Pith. sign in

REVIEW 3 major objections 6 minor 41 references

ARGON: A GNN-Empowered Compilation Framework for Scalable Neutral Atom Computing

T0 review · 3 major / 6 minor · reviewed 2026-08-01 · deepseek-v4-flash

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

Pith's one-line read ARGON breaks the neutral-atom compilation bottleneck by decoupling static layout from dynamic routing, finishing under 10 seconds with up to 100x fidelity gains.

desk verdict A real decoupling idea with a plausible compile-time speedup; the fidelity headline is not yet grounded outside its own simulator. read the letter →

arxiv 2607.21216 v1 pith:WLBB6Z6V submitted 2026-07-23 cs.ET quant-ph

Wenjie Sun , Xiaoyu Li , Zhigang Wang , Lianhui Yu , Geng Chen , Guowu Yang This is my paper
classification cs.ETquant-ph
keywords neutralatomcomputingquantumcompilationgraphneuralnetworkspatiotemporalschedulingRydberggateslayoutsynthesisAODroutingscalability
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

ARGON is a compiler for neutral-atom quantum computers that aims to remove the scalability wall caused by joint spatiotemporal scheduling. Rather than solving placement and routing together at compile time, it precomputes a library of hardware-validated, maximum-parallelism layouts offline, uses a graph neural network to select a layout that anticipates future routing congestion, and then runs a lightweight collision-free router. The paper reports sub-10-second compilation with an average speedup above 600x over prior compilers, and simulated execution fidelity improved by up to two orders of magnitude on dense circuits. If correct, the exponential search over coupled placement and movement collapses into constant-time neural inference plus light routing.

What carries the argument

The key machinery is the offline layout library combined with a GNN predictor. The layout library is built by an SMT solver solving a Maximum Independent Set problem over a conflict graph of gate placements, so every stored layout is hardware-certified, maximum-parallelism, and generated once per hardware geometry. The GNN—a three-layer graph isomorphism network with edge embeddings for future gate time-lags and sum-pooling over qubits—predicts a cumulative routing cost for a candidate layout across a lookahead horizon, giving the compiler topological foresight in constant time. The heuristic router handles the remaining kinematic pathfinding, using temporary parking to resolve cyclic depend

What would settle it

Run ARGON's generated schedules on a real neutral-atom processor and compare measured circuit fidelity to the Eq. (7) prediction. Specifically, compile a 25-qubit random-gates circuit with ARGON and with a leading prior compiler, execute both on the same hardware, and check whether the measured fidelity gap approaches the simulated two-order-of-magnitude gap. If the ordering reverses or the gap shrinks drastically, the error model is missing dominant physical errors.

Watch

Extended reading notes

Core claim

ARGON's central claim is that the 'spatiotemporal coupling' of neutral-atom compilation—where a gate placement must anticipate future AOD atom-transport trajectories—can be structurally decoupled without sacrificing quality. The framework resolves static geometric constraints offline via an SMT solver that builds a library of maximum-parallelism layouts (formulated as a maximum independent set on a conflict graph), and then reduces per-layer placement to a constant-time GNN forward pass over the current physical topology plus a lookahead window of future gates. A final heuristic router synthesizes collision-free AOD moves, parking atoms to break deadlocks and coloring non-conflicting traject

Load-bearing premise

The strongest assumption is that the fidelity model in Eq. (7), with the numerical fidelities and timings in Table 1, accurately predicts real execution; if actual neutral-atom hardware suffers errors the model omits—such as AOD movement crosstalk, atom loss during transfer, or correlated Rydberg errors—the claimed 100x fidelity gains may not materialize.

Editorial extensions

If this is right

  • Compilation time becomes a linear function of the number of Rydberg stages; ARGON stays under 3 seconds even for circuits with over 300 stages.
  • The decoupling makes compilation responsive enough for large-scale use: all evaluated benchmarks compile in under 10 seconds, with an average of 0.68 seconds.
  • Fewer Rydberg stages and reduced routing decoherence raise simulated end-to-end fidelity by up to two orders of magnitude on dense circuits, such as 25-qubit random gates.
  • The GNN predictor trained on a 16x16 array transfers zero-shot to 20x20 and 28x28 arrays, indicating the layout library acts as a scale-invariant microarchitectural primitive.
  • One-time offline costs—SMT layout generation, dataset rollout, and GNN training—are amortized because they depend only on hardware geometry, not on the circuit.

Reading between the lines

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

  • The same decoupling principle—offline hardware-certified candidate generation plus learned temporal selection—could plausibly be applied to other reconfigurable quantum architectures, such as trapped-ion systems with transport gates, though the paper does not claim this.
  • Because the GNN's lookahead window is fixed at K=4, a natural extension is to test whether variable or adaptive horizons, or training with reinforcement learning instead of supervised rollout labels, further improves routing decisions on workloads with longer-range dependencies.
  • The zero-shot generalization results suggest a compositional scaling strategy: compiling very large arrays by tiling the precomputed local layouts and using the GNN to stitch them, rather than recomputing layouts for the full device; this is our inference, not the paper's explicit claim.
  • The fidelity estimates depend entirely on the product-form error model in Eq. (7); real hardware may introduce error correlations, such as crosstalk during AOD sweeps, that the model misses, so the 100x fidelity claim is best interpreted as a simulator-level bound until measured on a physical device.
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 / 6 minor

Summary. The paper proposes ARGON, a compilation framework for neutral atom quantum processors built on spatiotemporal decoupling. An offline SMT-based phase precomputes a library of high-parallelism spatial layouts; a GNN trained on rollout data selects layouts with temporal foresight; a heuristic router converts the selected layouts into collision-free AOD transport. The evaluation on MQT Bench, QASMBench, and synthetic workloads reports sub-10s compilation, >600x average speedup over Enola/DasAtom/PowerMove, reduced Rydberg stages, and up to 10^2x simulated end-to-end fidelity improvement. The central claimed contribution is replacing exponential joint spatiotemporal search with constant-time neural inference plus a light routing backend.

Significance. If the fidelity claims withstand scrutiny, this is a significant architectural contribution: it identifies the spatiotemporal coupling bottleneck in neutral-atom compilation and replaces joint search with an offline geometric layout library plus learned layout selection. The evaluation design is systematic: multiple benchmark suites, held-out generalization tasks, ablations, hardware-like parameters, and zero-shot grid scaling are all present. The compilation-speed result is plausible because the offline NRE amortization is explicitly discussed. The main risk is that all physical-fidelity claims rest on the paper's own simulation model, Eq. (7), with no hardware measurement or sensitivity analysis; this risk is load-bearing because the abstract advertises execution-fidelity gains as a headline result.

major comments (3)
  1. [§4.1, Eq. (7); §4.2; §4.4] The headline fidelity improvement (up to 10^2x) is computed with the paper's own penalty model, Eq. (7), and no hardware calibration or measurement is reported. For dense circuits the dominant term is f_exc^(|Q|L - 2|G2|); L is precisely what the offline SMT library is constructed to minimize and what the GNN rollout cost in Eq. (5) targets. The same simulation stack supplies the GNN training labels and the evaluation metric, so the result is at least partially in-model circularity. Section 4.4 shows ARGON's advantage is concentrated in idle fidelity while DasAtom has higher transfer and movement fidelity; under a transfer-dominated or distance-dependent error budget the advantage could shrink or reverse. Please add a sensitivity analysis over f_exc and f_trans, an error model with spatial crosstalk, or hardware validation before claiming physical fidelity gains.
  2. [§4.2, Fig. 7(h)] The paper acknowledges that PowerMove yields higher fidelity on some larger QV circuits and explains this by PowerMove not imposing a strict upper bound on hardware spatial parallelism. This means those baseline schedules violate the same exclusion-zone constraint used to define the hardware model elsewhere. Comparing ARGON against schedules that are physically invalid under the paper's own model inflates ARGON's relative fidelity; if PowerMove's zoned-architecture assumptions are instead accepted, the comparison is not apples-to-apples. Please enforce identical physical constraints for all baselines or report the QV comparison as a separately labeled assumption set.
  3. [§4.5, Fig. 11 and Eq. (2)] The zero-shot generalization claim is not fully supported. The GNN node features are absolute coordinates (Eq. (2), X = [x_old, y_old, x_new, y_new]) and no normalization or coordinate-relative encoding is described. A model trained only on a 16x16 grid may not transfer to 20x20 or 28x28 for reasons unrelated to physical scale invariance. Moreover, Fig. 11 compares ARGON only against random layout selection on the larger grids; no existing compiler baseline or absolute fidelity is reported there. Please report actual fidelity values on scaled grids, compare with DasAtom/Enola/PowerMove on those grids, and specify the coordinate encoding used for transfer.
minor comments (6)
  1. [Abstract / §3.2] Typo: 'an Graph Neural Network' should be 'a Graph Neural Network'.
  2. [Eq. (2)] The matrices X, tau, and E in Eq. (2) are hard to parse. The relationship between the four-node encoding and the 2x3 edge tensor should be explained in the text or a caption.
  3. [§4.2, Table 2] The 'average speedup above 600x' calculation is not specified (arithmetic mean? geometric mean? how are timeouts treated?). Please state the aggregation rule and the number of runs.
  4. [§3.2.2] The sentence 'approximately 3 hours and 30 minutes, respectively' is ambiguous: is 3h30m the total for dataset generation and training combined, or separately? Please clarify.
  5. [Abstract vs §4.2] The abstract says 'improving execution fidelity by up to 10^2x', but §4.2 reports fidelity gains of over 8 to 10 orders of magnitude on the 25-qubit Random gates benchmark. Please harmonize the headline claim with the actual numbers.
  6. [Fig. 7] The multi-panel figure is very dense. Splitting the time, fidelity, and Rydberg-stage panels into separate figures would improve readability.

Circularity Check

0 steps flagged · score 0.0 of 10

No load-bearing circularity: the GNN is trained on routing-cost rollouts, not on the fidelity metric; the fidelity model uses externally calibrated parameters and held-out benchmarks are reported.

full rationale

ARGON's central speedup claim is wall-clock compilation time measured against independent baselines (DasAtom, Enola, PowerMove) and does not depend on the GNN predictions for its latency advantage; the offline library lookup plus neural forward pass plus light router are the actual compile-time pipeline. The fidelity claim is computed from Eq. (7), a product model whose parameters (f2, fexc, ftrans, T2, timings) are taken from published neutral-atom hardware references, not fitted to ARGON. The GNN's training target (Eq. (5)) is a discounted routing-cost rollout, not the fidelity F of Eq. (7); lowering C is correlated with lowering T_q and Rydberg stages, but the evaluation F is not identical to the training loss. The offline library maximizes per-stage gate count through Eq. (1), but the reported Rydberg-stage count L arises from the full schedule and varies across circuits; it is not set equal to the MIS objective. Held-out benchmarks (QV, 3-Regular, QASMBench) and a random-layout ablation provide independent evidence. The reported W-state/Random-gate fidelity gains are on circuit families used for rollout training, which is a mild validation weakness, but it does not make the derived fidelity equal to an input, and the held-out results plus ablation support the qualitative claim. No self-citation chain is load-bearing. The PowerMove-favorable QV cases and the omitted Stade fidelity comparison are honest caveats, not circularity.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The paper introduces no new physical entity. Its load-bearing assumptions are the hardware mobility rules, the spatial interference model, the analytic fidelity formula, and the scale-invariance of local layouts. The free parameters are mainly GNN training choices and the lookahead horizon; the discount factor value is not disclosed.

free parameters (3)
  • GNN lookahead window K = 4
    K=4 chosen by empirical profiling as the latency-accuracy trade-off (§4.2); directly affects the predictor's temporal horizon.
  • Discount factor gamma
    Introduced in Eq. (5) to attenuate future rollout costs; value not reported, but it shapes the training labels and thus the learned policy.
  • GNN hyperparameters
    Hidden dimensions, learning rate, number of training examples, and epochs are not stated; required to reproduce the trained predictor.
assumptions (4)
  • domain assumption AOD traps sharing a row or column move in tandem, and rows/columns cannot cross during transport
    Hardware mobility rule from [17,32], used throughout Section 2.2 and in the router design.
  • domain assumption Parallel two-qubit gates are valid only if atoms are within R_int and exclusion zones of radius R_r are disjoint
    Static spatial model used for the offline MIS/SMT formulation in Section 3.1.
  • domain assumption End-to-end fidelity equals the product of two-qubit, idle-excitation, transfer, and T2 decoherence penalties
    Eq. (7) is the evaluation metric; all fidelity claims rest on this model, which is not validated against hardware.
  • ad hoc to paper Layout templates are local and transfer zero-shot to larger grids
    Asserted in Section 4.5 and Section 5; supported only by two simulation transfer experiments, not by a derivation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of ARGON: A GNN-Empowered Compilation Framework for Scalable Neutral Atom Computing." pith.science (2026). https://pith.science/paper/WLBB6Z6V

@misc{pith2026260721216,
  author       = {Pith},
  title        = {Pith review of: ARGON: A GNN-Empowered Compilation Framework for Scalable Neutral Atom Computing},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WLBB6Z6V}},
  note         = {Machine review of arXiv:2607.21216}
}
read the original abstract

Neutral atom quantum systems offer a promising pathway to large-scale quantum computing due to high qubit uniformity and flexible connectivity. To exploit this architecture, compilers must coordinate dynamic atom transport alongside highly parallel entangling gates. As circuits scale, the interplay between these operations becomes a system bottleneck, introducing denser logical interactions and longer temporal dependencies. Compilers must simultaneously satisfy rigid spatial constraints and complex movement schedules. Existing joint spatiotemporal compilation methods face an exponentially expanding search space, incurring substantial overheads or compromising fidelity as circuit size grows. In this work, we propose ARGON, a scalable compilation framework that introduces a spatiotemporal decoupling paradigm for neutral atom processors. Our key novelty is offloading static geometric conflict resolution to an offline phase, precomputing a library of hardware-certified, high-parallelism spatial layouts. To guide temporal routing, we deploy a Graph Neural Network (GNN) predictor to evaluate candidate layouts against deep temporal horizons, proactively evading downstream kinematic bottlenecks. Finally, a heuristic router translates the selected sequence into collision-free physical transport. Evaluations show ARGON completes compilation in under 10 seconds, delivering up to a >10^4x and 600x average speedup over state-of-the-art baselines. ARGON also minimizes routing decoherence and reduces Rydberg stages, improving execution fidelity by up to 10^2x on dense circuits.

Figures

Figures reproduced from arXiv: 2607.21216 by the authors.

Figure 1
Figure 1. The scalability bottleneck of SOTA NA compilers. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Spatiotemporal scheduling constraints in reconfigurable neutral atom arrays. Blue regions represent conflict regions [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Overview of the ARGON Framework. The compilation pipeline isolates spatiotemporal evaluations into three [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: Visualization of spatial connectivity and solver opti [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: The architectural structure of the GNN-guided pre [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: An example of encoding X, 𝜏, and E [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: Experimental results on Enola, DasAtom, PowerMove, and ARGON. The first row [PITH_FULL_IMAGE:figures/full_fig_p010_7.png]
Figure 9
Figure 9. Figure 9: Detailed breakdown of fidelity components for dif [PITH_FULL_IMAGE:figures/full_fig_p011_9.png]
Figure 10
Figure 10. Figure 10: Ablation study evaluating the efficacy of the GNN-guided predictive policy. [PITH_FULL_IMAGE:figures/full_fig_p012_10.png]
Figure 11
Figure 11. Figure 11: Zero-shot generalization evaluation of the GNN [PITH_FULL_IMAGE:figures/full_fig_p012_11.png]
Figure 12
Figure 12. Figure 12: Evaluation of compile-time overhead. (a) Compi [PITH_FULL_IMAGE:figures/full_fig_p012_12.png]
Figure 13
Figure 13. Figure 13: Performance comparison between ARGON, DasAtom, and the zoned-architecture heuristic by Stade et al. [30] on QASMBench. data-driven explorations. While ARGON targets dense monolithic architectures, full zoned systems and FTQC stacks additionally require storage managem…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

41 extracted references · 4 canonical work pages

  1. [1]

    Egger, Bruce G

    Amira Abbas, Andris Ambainis, Brandon Augustino, Andreas Bärtschi, Harry Buhrman, Carleton Coffrin, Giorgio Cortiana, Vedran Dunjko, Daniel J. Egger, Bruce G. Elmegreen, Nicola Franco, Filippo Fratini, Bryce Fuller, Julien Gacon, Constantin Gonciulea, Sander Gribling, Swati Gupta, Stuart Hadfield, Raoul Heese, Gerhard Kircher, Thomas Kleinert, Thorsten Ko...

    2024

  2. [2]

    Bardin, Rami Barends, Rupak Biswas, Sergio Boixo, Fernando G

    Frank Arute, Kunal Arya, Ryan Babbush, Dave Bacon, Joseph C. Bardin, Rami Barends, Rupak Biswas, Sergio Boixo, Fernando G. S. L. Brandao, David A. Buell, Brian Burkett, Yu Chen, Zijun Chen, Ben Chiaro, Roberto Collins, William Court- ney, Andrew Dunsworth, Edward Farhi, Brooks Foxen, Austin Fowler, Craig Gid- ney, Marissa Giustina, Rob Graff, Keith Guerin...

    2019

  3. [3]

    Baker, Andrew Litteken, Casey Duckering, Henry Hoffmann, Hannes Bernien, and Frederic T

    Jonathan M. Baker, Andrew Litteken, Casey Duckering, Henry Hoffmann, Hannes Bernien, and Frederic T. Chong. 2021. Exploiting long-distance interactions and tolerating atom loss in neutral atom quantum architectures. InProceedings of the 48th Annual International Symposium on Computer Architecture (ISCA ’21). IEEE Press, Virtual Event, Spain, 818–831. http...

    arXiv 2021

  4. [4]

    Bernstein and Tanja Lange

    Daniel J. Bernstein and Tanja Lange. 2017. Post-Quantum Cryptography.Nature 549, 7671 (Sept. 2017), 188–194. https://doi.org/10.1038/nature23461

    doi:10.1038/nature23461 2017

  5. [5]

    Evered, Alexandra A

    Dolev Bluvstein, Simon J. Evered, Alexandra A. Geim, Sophie H. Li, Hengyun Zhou, Tom Manovitz, Sepehr Ebadi, Madelyn Cain, Marcin Kalinowski, Dominik Hangleiter, J. Pablo Bonilla Ataides, Nishad Maskara, Iris Cong, Xun Gao, Pe- dro Sales Rodriguez, Thomas Karolyshyn, Giulia Semeghini, Michael J. Gullans, Markus Greiner, Vladan Vuletić, and Mikhail D. Luki...

    work page doi:10.1038/s41586-023-0 2024

  6. [6]

    Geim, Sophie H

    Dolev Bluvstein, Alexandra A. Geim, Sophie H. Li, Simon J. Evered, J. Pablo Bonilla Ataides, Gefen Baranes, Andi Gu, Tom Manovitz, Muqing Xu, Marcin Kalinowski, Shayan Majidy, Christian Kokail, Nishad Maskara, Elias C. Trapp, Luke M. Stewart, Simon Hollerith, Hengyun Zhou, Michael J. Gullans, Susanne F. Yelin, Markus Greiner, Vladan Vuletić, Madelyn Cain,...

    2025

  7. [7]

    Wang, Sepehr Ebadi, Marcin Kalinowski, Alexander Keesling, Nishad Maskara, Hannes Pichler, Markus Greiner, Vladan Vuletić, and Mikhail D

    Dolev Bluvstein, Harry Levine, Giulia Semeghini, Tout T. Wang, Sepehr Ebadi, Marcin Kalinowski, Alexander Keesling, Nishad Maskara, Hannes Pichler, Markus Greiner, Vladan Vuletić, and Mikhail D. Lukin. 2022. A quantum processor based on coherent transport of entangled atom arrays.Nature604, 7906 (April 2022), 451–456. https://doi.org/10.1038/s41586-022-04...

    doi:10.1038/s41586-022-04592-6 2022

  8. [8]

    Trapp, Jinen Guo, Mohamed H

    Neng-Chun Chiu, Elias C. Trapp, Jinen Guo, Mohamed H. Abobeih, Luke M. Stewart, Simon Hollerith, Pavel L. Stroganov, Marcin Kalinowski, Alexandra A. Geim, Simon J. Evered, Sophie H. Li, Xingjian Lyu, Lisa M. Peters, Dolev Bluvstein, Tout T. Wang, Markus Greiner, Vladan Vuletić, and Mikhail D. Lukin. 2025. Continuous operation of a coherent 3,000-qubit sys...

    doi:10.1038/s41586-025-09596-6 2025

Show all 41 references
  1. [9]

    Santos, and Evan Sheridan

    Laura Clinton, Toby Cubitt, Brian Flynn, Filippo Maria Gambetta, Joel Klassen, Ashley Montanaro, Stephen Piddock, Raul A. Santos, and Evan Sheridan. 2024. Towards near-term quantum simulation of materials.Nature Communications15, 1 (Jan. 2024), 211. https://doi.org/10.1038/s41...

    2024 doi

  2. [10]

    Covey, Harald Weinfurter, and Hannes Bernien

    Jacob P. Covey, Harald Weinfurter, and Hannes Bernien. 2023. Quantum Networks with Neutral Atom Processing Nodes.npj Quantum Information9, 1 (Sept. 2023),

    2023

  3. [11]

    Cross, Lev S

    Andrew W. Cross, Lev S. Bishop, Sarah Sheldon, Paul D. Nation, and Jay M. Gambetta. 2019. Validating quantum computers using randomized model circuits. Physical Review A100, 3 (Sept. 2019), 032328. https://doi.org/10.1103/PhysRevA. 100.032328

    2019 doi

  4. [12]

    Ebadi, A

    S. Ebadi, A. Keesling, M. Cain, T. T. Wang, H. Levine, D. Bluvstein, G. Semeghini, A. Omran, J.-G. Liu, R. Samajdar, X.-Z. Luo, B. Nash, X. Gao, B. Barak, E. Farhi, S. Sachdev, N. Gemelke, L. Zhou, S. Choi, H. Pichler, S.-T. Wang, M. Greiner, V. Vuletić, and M. D. Lukin. 2022....

    2022 doi

  5. [13]

    Evered, Dolev Bluvstein, Marcin Kalinowski, Sepehr Ebadi, Tom Manovitz, Hengyun Zhou, Sophie H

    Simon J. Evered, Dolev Bluvstein, Marcin Kalinowski, Sepehr Ebadi, Tom Manovitz, Hengyun Zhou, Sophie H. Li, Alexandra A. Geim, Tout T. Wang, Nishad Maskara, Harry Levine, Giulia Semeghini, Markus Greiner, Vladan Vuletić, and Mikhail D. Lukin. 2023. High-fidelity parallel enta...

    2023 doi

  6. [14]

    T. M. Graham, Y. Song, J. Scott, C. Poole, L. Phuttitarn, K. Jooya, P. Eichler, X. Jiang, A. Marra, B. Grinkemeyer, M. Kwon, M. Ebert, J. Cherek, M. T. Lichtman, M. Gillette, J. Gilbert, D. Bowman, T. Ballance, C. Campbell, E. D. Dahl, O. Crawford, N. S. Blunt, B. Rogers, T. N...

    2022 doi

  7. [15]

    Harrigan, Kevin J

    Matthew P. Harrigan, Kevin J. Sung, Matthew Neeley, Kevin J. Satzinger, Frank Arute, Kunal Arya, Juan Atalaya, Joseph C. Bardin, Rami Barends, Sergio Boixo, Michael Broughton, Bob B. Buckley, David A. Buell, Brian Burkett, Nicholas Bushnell, Yu Chen, Zijun Chen, Ben Chiaro, Ro...

  8. [16]

    Loïc Henriet, Lucas Beguin, Adrien Signoles, Thierry Lahaye, Antoine Browaeys, Georges-Olivier Reymond, and Christophe Jurczak. 2020. Quantum Computing with Neutral Atoms.Quantum4 (Sept. 2020), 327. https://doi.org/10.22331/q- 2020-09-21-327

    2020 doi

  9. [17]

    Yunqi Huang, Dingchao Gao, Shenggang Ying, and Sanjiang Li. 2025. DasAtom: A Divide-and-Shuttle Atom Approach to Quantum Circuit Transformation.IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems(2025), 1–1. https://doi.org/10.1109/TCAD.2025.3532818

    2025

  10. [18]

    Giacomo Lancellotti, Simone Perriello, Alessandro Barenghi, and Gerardo Pelosi

  11. [19]

    Ang Li, Samuel Stein, Sriram Krishnamoorthy, and James Ang. 2023. QASMBench: A Low-Level Quantum Benchmark Suite for NISQ Evaluation and Simulation. ACM Transactions on Quantum Computing4, 2 (Feb. 2023), 10:1–10:26. https: //doi.org/10.1145/3550488

    2023 doi

  12. [20]

    Yongshang Li, Yu Zhang, Mingyu Chen, Xiangyang Li, and Peng Xu. 2023. Timing- Aware Qubit Mapping and Gate Scheduling Adapted to Neutral Atom Quantum Computing.IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems42, 11 (Nov. 2023), 3768–3780. https://d...

    2023 doi

  13. [21]

    Wan-Hsuan Lin, Jason Kimko, Bochen Tan, Nikolaj Bjørner, and Jason Cong

  14. [22]

    Jason Ludmir and Tirthak Patel. 2024. Parallax: A Compiler for Neutral Atom Quantum Computers under Hardware Constraints. InProceedings of the Inter- national Conference for High Performance Computing, Networking, Storage, and Analysis (SC ’24). IEEE Press, Atlanta, GA, USA, 1...

    2024 arXiv

  15. [23]

    Tirthak Patel, Daniel Silver, and Devesh Tiwari. 2022. Geyser: a compilation framework for quantum computing with neutral atoms. InProceedings of the 49th Annual International Symposium on Computer Architecture (ISCA ’22). Association for Computing Machinery, New York, NY, USA...

    2022

  16. [24]

    Tirthak Patel, Daniel Silver, and Devesh Tiwari. 2023. GRAPHINE: Enhanced Neutral Atom Quantum Computing Using Application-Specific Rydberg Atom 14 ARGON: A GNN-Empowered Compilation Framework for Scalable Neutral Atom Computing Conference’17, July 2017, Washington, DC, USA Ar...

    2023

  17. [25]

    Nils Quetschlich, Lukas Burgholzer, and Robert Wille. 2023. MQT Bench: Bench- marking Software and Design Automation Tools for Quantum Computing.Quan- tum7 (July 2023), 1062. https://doi.org/10.22331/q-2023-07-20-1062 Pub- lisher: Verein zur Förderung des Open Access Publizier...

    2023 doi

  18. [26]

    Jixuan Ruan, Xiang Fang, Hezi Zhang, Ang Li, Travis Humble, and Yufei Ding

  19. [27]

    Irfansha Shaik and Jaco van de Pol. 2024. Optimal Layout Synthesis for Deep Quantum Circuits on NISQ Processors with 100+ Qubits. InDROPS- IDN/v2/document/10.4230/LIPIcs.SAT.2024.26. https://doi.org/10.4230/LIPIcs.SAT. 2024.26

    2024 doi

  20. [28]

    Lukin, and Vladan Vuletić

    Josiah Sinclair, Joshua Ramette, Brandon Grinkemeyer, Dolev Bluvstein, Mikhail D. Lukin, and Vladan Vuletić. 2025. Fault-tolerant optical intercon- nects for neutral-atom arrays.Physical Review Research7, 1 (March 2025), 013313. https://doi.org/10.1103/PhysRevResearch.7.013313

    2025 doi

  21. [29]

    Kemp, and Hannes Bernien

    Kevin Singh, Shraddha Anand, Andrew Pocklington, Jordan T. Kemp, and Hannes Bernien. 2022. Dual-Element, Two-Dimensional Atom Array with Continuous- Mode Operation.Physical Review X12, 1 (March 2022), 011040. https://doi.org/ 10.1103/PhysRevX.12.011040

    2022 doi

  22. [30]

    Yannick Stade, Wan-Hsuan Lin, Jason Cong, and Robert Wille. 2025. Routing- Aware Placement for Zoned Neutral Atom-based Quantum Computing. In2025 IEEE/ACM International Conference On Computer Aided Design (ICCAD). 1–9. https://doi.org/10.1109/ICCAD66269.2025.11240721 ISSN: 1558-2434

    2025

  23. [31]

    Bochen Tan, Dolev Bluvstein, Mikhail Lukin, and Jason Cong. 2024. Compiling Quantum Circuits for Dynamically Field-Programmable Neutral Atoms Array Processors.Quantum8 (March 2024), 1281. https://doi.org/10.22331/q-2024-03- 14-1281

    2024 doi

  24. [32]

    Lukin, and Jason Cong

    Bochen Tan, Dolev Bluvstein, Mikhail D. Lukin, and Jason Cong. 2022. Qubit Mapping for Reconfigurable Atom Arrays. InProceedings of the 41st IEEE/ACM International Conference on Computer-Aided Design. ACM, San Diego California, 1–9. https://doi.org/10.1145/3508352.3549331

    2022

  25. [33]

    Bochen Tan and Jason Cong. 2020. Optimal Layout Synthesis for Quantum Computing. In2020 IEEE/ACM International Conference On Computer Aided Design (ICCAD). 1–9

    2020

  26. [34]

    Daniel Bochen Tan, Wan-Hsuan Lin, and Jason Cong. 2025. Compilation for Dynamically Field-Programmable Qubit Arrays with Efficient and Provably near- Optimal Scheduling. InProceedings of the 30th Asia and South Pacific Design Automation Conference. Association for Computing Ma...

    2025

  27. [35]

    Pan, Jason Cong, Umut A

    Hanrui Wang, Pengyu Liu, Daniel Bochen Tan, Yilian Liu, Jiaqi Gu, David Z. Pan, Jason Cong, Umut A. Acar, and Song Han. 2024. Atomique: A Quantum Compiler for Reconfigurable Neutral Atom Arrays. In2024 ACM/IEEE 51st Annual International Symposium on Computer Architecture (ISCA...

    2024

  28. [36]

    Pablo Bonilla Ataides, Christopher A

    Qian Xu, J. Pablo Bonilla Ataides, Christopher A. Pattison, Nithin Raveendran, Dolev Bluvstein, Jonathan Wurtz, Bane Vasić, Mikhail D. Lukin, Liang Jiang, and Hengyun Zhou. 2024. Constant-overhead fault-tolerant quantum computation with reconfigurable atom arrays.Nature Physic...

    2024 doi

  29. [90]

    https://doi.org/10.1038/s41534-023-00759-9

    doi

  30. [2021]

    https://doi.org/10.1038/s41567-020-01105-y

    Quantum approximate optimization of non-planar graph problems on a planar superconducting processor.Nature Physics17, 3 (March 2021), 332–336. https://doi.org/10.1038/s41567-020-01105-y

    2021 doi

  31. [2023]

    In2023 60th ACM/IEEE Des

    Scalable Optimal Layout Synthesis for NISQ Quantum Processors. In2023 60th ACM/IEEE Des. Autom. Conf. (DAC). 1–6. https://doi.org/10.1109/DAC56929. 2023.10247760

    2023

  32. [2024]

    InProceedings of the 61st ACM/IEEE Design Automation Conference (DAC ’24)

    Design of a Quantum Walk Circuit to Solve the Subset-Sum Problem. InProceedings of the 61st ACM/IEEE Design Automation Conference (DAC ’24). Association for Computing Machinery, New York, NY, USA, 1–6. https://doi. org/10.1145/3649329.3657337

  33. [2025]

    InProceedings of the 30th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3 (ASPLOS ’25)

    PowerMove: Optimizing Compilation for Neutral Atom Quantum Com- puters with Zoned Architecture. InProceedings of the 30th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3 (ASPLOS ’25). Association for Computing Mac...

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.