by 1 Pith paper

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

Mind the gaps: The fraught road to quantum advantage
quant-ph 2025-10 unverdicted novelty 4.0

The authors identify four transitions needed to reach fault-tolerant application-scale quantum computing from current NISQ devices.

Reference graph

Works this paper leans on

141 extracted references · 141 canonical work pages · cited by 1 Pith paper · 6 internal anchors

[1]

(E27) The invertibility of the design matrix follows from the invertibility of allDi, the transformation from layer noise to germ noise

Exact solution of linearized least squares equa- tion: The MSE cost function can be solved analytically by taking the logarithm of the cost function and equat- ing it to zero, i.e., log⟨ ⃗O⟩ = A⃗ γ⇒ ⃗ γ= A−1 log⟨ ⃗O⟩ . (E27) The invertibility of the design matrix follows from the invertibility of allDi, the transformation from layer noise to germ noise. T...

work page
[2]

(E26) Here, we discussed the inference, ignoring any Clifford two-qubit gate layers that may participate in the algorithm

Gradient descent for the non-linear MSE: Using the linearized least squares solution as the ini- tial condition, QESEM optimizes the parameters ⃗ γ further using the full non-linear MSE in Eq. (E26) Here, we discussed the inference, ignoring any Clifford two-qubit gate layers that may participate in the algorithm. If any such layers are present, they requ...

work page
[3]

Clifford two-qubit gate layers The set of characterization circuits executed by QESEM for Clifford two-qubit gate layers can be classified into two families:

work page
[4]

Circuits applying only one unique Clifford two-qubit gate layer

work page
[5]

The former provides sensitivity to all inter-layer learnable combinations of Pauli generator rates and amplifies all intra-layer amplifiable combinations [38, 108]

Circuits mixing more than one unique Clifford two- qubit gate layer. The former provides sensitivity to all inter-layer learnable combinations of Pauli generator rates and amplifies all intra-layer amplifiable combinations [38, 108]. The latter 46 amplifies inter-layer combinations that are built from intra- layer non-amplifiable, but learnable, ones [104...

work page
[6]

Device Familiarization

Qubit selection At the start of every QESEM run, the first calculation performed on the QPU is called "Device Familiarization" (DFAM) - a compact form of the characterization described in App. E, targeted to characterize the infidelity of all the 2-qubit gates on the device in layer context. Meaning, the effects of crosstalk and large layers are taken int...

work page
[7]

Circuit parallelization For small enough circuits that can be fitted multiple times onto the device, QESEM enables parallel execution of the circuit, as was demonstrated in Sec. IV. The number of patches is selected using the backend error per layered gate (EPLG) data [111], to optimize the value of T (EPLG(n × nq)) n . (F1) Inotherwords, sothatwegainmore...

work page
[8]

To isolate the tar- geted term, we construct a germG using a concate- nation of multiple Pauli DD pulses D1, D2, D3,

Concatenated Dynamical Decoupling (DD): DD sequences can suppress Pauli terms that anti- commute with the applied decoupling pulses, while commuting terms are preserved. To isolate the tar- geted term, we construct a germG using a concate- nation of multiple Pauli DD pulses D1, D2, D3, . . ., commuting with the targeted term[Di, Pt] = 0. As- suming the tw...

work page
[9]

For OR, the germ is: G(Pi) = Pi g Pi

Partial Twirling: Partial twirling involves averaging over randomized Pauli operators Pi that commute with the targeted term Pt, i.e., [Pi, Pt] = 0. For OR, the germ is: G(Pi) = Pi g Pi . (G2) For other coherent errors, we combine partial twirling with DD: G0 = g D1 g D1 , G(Pi) = Pi G0 Pi , where D1 is a Pauli operator that anti-commutes with g, i.e., {D...

work page
[10]

Here, we exemplify our protocol by calibrating aRZZ (π/6) on all the two-qubit gate pairs ofibm_torino

Calibration On the Heron devices, we use the partial twirling method to accurately measure and calibrate the over-rotation error. Here, we exemplify our protocol by calibrating aRZZ (π/6) on all the two-qubit gate pairs ofibm_torino. All the two- qubit gates in the device are divided into three layers. The Germ (G2) (applied to each pair in the layer) is ...

work page
[11]

2 of the main text, we considered the average magnetization for the different steps

Single-qubit Magnetization In Fig. 2 of the main text, we considered the average magnetization for the different steps. Here, in Fig. H1, we present the QESEM mitigation results per site for steps one, three, five, and seven

work page
[12]

The weight-n observable is defined as ⟨Wn⟩ = 1 Nn X ⟨i0,i1,...,in−1⟩ ⟨Zi0

Heavy-weight observables In addition to the magnetization expectation values, we can consider averages of higher-weight observables. The weight-n observable is defined as ⟨Wn⟩ = 1 Nn X ⟨i0,i1,...,in−1⟩ ⟨Zi0 . . . Zin−1 ⟩, where the sum runs over all the length-n chains embedded in the device’s geometry, andNn is the number of such chains. The results are ...

work page
[13]

ZNE error mitigation We ran the same circuits using several error mitigation solutions freely available on IBM hardware

work page
[14]

E1), and the noise was amplified locally (for each gate) via the gate-folding method

Twirled ZNE.The RZZ gates were partially Pauli- twirled (see Sec. E1), and the noise was amplified locally (for each gate) via the gate-folding method. Specifically, anRZZ gate with noise leveln is realized as Rn ZZ = RZZ (RZZ X0RZZ X0) n−1 2 , 52 0.0 0.2 0.4 0.6<Z> (a) Step - 1 0.0 0.2 0.4<Z> (b) Step - 3 0.0 0.2 0.4 0.6<Z> (c) Step - 5 Qubits 0.0 0.5 1....

work page
[15]

This performs {1, 3, 5} gate- folding noise amplification, and the zero-noise Pauli observables are extracted via exponential fitting

qiskit – fractional ZNE.We ran the circuits via the qiskitestimator, with the ZNE option turned on (with default options). This performs {1, 3, 5} gate- folding noise amplification, and the zero-noise Pauli observables are extracted via exponential fitting. We then regularized the noiseless observables and aver- aged for the magnetization. TREX measuremen...

work page
[16]

ZNE was performed via the qiskit estimator in a similar way to the fractional ZNE

qiskit – Clifford ZNE.Here, we compiled the same logical circuits to have only Clifford (CZ) two-qubit gates. ZNE was performed via the qiskit estimator in a similar way to the fractional ZNE

work page
[17]

The gates are characterized, and the noise is amplified with Pauli insertion

qiskit – PEA.Here, we used the qiskit estimator Pauli Error Amplification (PEA) [29] option to miti- gate the noise. The gates are characterized, and the noise is amplified with Pauli insertion. Again, we used the default parameters, consisting of {1, 2, 3} noise levels. Measurement errors are mitigated with TREX. Here, due to a qiskit limitation, we ran ...

work page
[18]

Classical simulation details In order to simulate the noiseless values of the magne- tization we run on the heavy-hex lattice, we use the re- centlyintroducedmethodsforapproximatecontractionand compressionoftensornetworksusingthebelief-propagation (BP) method [58–60]. Specifically, we follow the approach described in [58] for Schrödinger evolution of a pr...

work page
[19]

noise factors: 1, 3, 5

work page
[20]

The experiment executed 320512 shots over 11.5 minutes of QPU time and yielded a large error bar,∼ 0.1, approx- imately five times larger than that of QESEM

extrapolator: best of exponential, linear (whichever produces the smaller error bar). The experiment executed 320512 shots over 11.5 minutes of QPU time and yielded a large error bar,∼ 0.1, approx- imately five times larger than that of QESEM. almost five 55 102 103 Number of circuits Figure I1: The optimal number of circuits per measurement base in the V...

work page
[21]

layer- ing

IBM Eagle To validate QESEM’s hardware-adaptability, we show large-scale Hamiltonian simulations performed on IBM’s 127-qubit fixed-frequency superconducting Eagle QPUs. While Eagle is also an IBM device, they are based on microwave-driven, echoed cross-resonance gates [114], which result in a very different noise model. Cross- resonance gates have more c...

work page
[22]

To the best of our knowledge, these ex- periments are the first large-scale demonstrations of error mitigation on trapped-ion hardware based on an unbiased method

IonQ Aria To demonstrate QESEM’s hardware-agnostic capabil- ities, we executed several quantum circuits on IonQ’s trapped-ion QPUs. To the best of our knowledge, these ex- periments are the first large-scale demonstrations of error mitigation on trapped-ion hardware based on an unbiased method. They were performed via both IonQ’s direct in- terface and Am...

work page
[23]

A framework for quantum advantage,

Olivia Lanes, Mourad Beji, Antonio D. Corcoles, Con- stantin Dalyac, Jay M. Gambetta, Loïc Henriet, Ali Javadi-Abhari, Abhinav Kandala, Antonio Mezzacapo, Christopher Porter, Sarah Sheldon, John Watrous, ChristaZoufal, AlexandreDauphin, andBorjaPeropadre, “A framework for quantum advantage,” arXiv preprint arXiv:2506.20658 (2025), arXiv:2506.20658 [quant-ph]

work page arXiv 2025
[24]

On the importance of er- ror mitigation for quantum computation,

Dorit Aharonov, Ori Alberton, Itai Arad, Yosi Atia, Eyal Bairey, Zvika Brakerski, Itsik Cohen, Omri Golan, Ilya Gurwich, Oded Kenneth, Eyal Leviatan, Ne- tanel H. Lindner, Ron Aharon Melcer, Adiel Meyer, Gili Schul, and Maor Shutman, “On the importance of er- ror mitigation for quantum computation,” arXiv preprint arXiv:2503.17243 (2025), 10.48550/arXiv.2...

work page doi:10.48550/arxiv.2503.17243 2025
[25]

Scheme for reducing decoherence in quantum computer memory,

Shor and Peter W., “Scheme for reducing decoherence in quantum computer memory,” PhRvA52, R2493–R2496 (1995)

work page 1995
[26]

Error correcting codes in quantum theory,

A. M. Steane, “Error correcting codes in quantum theory,” Phys. Rev. Lett.77, 793–797 (1996)

work page 1996
[27]

Fault-Tolerant Quantum Computation with Constant Error Rate,

Dorit Aharonov and Michael Ben-Or, “Fault-Tolerant Quantum Computation with Constant Error Rate,” https://doi.org/10.1137/S0097539799359385 38, 1207– 1282 (2008), arXiv:9906129 [quant-ph]

work page doi:10.1137/s0097539799359385 2008
[28]

Fault-TolerantPostselectedQuantumComputa- tion: Threshold Analysis,

E.Knill,“Fault-TolerantPostselectedQuantumComputa- tion: Threshold Analysis,” (2004), arXiv:0404104 [quant- ph]

work page 2004
[29]

Hybrid Quantum-Classical Algorithms and Quan- tum Error Mitigation,

Endo Suguru, Cai Zhenyu, Benjamin Simon C., and Yuan Xiao, “Hybrid Quantum-Classical Algorithms and Quan- tum Error Mitigation,” Journal of the Physical Society of Japan 90, 032001 (2021)

work page 2021
[30]

Quantum error miti- gation,

Zhenyu Cai, Ryan Babbush, Simon C. Benjamin, Sug- uru Endo, William J. Huggins, Ying Li, Jarrod R. Mc- Clean, and Thomas E. O’Brien, “Quantum error miti- gation,” Reviews of Modern Physics95, 045005 (2023), arXiv:2210.00921

work page arXiv 2023
[31]

Quantum Error Mitigation as a Universal Er- ror Reduction Technique: Applications from the NISQ to the Fault-Tolerant Quantum Computing Eras,

Yasunari Suzuki, Suguru Endo, Keisuke Fujii, and Yuuki Tokunaga, “Quantum Error Mitigation as a Universal Er- ror Reduction Technique: Applications from the NISQ to the Fault-Tolerant Quantum Computing Eras,” PRX Quantum 3, 010345 (2022)

work page 2022
[32]

Practical quan- tum advantage on partially fault-tolerant quantum com- puter,

Riki Toshio, Yutaro Akahoshi, Jun Fujisaki, Hirotaka Os- hima, Shintaro Sato, and Keisuke Fujii, “Practical quan- tum advantage on partially fault-tolerant quantum com- puter,” (2024), arXiv:2408.14848

work page arXiv 2024
[33]

Error Mitigation for Universal Gates on Encoded Qubits,

Christophe Piveteau, David Sutter, Sergey Bravyi, Jay M. Gambetta, and Kristan Temme, “Error Mitigation for Universal Gates on Encoded Qubits,” Physical Review Letters 127, 200505 (2021), arXiv:2103.04915

work page arXiv 2021
[34]

Zero Noise Extrapolation on Logical Qubits by Scaling the Er- ror Correction Code Distance,

Misty A. Wahl, Andrea Mari, Nathan Shammah, William J. Zeng, and Gokul Subramanian Ravi, “Zero Noise Extrapolation on Logical Qubits by Scaling the Er- ror Correction Code Distance,” Proceedings - 2023 IEEE InternationalConferenceonQuantumComputingandEn- gineering, QCE 20231, 888–897 (2023), arXiv:2304.14985

work page arXiv 2023
[35]

Zhang, H

Aosai Zhang, Haipeng Xie, Yu Gao, Jia-Nan Yang, Zehang Bao, Zitian Zhu, Jiachen Chen, Ning Wang, Chuanyu Zhang, Jiarun Zhong, Shibo Xu, Ke Wang, Yaozu Wu, Feitong Jin, Xuhao Zhu, Yiren Zou, Ziqi Tan, Zhengyi Cui, Fanhao Shen, Tingting Li, Yihang Han, Yiyang He, Gongyu Liu, Jiayuan Shen, Han Wang, Yanzhe Wang, Hang Dong, Jinfeng Deng, Hekang Li, Zhen Wang,...

work page arXiv 2025
[37]

The future of quantum computing with superconducting qubits,

Sergey Bravyi, Oliver Dial, Jay M Gambetta, Darío Gil, and Zaira Nazario, “The future of quantum computing with superconducting qubits,” Journal of Applied Physics 132, 160902 (2022)

work page 2022
[38]

Fundamental limits of quantum error mitiga- tion,

Ryuji Takagi, Suguru Endo, Shintaro Minagawa, and Mile Gu, “Fundamental limits of quantum error mitiga- tion,” npj Quantum Information 2022 8:18, 1–11 (2022), arXiv:2109.04457

work page arXiv 2022
[39]

Univer- sal sampling lower bounds for quantum error mitigation,

Ryuji Takagi, Hiroyasu Tajima, and Mile Gu, “Univer- sal sampling lower bounds for quantum error mitigation,” Phys. Rev. Lett.131, 210602 (2023)

work page 2023
[40]

Universal cost bound of quantum error mitigation based on quantum estimation theory,

Kento Tsubouchi, Takahiro Sagawa, and Nobuyuki Yosh- ioka, “Universal cost bound of quantum error mitigation based on quantum estimation theory,” Phys. Rev. Lett. 131, 210601 (2023)

work page 2023
[41]

Exponen- tially tighter bounds on limitations of quantum er- ror mitigation,

Yihui Quek, Daniel Stilck França, Sumeet Khatri, Jo- hannes Jakob Meyer, and Jens Eisert, “Exponen- tially tighter bounds on limitations of quantum er- ror mitigation,” Nature Physics 2024 , 1–11 (2024), arXiv:2210.11505

work page arXiv 2024
[42]

Zimborás et al., Myths around quantum computation before full fault tolerance: What no-go theorems rule out and what they don’t (2025), arXiv:2501.05694 [quant- ph]

Zoltán Zimborás, Bálint Koczor, Zoë Holmes, Elsi-Mari Borrelli, András Gilyén, Hsin-Yuan Huang, Zhenyu Cai, Antonio Acín, Leandro Aolita, Leonardo Banchi, Fer- nando G. S. L. Brandão, Daniel Cavalcanti, Toby Cu- bitt, Sergey N. Filippov, Guillermo García-Pérez, John Goold, Orsolya Kálmán, Elica Kyoseva, Matteo A. C. Rossi, Boris Sokolov, Ivano Tavernelli,...

work page arXiv 2025
[43]

Quantum Computing for High-Energy Physics: State of the Art and Challenges,

Alberto Di Meglio, Karl Jansen, Ivano Tavernelli, Con- stantia Alexandrou, Srinivasan Arunachalam, Chris- tian W. Bauer, Kerstin Borras, Stefano Carrazza, Arianna Crippa, Vincent Croft, Roland De Putter, Andrea Del- gado, Vedran Dunjko, Daniel J. Egger, Elias Fernández- Combarro, Elina Fuchs, Lena Funcke, Daniel González- Cuadra, Michele Grossi, Jad C. Ha...

work page 2024
[44]

Quantum simulations of hadron dy- namics in the Schwinger model using 112 qubits,

Roland C. Farrell, Marc Illa, Anthony N. Ciavarella, and Martin J. Savage, “Quantum simulations of hadron dy- namics in the Schwinger model using 112 qubits,” Physical Review D109, 114510 (2024), arXiv:2401.08044

work page arXiv 2024
[45]

First-order phase transition of the Schwinger model with a quantum computer,

Takis Angelides, Pranay Naredi, Arianna Crippa, Karl Jansen, Stefan Kühn, Ivano Tavernelli, and Derek S. Wang, “First-order phase transition of the Schwinger model with a quantum computer,” npj Quantum Infor- mation 2025 11:111, 1–12 (2025)

work page 2025
[46]

Unveiling clean two- dimensional discrete time quasicrystals on a digital quan- tum computer,

Kazuya Shinjo, Kazuhiro Seki, Tomonori Shirakawa, Rong-Yang Sun, and Seiji Yunoki, “Unveiling clean two- dimensional discrete time quasicrystals on a digital quan- tum computer,” (2024), arXiv:2403.16718

work page arXiv 2024
[47]

Error mitigation extends the computational reach of a noisy quantum processor,

Abhinav Kandala, Kristan Temme, Antonio D. Córcoles, Antonio Mezzacapo, Jerry M. Chow, and Jay M. Gam- betta, “Error mitigation extends the computational reach of a noisy quantum processor,” Nature 2019 567:7749567, 491–495 (2019)

work page 2019
[48]

Coupled cluster downfolding theory in simulations of chemical systems on quantum hardware,

Nicholas P. Bauman, Muqing Zheng, Chenxu Liu, Nathan M. Myers, Ajay Panyala, Bo Peng, Ang Li, and Karol Kowalski, “Coupled cluster downfolding theory in simulations of chemical systems on quantum hardware,” (2025), arXiv:2507.01199 [quant-ph]

work page arXiv 2025
[49]

Quantum computa- tional chemistry,

Sam McArdle, Suguru Endo, Alán Aspuru-Guzik, Si- mon C. Benjamin, and Xiao Yuan, “Quantum computa- tional chemistry,” Reviews of Modern Physics92, 015003 (2020), arXiv:1808.10402

work page arXiv 2020
[50]

Quantum supremacy using a programmable super- conducting processor,

Frank Arute, Kunal Arya, Ryan Babbush, Dave Bacon, Joseph C. Bardin, Rami Barends, Rupak Biswas, Ser- gio Boixo, Fernando G.S.L. Brandao, David A. Buell, BrianBurkett, YuChen, ZijunChen, BenChiaro, Roberto Collins, William Courtney, Andrew Dunsworth, Ed- ward Farhi, Brooks Foxen, Austin Fowler, Craig Gidney, Marissa Giustina, Rob Graff, Keith Guerin, Stev...

work page 2019
[51]

Evidence for the util- ity of quantum computing before fault tolerance,

Youngseok Kim, Andrew Eddins, Sajant Anand, Ken Xuan Wei, Ewout van den Berg, Sami Rosenblatt, Hasan Nayfeh, Yantao Wu, Michael Zaletel, Kristan Temme, and Abhinav Kandala, “Evidence for the util- ity of quantum computing before fault tolerance,” Nature 2023 618:7965618, 500–505 (2023)

work page 2023
[52]

Constructive inter- ference at the edge of quantum ergodic dynamics,

Dmitry A. Abanin, Rajeev Acharya, Laleh Aghababaie- Beni, Georg Aigeldinger, Ashok Ajoy, Ross Alcaraz, Igor Aleiner, Trond I. Andersen, Markus Ansmann, Frank Arute, Kunal Arya, Abraham Asfaw, Nikita Astrakhant- sev, Juan Atalaya, Ryan Babbush, Dave Bacon, Brian Ballard, Joseph C. Bardin, Christian Bengs, Andreas Bengtsson, Alexander Bilmes, Sergio Boixo, ...

work page arXiv 2025
[53]

Digital quantum magnetism on a trapped-ion quantum computer

Reza Haghshenas, Eli Chertkov, Michael Mills, Wil- helm Kadow, Sheng-Hsuan Lin, Yi-Hsiang Chen, Chris Cade, Ido Niesen, Tomislav Begušić, Manuel S. Rudolph, Cristina Cirstoiu, Kevin Hemery, Conor Mc Keever, Michael Lubasch, Etienne Granet, Charles H. Baldwin, John P. Bartolotta, Matthew Bohn, Julia Cline, Matthew DeCross, Joan M. Dreiling, Cameron Foltz, ...

work page internal anchor Pith review Pith/arXiv arXiv 2025
[54]

Multi-exponential error extrapolation and combining error mitigation techniques for NISQ applica- tions,

Zhenyu Cai, “Multi-exponential error extrapolation and combining error mitigation techniques for NISQ applica- tions,” npj Quantum Information 2021 7:17, 1–12 (2021), arXiv:2007.01265

work page arXiv 2021
[55]

Gambetta

Kristan Temme, Sergey Bravyi, and Jay M. Gam- betta, “Error Mitigation for Short-Depth Quantum Cir- cuits,” Physical Review Letters 119, 180509 (2017), arXiv:1612.02058

work page internal anchor Pith review Pith/arXiv arXiv 2017
[56]

Practical Quantum Error Mitigation for Near-Future Applications,

Suguru Endo, Simon C Benjamin, and Ying Li, “Practical Quantum Error Mitigation for Near-Future Applications,” Phys. Rev. X8, 031027 (2018)

work page 2018
[57]

Mitiq: A software package for error mitigation on noisy quantum computers,

Ryan LaRose, Andrea Mari, Sarah Kaiser, Peter J. Kar- alekas, Andre A. Alves, Piotr Czarnik, Mohamed El Man- douh, Max H. Gordon, Yousef Hindy, Aaron Robertson, Purva Thakre, Misty Wahl, Danny Samuel, Rahul Mis- tri, Maxime Tremblay, Nick Gardner, Nathaniel T. Ste- men, Nathan Shammah, and William J. Zeng, “Mitiq: A software package for error mitigation o...

work page arXiv 2022
[58]

Error mitigation with Clifford quantum- circuit data,

Piotr Czarnik, Andrew Arrasmith, Patrick J. Coles, and Lukasz Cincio, “Error mitigation with Clifford quantum- circuit data,” Quantum5 (2020), 10.22331/q-2021-11-26- 592, arXiv:2005.10189v3

work page doi:10.22331/q-2021-11-26- 2020
[59]

Practical Quantum Error Mitigation for Near-Future Applications

SuguruEndo, SimonC.Benjamin, andYingLi,“Practical Quantum Error Mitigation for Near-Future Applications,” Physical Review X8, 031027 (2018), arXiv:1712.09271

work page internal anchor Pith review Pith/arXiv arXiv 2018
[60]

Probabilistic error cancellation with sparse Pauli–Lindblad models on noisy quantum proces- sors,

Ewout van den Berg, Zlatko K. Minev, Abhinav Kandala, and Kristan Temme, “Probabilistic error cancellation with sparse Pauli–Lindblad models on noisy quantum proces- sors,” Nature Physics 2023 19:819, 1116–1121 (2023), arXiv:2201.09866

work page arXiv 2023
[61]

Efficiently improving the performance of noisy quantum computers,

Samuele Ferracin, Akel Hashim, Jean Loup Ville, Ravi Naik, Arnaud Carignan-Dugas, Hammam Qassim, Alexis Morvan, David I. Santiago, Irfan Siddiqi, and Joel J. Wallman, “Efficiently improving the performance of noisy quantum computers,” Quantum 8, 1410 (2024), arXiv:2201.10672v5

work page arXiv 2024
[62]

Effective quantum volume, fi- delityandcomputationalcostofnoisyquantumprocessing experiments,

Kostyantyn Kechedzhi, Sergei V. Isakov, Salvatore Man- drà, Benjamin Villalonga, Xiao-Si Mi, Sergio Boixo, and Vadim N. Smelyanskiy, “Effective quantum volume, fi- delityandcomputationalcostofnoisyquantumprocessing experiments,” Future Generation Computer Systems153, 431–441 (2024)

work page 2024
[63]

Exper- imental Pauli-frame randomization on a superconducting qubit,

Matthew Ware, Guilhem Ribeill, Diego Ristè, Colm A Ryan, Blake Johnson, and Marcus P da Silva, “Exper- imental Pauli-frame randomization on a superconducting qubit,” Phys. Rev. A103, 42604 (2021)

work page 2021
[64]

Simulating physics with comput- ers,

Richard P. Feynman, “Simulating physics with comput- ers,” International Journal of Theoretical Physics21, 467– 488 (1982)

work page 1982
[65]

Universal quantum simulators,

Seth Lloyd, “Universal quantum simulators,” Science273, 1073–1078 (1996)

work page 1996
[66]

Efficient quantum algorithms for sim- ulating sparse hamiltonians,

Dominic W. Berry, Graeme Ahokas, Richard Cleve, and Barry C. Sanders, “Efficient quantum algorithms for sim- ulating sparse hamiltonians,” Communications in Mathe- matical Physics270, 359–371 (2007)

work page 2007
[67]

Chemical basis of trotter-suzuki errors in quantum chemistry simulation,

Ryan Babbush, Jarrod R. McClean, Dave Wecker, Alán Aspuru-Guzik, and Nathan Wiebe, “Chemical basis of trotter-suzuki errors in quantum chemistry simulation,” Physical Review A91, 022311 (2015)

work page 2015
[68]

Quantum computing in the NISQ era and beyond,

John Preskill, “Quantum computing in the NISQ era and beyond,” Quantum2, 79 (2018)

work page 2018
[69]

Time-crystalline eigenstate order on a quantum processor,

Xiao Mi, Matteo Ippoliti, Chris Quintana, Ami Greene, Zijun Chen, Jonathan Gross, Frank Arute, Kunal Arya, Juan Atalaya, Ryan Babbush, et al., “Time-crystalline eigenstate order on a quantum processor,” Nature601, 531–536 (2022)

work page 2022
[70]

Noise- resilientedgemodesonachainofsuperconductingqubits,

Xiao Mi, Michael Sonner, M Yuezhen Niu, Kenneth W Lee, Brooks Foxen, Rajeev Acharya, Igor Aleiner, Trond I Andersen, Frank Arute, Kunal Arya, et al., “Noise- resilientedgemodesonachainofsuperconductingqubits,” Science 378, 785–790 (2022)

work page 2022
[71]

Error-mitigated simulation of quantum many-body scars on quantum computers with pulse-level control,

I-Chi Chen, Benjamin Burdick, Yongxin Yao, Peter P Orth, and Thomas Iadecola, “Error-mitigated simulation of quantum many-body scars on quantum computers with pulse-level control,” Physical Review Research4, 043027 (2022). 62

work page 2022
[72]

Exact time-correlation functions of quan- tum ising chain in a kicking transversal magnetic field: Spectral analysis of the adjoint propagator in heisenberg picture,

Tomaž Prosen, “Exact time-correlation functions of quan- tum ising chain in a kicking transversal magnetic field: Spectral analysis of the adjoint propagator in heisenberg picture,” Progress of Theoretical Physics Supplement139, 191–203 (2000)

work page 2000
[73]

A floquet model for the many-body localization transi- tion,

Liangsheng Zhang, Vedika Khemani, and David A. Huse, “A floquet model for the many-body localization transi- tion,” Phys. Rev. B94, 224202 (2016)

work page 2016
[74]

Probing non-equilibrium topological or- der on a quantum processor,

M. Will, T. A. Cochran, E. Rosenberg, B. Jobst, N. M Eassa, P. Roushan, M. Knap, A. Gammon-Smith, and F. Pollmann, “Probing non-equilibrium topological or- der on a quantum processor,” (2025), arXiv:2501.18461 [quant-ph]

work page arXiv 2025
[75]

Large-scale simulations of floquet physics on near-term quantum computers,

Timo Eckstein, Refik Mansuroglu, Piotr Czarnik, Jian- Xin Zhu, Michael J. Hartmann, Lukasz Cincio, Andrew T. Sornborger, and Zoë Holmes, “Large-scale simulations of floquet physics on near-term quantum computers,” npj Quantum Information10, 84 (2024)

work page 2024
[76]

Entan- glement spreading in a minimal model of maximal many- body quantum chaos,

Bruno Bertini, Pavel Kos, and Toma ž Prosen, “Entan- glement spreading in a minimal model of maximal many- body quantum chaos,” Phys. Rev. X9, 021033 (2019)

work page 2019
[77]

Fast classi- cal simulation of evidence for the utility of quantum com- puting before fault tolerance,

Tomislav Begušić and Garnet Kin-Lic Chan, “Fast classi- cal simulation of evidence for the utility of quantum com- puting before fault tolerance,” (2023), arXiv:2306.16372 [quant-ph]

work page arXiv 2023
[78]

A polynomial-time classical algo- rithmfornoisyrandomcircuitsampling,

Dorit Aharonov, Xun Gao, Zeph Landau, Yunchao Liu, and Umesh Vazirani, “A polynomial-time classical algo- rithmfornoisyrandomcircuitsampling,” in Proceedings of the 55th Annual ACM Symposium on Theory of Comput- ing, STOC 2023 (Association for Computing Machinery, New York, NY, USA, 2023) p. 945–957

work page 2023
[79]

Rudolph, M

Armando Angrisani, Alexander Schmidhuber, Manuel S. Rudolph, M. Cerezo, Zoë Holmes, and Hsin-Yuan Huang, “Classically estimating observables of noiseless quantum circuits,” (2024), arXiv:2409.01706 [quant-ph]

work page arXiv 2024
[80]

Beguši´c, J

Tomislav Begušić, Johnnie Gray, and Garnet Kin-Lic Chan, “Fast and converged classical simulations of evidence for the utility of quantum computing before fault tolerance,” Science Advances10, eadk4321 (2024), https://www.science.org/doi/pdf/10.1126/sciadv.adk4321

work page doi:10.1126/sciadv.adk4321 2024
[81]

Efficient tensor network simulation of ibm’s eagle kicked ising experiment,

Joseph Tindall, Matthew Fishman, E. Miles Stoudenmire, and Dries Sels, “Efficient tensor network simulation of ibm’s eagle kicked ising experiment,” PRX Quantum5, 010308 (2024)

work page 2024

Showing first 80 references.