Pith. sign in

REVIEW 3 major objections 4 minor 39 references

Breaking Memory Bottlenecks in Quantum Control Systems for More Precise Experiments and Higher Throughput Computing

T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Ant-Q claims that a DRAM-plus-BRAM memory hierarchy can pipeline quantum circuit execution on a commodity FPGA control board, cutting circuit-loading and readout overhead from up to 1417 percent of execution time to near zero while…

desk verdict A genuinely useful memory-hierarchy design for FPGA-based quantum control that claims full-system results for a downlink path the paper itself says is not yet integrated. read the letter →

arxiv 2608.06318 v1 pith:RI324AQI submitted 2026-08-06 cs.AR cs.ETquant-ph

Yicheng Guang , Neel Vora , Yilun Xu , Yueqi Chen , Gang Huang This is my paper
classification cs.ARcs.ETquant-ph
keywords quantumcontrolsystemmemoryhierarchyDRAMBRAMping-pongbufferpipelineworkflowrandomizedbenchmarkingearlytermination
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

Quantum control systems today work like punched-card computers: a circuit is loaded, run to completion, its readout is sent back, and only then does the next circuit start. Ant-Q claims to break the memory bottleneck that forces this workflow by using the board's DRAM as a large command and readout pool, with BRAM kept as a fast cache, joined by ping-pong buffers that absorb DRAM's non-deterministic latency. If the claim holds, control systems on commodity RFSoCs can run much deeper experiments—randomized benchmarking at thousands of gates, repeated-Ramsey noise spectroscopy with dense shot counts—and can keep the quantum processor busy instead of idle during classical transfers. Ant-Q also uses the streamed readout to terminate noise-dominated validation circuits early, before their full shot budget is spent. The paper reports overhead falling from $22.90\%$ to $1417.05\%$ of execution time down to near zero and a $42.98\%$ batch-throughput improvement, while noting that the full downlink path is not yet integrated and is still undergoing stress testing.

What carries the argument

The load-bearing mechanism is a memory hierarchy with DRAM as the main pool and BRAM as a cache, connected by ping-pong buffers: each DSP's BRAM bank is split into two regions, one being drained by the controller while the other is filled from DRAM, with Write-Done and Read-Done handshakes governing the switch. The downlink removes padding bytes and uses a qMap bitmap to lay out only active per-DSP command pipelines contiguously in DRAM, so unused qubit channels do not waste transfer bandwidth. The uplink aggregates tagged IQ readout into a shared ping-pong BRAM buffer and drains it into a DRAM ring buffer, decoupling host-side transfer from execution. A host-side monitor computes total variation distance and Hellinger fidelity at ten-percent checkpoints and sends a Stop signal back through a cID-tagged control path when results have converged to an unreadable distribution.

What would settle it

Run the complete Ant-Q downlink and uplink on a ZCU216 executing a 5101-Clifford 1Q RB sequence and a 30-circuit batch, instrumenting the DSP start and stop times: if the DSP ever waits for a ping-pong half to fill, or the measured inter-circuit gap exceeds the declared deterministic bound, the near-zero-overhead and deep-circuit claims fail.

Watch

Extended reading notes

Core claim

The central claim is that DRAM can be brought into the real-time control path of FPGA-based quantum control systems without breaking deterministic timing, if BRAM is repurposed as a cache and the two are joined by ping-pong buffers. On a ZCU216 RFSoC board, Ant-Q stores circuit commands and readout results in 4 GB of PL-side DRAM and streams them into per-DSP BRAM banks in alternating halves, so the next circuit is already resident when the current one finishes. The paper reports this supports 1Q randomized benchmarking at 5101 Clifford gates and 2Q RB at 500 gates, repeated-Ramsey noise spectroscopy at $5\times 10^4$ shots with $2\,\mathrm{ms}$ spacing, and charge-parity monitoring sampled every $6\,\mu\mathrm{s}$ for 8000 samples; it also reduces circuit-loading and readout-uplink overhead from $22.90\%$ to $1417.05\%$ of execution time to near zero and raises batch throughput by $42.98\%$. The full downlink path is not yet integrated into QubiC 3.0; the paper states it is undergoing stress testing.

Load-bearing premise

The load-bearing premise is that the not-yet-integrated circuit-download path can keep the processor supplied with commands on time; the paper says that path is still undergoing stress testing.

Editorial extensions

If this is right

  • Deep randomized benchmarking sequences that exceed BRAM capacity, such as 5101-gate 1Q RB and 500-gate 2Q RB, become executable on a single commodity control board.
  • Long-duration, high-sampling-rate noise experiments such as repeated Ramsey spectroscopy and charge-parity monitoring no longer exhaust on-board memory, because readout streams to DRAM and then the host during execution.
  • Short computational circuits, where classical overhead is largest, see their loading and uplink overhead drop to near zero, so quantum-processor idle time shrinks correspondingly.
  • Noise-dominated validation circuits can be stopped after roughly 30 to 50 percent of their shots when the streamed distribution has converged to an unreadable one, saving quantum-processor time without waiting for the full budget.
  • Batch throughput on a mixed workload of 30 circuits improves by about 43 percent, with the gain concentrated in circuits whose execution time is short relative to transfer time.

Reading between the lines

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

  • Editorial extension: if the downlink passes its stress tests, the same DRAM-plus-BRAM ping-pong pattern could be ported to other FPGA control systems with punched-card workflows, since the design depends only on board-level DRAM and BRAM resources.
  • Editorial extension: the paper evaluates executability and throughput, not physical estimation precision; a natural next experiment is to compare 1/f noise-spectroscopy fits and RB decay-curve variance with and without Ant-Q's deterministic timing on a real qubit.
  • Editorial extension: the host-side early-termination rule requires a known ideal distribution, so it applies to validation circuits only; a distribution-free stopping rule would extend the mechanism to exploratory experiments.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 4 minor

Summary. The paper presents Ant-Q, a DRAM-BRAM memory-hierarchy design for FPGA-based quantum control systems. Ant-Q decouples circuit downlink, execution, and readout uplink to replace the punched-card workflow with a pipeline, using ping-pong buffering to hide DRAM non-determinism. The authors evaluate Ant-Q against the QubiC2.0 baseline on a ZCU216 board, reporting support for deep 1Q and 2Q randomized benchmarking circuits, reduction of circuit-loading and readout-uplink overhead from 22.90%–1417.05% to near zero, a 42.98% batch throughput improvement, and effective early termination of noise-dominated circuits.

Significance. If the claims hold, Ant-Q addresses a real practical bottleneck in quantum control systems: limited BRAM capacity forces idle QPU time during circuit loading and readout transfer. The paper's use of commodity RFSoC hardware, open-source benchmarks, and a direct comparison to an external baseline (QubiC2.0) is appropriate, and the design ideas—DRAM as a main memory pool, BRAM as a cache, and decoupled data paths—are plausible and of interest to the quantum-control and FPGA systems communities. The throughput and deep-circuit claims are not derived from fitted parameters; they are measured or benchmarked quantities, which is a strength. However, the current manuscript cannot be fully endorsed because a load-bearing part of the evaluation rests on an implementation component whose integration status is stated inconsistently.

major comments (3)
  1. [IV] Section IV states that 'The Ant-Q Downlink is currently undergoing stress testing and will be integrated subsequently,' yet Tables 2, 3, and 4 report results under the configuration 'Ant-Q Downlink + Uplink.' The last two rows of Table 2 (1Q RB with 5101 Clifford gates and 2Q RB with 500 Clifford gates) and the full Ant-Q columns of Tables 3 and 4 depend on the downlink to achieve the claimed near-zero overhead and deep-circuit support. The paper does not state that these rows are simulations, projections, or measurements from a test harness. This is an internal inconsistency that affects the central claims in the abstract. Please clarify which entries are hardware measurements from a fully integrated system, which are from the integrated uplink alone, and which are projected; if the downlink is not yet integrated, the relevant claims must be reworded or supported by measurements from a prototype that includes the downlink.
  2. [III-B] The abstract and Section III claim that Ant-Q ensures 'deterministic inter-circuit timing,' but the evaluation in Section V only reports PS-layer end-to-end timestamps and does not measure inter-circuit timing jitter or the worst-case latency of DRAM refresh and row/bank conflicts. The ping-pong buffer argument shows how non-determinism is absorbed in principle, but it does not demonstrate that the deterministic-timing bound holds under real workloads with the reported DRAM bandwidth and buffer sizes. Please add a timing-jitter characterization (e.g., distribution of inter-circuit gaps across a workload) or explicitly scope the deterministic-timing claim to the conditions that were actually verified.
  3. [V-A] In Section V-A, the paper states that with the addition of the downlink design, Ant-Q 'can now support arbitrarily long Clifford sequences' for 1Q and 2Q RB. This statement is presented as a conclusion from Table 2, but the downlink is the same component whose integration is deferred per Section IV. As written, this conflates a design argument with an experimental result. Please decouple the experimental contribution of the integrated uplink from the projected or simulated contribution of the downlink, and report each accordingly.
minor comments (4)
  1. [I] The phrase 'first open-sourced design' in the introduction would be clearer as 'first open-source design' or 'first design released as open source.'
  2. [Table 3] For QAOA MaxCut, the reported full Ant-Q overhead is negative (-0.08 ms, -0.02%), which the text attributes to the non-deterministic nature of PS-layer timestamping. Please include the measurement uncertainty or the number of repeated runs so readers can judge whether the negative value is statistically distinguishable from zero.
  3. [III-D] The early-termination thresholds (epsilon_TVD = epsilon_HF = 0.03, HF < 0.4, TVD > 0.6, and the 20% minimum shot requirement) are described as design choices without sensitivity analysis. A brief justification or citation for these specific values would strengthen the evaluation.
  4. [Figure 5] The caption of Figure 5 is clipped and the bar labels are dense; expand the caption and consider a table format for the shot percentages to improve readability.

Circularity Check

0 steps flagged · score 1.0 of 10

No material circularity found; the main risk is an implementation-completeness gap (downlink not yet integrated), not a derivation that reduces to its own inputs.

full rationale

The paper's load-bearing results are benchmark comparisons, not quantities derived from fitted parameters or from the authors' prior theorems. The overhead numbers in Tables 3-4 are computed as (measured total time - QPU time)/QPU time, where QPU time is independently defined as Shot # x Per-Shot time; Ant-Q rows are reported against the QubiC 2.0 baseline on the same ZCU216 board, so the improvement is externally anchored rather than self-referential. The deep-circuit claims (5101-Clifford 1Q RB, 500-Clifford 2Q RB) follow from the design's command decomposition and ping-pong buffering and are checked against the baseline's 2048-command BRAM limit; they do not import a uniqueness theorem. Early-stop thresholds (epsilon=0.03, HF<0.4, TVD>0.6) are hand-selected but affect only RQ3 and do not enter the throughput or deep-RB claims. Self-citations exist ([8] baseline, [17]/[32] open-source artifacts, [34] related scaling work) but are not load-bearing justifications. The one serious concern is implementation completeness, not circularity: Section IV says 'The Ant-Q Downlink is currently undergoing stress testing and will be integrated subsequently,' while Tables 2-4 label full-system rows 'Ant-Q Downlink + Uplink.' If those rows are not actual measurements of the integrated system, the abstract's near-zero-overhead and deep-RB claims would be unsupported; however, that would be an evidentiary gap or an unsupported projection, not a reduction of the result to its inputs. Because no step exhibits Eq. X = Eq. Y by construction or a fitted parameter renamed as a prediction, circularity burden is low.

Assumptions & free parameters 3 free parameters · 3 assumptions · 0 invented entities

The central claims rest on hardware behavior assumptions about BRAM and DRAM timing that are stated but not formally verified, and on hand-chosen early-stop thresholds that do not feed the throughput claims. No new physical entities are introduced.

free parameters (3)
  • Early-termination thresholds (TVD, HF, epsilons, checkpoint interval) = HF < 0.4, TVD > 0.6, epsilon_TVD = epsilon_HF = 0.03, checkpoints every 10%, minimum 20% shots
    Hand-chosen in Section III.D to classify unreadable results and convergence; no statistical model or calibration justifies these values, and they affect the early-termination results in RQ3.
  • Per-shot timing parameters = fast reset 2.83 us, readout 2.6 us, 1Q gate 24 ns, 2Q gate 332 ns
    Taken from the QubiC 2.0 default configuration in Section V.A; these determine the per-shot times and QPU-time denominators in Table 3, so the overhead percentages depend on them.
  • Buffer sizes = 64 KB downlink BRAM per buffer, 2048 cmds/half; 16 KB uplink BRAM, 1024 entries/half; 2 GB DRAM per direction
    Chosen to fit ZCU216 and absorb DRAM latency; the paper does not analyze sensitivity to these choices.
assumptions (3)
  • domain assumption Only BRAM can provide the single-cycle deterministic access needed for real-time control, so DRAM must be buffered through BRAM.
    Section II.C argues this is the memory bottleneck; it is not proven, and the central design builds on it.
  • domain assumption Ping-pong buffers with large DRAM can absorb worst-case DRAM latency without stalling the DSP.
    Section III.A/B asserts this as the key idea; no worst-case timing measurement or formal bound is provided.
  • domain assumption QPU time equals shot count times per-shot duration from the compiled circuit, and PS-layer timestamps cleanly separate overhead from execution.
    Section V.B uses this model to compute overhead percentages; the negative Delta values in Table 3 show the timestamp measurement has non-determinism.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Breaking Memory Bottlenecks in Quantum Control Systems for More Precise Experiments and Higher Throughput Computing." pith.science (2026). https://pith.science/paper/RI324AQI

@misc{pith2026260806318,
  author       = {Pith},
  title        = {Pith review of: Breaking Memory Bottlenecks in Quantum Control Systems for More Precise Experiments and Higher Throughput Computing},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RI324AQI}},
  note         = {Machine review of arXiv:2608.06318}
}
read the original abstract

As quantum computing continues to demonstrate promise and attract growing attention, there is an increasing need for more precise experiments to advance the development of quantum devices, as well as higher circuit throughput to validate more domain applications. However, this need is hindered by a memory bottleneck at the quantum control system layer, arising from limited on-chip BRAM capacity and the non-deterministic latency of DRAM. To break this bottleneck, we present Ant-Q, a memory hierarchy design that integrates DRAM with BRAM to support pipelined quantum circuit execution while ensuring deterministic inter-circuit timing. We evaluated Ant-Q using 26 real-world experimental and computing circuits. The results show that Ant-Q supports deep circuits for 1Q and 2Q Randomized Benchmarking and reduces the overhead of circuit loading and readout uplink relative to execution time from 22.90%-1417.05% to near zero. Ant-Q is being integrated into QubiC 3.0, with part of its functionalities already available.

Figures

Figures reproduced from arXiv: 2608.06318 by the authors.

Figure 1
Figure 1. Ant-Q transforms the workflow of quantum control systems. In the pipeline workflow of Ant-Q, circuit loading, execution, and readout uplink are decoupled and overlapped in time. As a result, readout data can be uplinked shortly after execution begins. the achievable clock frequency [10]. With so many memory constraints, punched-card workflow becomes a natural choice. In this work, we develop Ant-Q, a memory-hierarch… view at source ↗
Figure 2
Figure 2. shows the design overview of Ant-Q that integrates DRAM with BRAM and FF (Flip-Flop, which constitutes the registers) on the PL, to support data transfer at both downlink and uplink directions. From left to right, the downlink direction includes three data paths. The first data path transfers circuit commands. After the remote host compiles the circuit and submits it to the PS, the commands are first transferred to … view at source ↗
Figure 3
Figure 3. Design details for Ant-Q downlink part. computation respectively can be unified into a single objective: to avoid stalling the DSP/QPU. The key to achieving it is to make sure the next circuit is already in the BRAM when the current circuit completes. This effectively becomes a race between command consump￾tion and data transfer. The faster circuit commands can be transferred, the more qubits and denser circuit stru… view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Design details for Ant-Q uplink part. C. Ant-Q Uplink As the downlink continuously streams circuits and DSP cores execute them to generate readout results, the uplink must efficiently transfer these results to the PS and then to the host. This is essential to prevent b…
Figure 5
Figure 5. Figure 5: Early-stop effectiveness on the six noise-dominated circuits from [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: Overhead reduction under three parameter sweeps, based on the VQE 1Q Eigensolver circuit. Each sweep varies one factor while holding the other two fixed at their [PITH_FULL_IMAGE:figures/full_fig_p010_6.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

39 extracted references · 32 canonical work pages

  1. [1]

    Real-time adaptive tracking of fluctuating relaxation rates in supercon- ducting qubits,

    F. Berritta, J. Benestad, J. A. Krzywda, O. Krause, M. A. Marciniak, S. Krøjer, C. W. Warren, E. Hogedal, A. Nylander, I. Ahmadet al., “Real-time adaptive tracking of fluctuating relaxation rates in supercon- ducting qubits,”Physical Review X, vol. 16, no. 1, p. 011025, 2026

    work page 2026

  2. [2]

    Spectroscopy of low-frequency noise and its temperature dependence in a superconducting qubit

    F. Yan, J. Bylander, S. Gustavsson, F. Yoshihara, K. Harrabi, D. G. Cory, T. P. Orlando, Y . Nakamura, J.-S. Tsai, and W. D. Oliver, “Spectroscopy of low-frequency noise and its temperature dependence in a superconducting qubit,”Physical Review B, vol. 85, no. 17, p. 174521, 2012. [Online]. Available: https://arxiv.org/abs/1201.5665

    work page Pith review arXiv 2012

  3. [3]

    Millisecond charge-parity fluctuations and induced decoherence in a superconducting transmon qubit,

    D. Rist `e, C. C. Bultink, M. J. Tiggelman, R. N. Schouten, K. W. Lehnert, and L. DiCarlo, “Millisecond charge-parity fluctuations and induced decoherence in a superconducting transmon qubit,” Nature Communications, vol. 4, p. 1913, 2013. [Online]. Available: https://www.nature.com/articles/ncomms2936

    work page 1913

  4. [4]

    Minimum quantum run-time characterization and calibration via restless measurements with dynamic repetition rates

    C. Tornow, N. Kanazawa, W. E. Shanks, and D. J. Egger, “Minimum quantum run-time characterization and calibration via restless measurements with dynamic repetition rates,”Physical Review Applied, vol. 17, no. 6, p. 064061, 2022. [Online]. Available: https://arxiv.org/abs/2202.06981

    work page Pith review arXiv 2022

  5. [5]

    Trapped-ion two-qubit gates with>99.99% fidelity without ground-state cooling,

    A. C. Hughes, R. Srinivas, C. M. L ¨oschnauer, H. M. Knaack, R. Matt, C. J. Ballance, M. Malinowski, T. P. Harty, and R. T. Sutherland, “Trapped-ion two-qubit gates with>99.99% fidelity without ground-state cooling,”arXiv preprint arXiv:2510.17286, 2025. [Online]. Available: https://arxiv.org/abs/2510.17286

    arXiv

  6. [6]

    Manipulating the quantum state of an electrical circuit,

    D. Vion, A. Aassime, A. Cottet, P. Joyez, H. Pothier, C. Urbina, D. Esteve, and M. H. Devoret, “Manipulating the quantum state of an electrical circuit,”Science, vol. 296, no. 5569, pp. 886–889, 2002

    work page 2002

  7. [7]

    Efficient measurement of quantum gate error by interleaved randomized benchmarking,

    E. Magesan, J. M. Gambetta, B. R. Johnson, C. A. Ryan, J. M. Chow, S. T. Merkel, M. P. da Silva, G. A. Keefe, M. B. Rothwell, T. A. Ohki, M. B. Ketchen, and M. Steffen, “Efficient measurement of quantum gate error by interleaved randomized benchmarking,”Phys. Rev. Lett., vol. 109, no. 8, p. 080505, 2012

    work page 2012

  8. [8]

    Qubic 2.0: An extensible open-source qubit control system capable of mid-circuit measurement and feed- forward,

    Y . Xu, G. Huang, N. Fruitwala, A. Rajagopala, R. K. Naik, K. Nowrouzi, D. I. Santiago, and I. Siddiqi, “Qubic 2.0: An extensible open-source qubit control system capable of mid-circuit measurement and feed- forward,”arXiv preprint arXiv:2309.10333, 2023

    arXiv 2023

Show all 39 references
  1. [9]

    The qick (quantum instrumentation control kit): Readout and control for qubits and detectors,

    L. Stefanazzi, K. Treptow, N. Wilcer, C. Stoughton, C. Bradford, S. Uemura, S. Zorzetti, S. Montella, G. Cancelo, S. Sussmanet al., “The qick (quantum instrumentation control kit): Readout and control for qubits and detectors,”Review of Scientific Instruments, vol. 93, no. 4, ...

    2022

  2. [10]

    Optimizing designs using several types of memories on modern fpgas,

    M. Gungor, K. Huang, S. Ioannidis, and M. Leeser, “Optimizing designs using several types of memories on modern fpgas,” in2022 IEEE High Performance Extreme Computing Conference (HPEC). IEEE, 2022, pp. 1–7

    2022

  3. [11]

    Quantum computing in the cloud: Analyzing job and machine characteristics,

    G. S. Ravi, K. N. Smith, P. Gokhale, and F. T. Chong, “Quantum computing in the cloud: Analyzing job and machine characteristics,” in2021 IEEE International Symposium on Workload Characterization (IISWC). IEEE, 2021, pp. 39–50. [Online]. Available: https: //arxiv.org/abs/2203.13121

    2021 arXiv

  4. [12]

    Rethinking quantum noise in quantum machine learning: When noise improves learning,

    L. Zhu, Y . Dong, Z. Zhang, and X. Li, “Rethinking quantum noise in quantum machine learning: When noise improves learning,” arXiv preprint arXiv:2601.13275, 2026. [Online]. Available: https: //arxiv.org/abs/2601.13275

    2026

  5. [13]

    Adaptive fidelity estimation for quantum programs with graph-guided noise awareness,

    T. Li, Z. Zhao, and J. Yin, “Adaptive fidelity estimation for quantum programs with graph-guided noise awareness,” inProceedings of the AAAI Conference on Artificial Intelligence, vol. 40, no. 1, 2026, pp. 695–703. [Online]. Available: https://arxiv.org/abs/2601.14713

    2026

  6. [14]

    Principled performance tunability in operating system kernels,

    Z. Chen, W. Zhang, Y . Tang, R. Shu, F. Ren, T. Xu, and J. Liu, “Principled performance tunability in operating system kernels,”arXiv preprint arXiv:2512.12530, 2025

    2025 arXiv

  7. [15]

    Modeling noisy quantum circuits using experimental characterization,

    M. L. Dahlhauser and T. S. Humble, “Modeling noisy quantum circuits using experimental characterization,”Physical Review A, vol. 103, no. 4, p. 042603, 2021. [Online]. Available: https://arxiv.org/abs/2001.08653

    2021 arXiv

  8. [16]

    Learning to rank quantum circuits for hardware-optimized performance enhancement,

    G. S. Hartnett, A. Barbosa, P. S. Mundada, M. Hush, M. J. Biercuk, and Y . Baum, “Learning to rank quantum circuits for hardware-optimized performance enhancement,”Quantum, vol. 8, p. 1542, 2024. [Online]. Available: https://arxiv.org/abs/2404.06535

    2024 arXiv

  9. [17]

    Ant-q: Breaking memory bottlenecks in quantum control systems,

    Y . Guang, “Ant-q: Breaking memory bottlenecks in quantum control systems,” https://github.com/a85tract/Ant-Q, 2026

    2026

  10. [18]

    End-to-end protocol for high-quality QAOA parameters with few shots,

    T. Hao, Z. He, R. Shaydulin, J. Larson, and M. Pistoia, “End-to-end protocol for high-quality QAOA parameters with few shots,”Physical Review Research, vol. 7, no. 3, p. 033179, 2025. [Online]. Available: https://arxiv.org/abs/2408.00557

    2025

  11. [19]

    Demonstrating the power of quantum computers, certification of highly entangled measurements and scalable quantum nonlocality,

    E. B ¨aumer, N. Gisin, and A. Tavakoli, “Demonstrating the power of quantum computers, certification of highly entangled measurements and scalable quantum nonlocality,”npj Quantum Information, vol. 7, no. 1, p. 117, 2021

    2021

  12. [20]

    Relativistic variational-quantum-eigensolver calculations of molecular electric dipole moments on quantum hardware,

    P. Chawla, Shweta, K. R. Swain, T. Patel, R. Bala, D. Shetty, K. Sugisaki, S. B. Mandal, J. Riu, J. Nogu ´e, V . S. Prasannaa, and B. P. Das, “Relativistic variational-quantum-eigensolver calculations of molecular electric dipole moments on quantum hardware,”Physical Review A,...

    2025 arXiv

  13. [21]

    Ground-state energy estimation on current quantum hardware through the variational quantum eigensolver: A practical study,

    N. E. Belaloui, A. Tounsi, A. R. Khamadja, M. M. Louamri, A. Benslama, D. E. Bernal Neira, and M. T. Rouabah, “Ground-state energy estimation on current quantum hardware through the variational quantum eigensolver: A practical study,”Journal of Chemical Theory and Computation,...

    2025

  14. [22]

    A scalable 5,6-qubit Grover’s quantum search algorithm,

    D. R. Vemula, D. Konar, S. Satheesan, S. M. Kalidasu, and A. Cangi, “A scalable 5,6-qubit Grover’s quantum search algorithm,” arXiv preprint arXiv:2205.00117, 2022. [Online]. Available: https: //arxiv.org/abs/2205.00117

    2022 arXiv

  15. [23]

    Grove VQE documentation,

    Rigetti Computing, “Grove VQE documentation,” https://grove-docs. readthedocs.io/en/latest/vqe.html, 2018

    2018

  16. [24]

    Artificial-intelligence- driven shot reduction in quantum measurement,

    S. Liang, L. Zhu, X. Liu, C. Yang, and X. Li, “Artificial-intelligence- driven shot reduction in quantum measurement,”Chemical Physics Reviews, vol. 5, no. 4, p. 041403, 2024

    2024

  17. [25]

    Quantum query algorithms — fundamentals of quantum algorithms,

    IBM Quantum Learning, “Quantum query algorithms — fundamentals of quantum algorithms,” https://quantum.cloud. ibm.com/learning/courses/fundamentals-of-quantum-algorithms/ quantum-query-algorithms, 2024

    2024

  18. [26]

    Practical evaluation of quantum kernel methods for radar micro-doppler classification on noisy intermediate-scale quantum (NISQ) hardware,

    V . Agnihotri, J. Kaur, and S. Kaushik, “Practical evaluation of quantum kernel methods for radar micro-doppler classification on noisy intermediate-scale quantum (NISQ) hardware,”arXiv preprint arXiv:2601.22194, 2026. [Online]. Available: https://arxiv.org/abs/2601. 22194

    2026

  19. [27]

    Experimental study of Shor’s factoring algorithm using the IBM Q Experience,

    M. Amico, Z. H. Saleem, and M. Kumph, “Experimental study of Shor’s factoring algorithm using the IBM Q Experience,”Physical Review A, vol. 100, no. 1, p. 012305, 2019. [Online]. Available: https://arxiv.org/abs/1903.00768

    2019 arXiv

  20. [28]

    Teleportation — utility-scale quantum computing,

    IBM Quantum Learning, “Teleportation — utility-scale quantum computing,” https://quantum.cloud.ibm.com/learning/courses/ utility-scale-quantum-computing/teleportation, 2024

    2024

  21. [30]

    Three months in the life of cloud quantum computing,

    D. Teegarden, A. Casey, F. G. Serpa, P. Becker, A. Brahme, S. Kataria, and P. Lopata, “Three months in the life of cloud quantum computing,”arXiv preprint arXiv:2601.09943, 2026. [Online]. Available: https://arxiv.org/abs/2601.09943

    2026

  22. [31]

    QAOA tutorial,

    Open Quantum Design, “QAOA tutorial,” https://docs. openquantumdesign.org/examples/qaoa.html, 2024

    2024

  23. [32]

    benchmark qce: Open-source quantum circuit benchmark suite,

    Y . Guang, “benchmark qce: Open-source quantum circuit benchmark suite,” https://gitlab.com/yguang1/benchmark qce, 2026

    2026

  24. [33]

    Qiskit: An open-source framework for quantum computing,

    Qiskit contributors, “Qiskit: An open-source framework for quantum computing,” 2023. [Online]. Available: https://github.com/Qiskit/qiskit

    2023

  25. [34]

    Multi-FPGA synchronization and data communication for quantum control and mea- surement,

    Y . Xu, A. D. Rajagopala, N. Fruitwala, and G. Huang, “Multi-FPGA synchronization and data communication for quantum control and mea- surement,”arXiv preprint arXiv:2506.09856, 2025

    2025 arXiv

  26. [35]

    Advanced quantum testbed,

    Lawrence Berkeley National Laboratory, “Advanced quantum testbed,” https://aqt.lbl.gov/, accessed: 2026-04-26

    2026

  27. [36]

    An adaptive optimizer for measurement-frugal variational algorithms,

    J. M. K ¨ubler, A. Arrasmith, L. Cincio, and P. J. Coles, “An adaptive optimizer for measurement-frugal variational algorithms,”Quantum, vol. 4, p. 263, 2020

    2020

  28. [37]

    Qtenon: Towards low-latency architecture inte- gration for accelerating hybrid quantum-classical computing,

    C. Tao, L. Lu, S. Zheng, L.-W. Chang, M. Shen, H. Zhang, F. Liu, K. Zhou, and J. Yin, “Qtenon: Towards low-latency architecture inte- gration for accelerating hybrid quantum-classical computing,” inPro- ceedings of the 52nd Annual International Symposium on Computer Architectu...

    2025

  29. [38]

    Integration of quantum accelerators into hpc: Toward a unified quantum platform,

    A. Elsharkawy, X. Guo, and M. Schulz, “Integration of quantum accelerators into hpc: Toward a unified quantum platform,” in2024 IEEE International Conference on Quantum Computing and Engineering (QCE), vol. 01, 2024, pp. 774–783

    2024

  30. [39]

    Manarat: A scalable qick-based control system for superconducting quantum processors supporting synchro- nized control of 10 flux-tunable qubits,

    A. Silva and A. Orgaz Fuertes, “Manarat: A scalable qick-based control system for superconducting quantum processors supporting synchro- nized control of 10 flux-tunable qubits,”Review of Scientific Instruments, vol. 97, no. 2, p. 024702, 2026

    2026

  31. [2025]

    Available: https://arxiv.org/abs/2504.10595

    [Online]. Available: https://arxiv.org/abs/2504.10595

    arXiv

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.