pith. sign in

arxiv: 2605.17145 · v1 · pith:IKNFI3NSnew · submitted 2026-05-16 · 🪐 quant-ph

Scaling Quantum Optimization for Unit Commitment via Pauli Correlation Encoding

Kien X. Nguyen , Ilya Safro , Xiaoyuan Liu This is my paper

Pith reviewed 2026-05-20 14:34 UTC · model grok-4.3

classification 🪐 quant-ph
keywords unit commitmentquantum optimizationPauli correlation encodinghybrid quantum-classicalpower systemsbinary encodingqubit reduction
0
0 comments X

The pith

Pauli correlation encoding lets hybrid quantum-classical optimization handle much larger unit commitment problems by using far fewer qubits for binary on/off decisions.

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

The paper develops a hybrid method for unit commitment, the problem of choosing which generators to turn on or off over multiple time periods while meeting power demand and respecting limits on ramping, reserves, and other constraints. A quantum optimizer proposes the binary on/off schedule and a classical optimizer then computes the actual power levels that satisfy the continuous constraints. The key innovation is Pauli correlation encoding, which represents many binary decisions through correlations among Pauli operators instead of assigning one qubit to each decision variable. This qubit reduction makes it possible to treat planning horizons that would otherwise exceed current quantum hardware limits. The authors demonstrate the approach on instances up to 312 binary variables and report that the resulting schedules remain feasible and competitive in cost.

Core claim

By framing unit commitment as a leader-follower problem and encoding the leader's binary decisions via Pauli correlations, the hybrid procedure scales to multi-period schedules using substantially fewer qubits than one-qubit-per-variable encodings while still returning feasible, near-optimal dispatch solutions.

What carries the argument

Pauli-Correlation Encoding, which maps binary variables to correlations among Pauli operators so that many decisions share the same qubits.

If this is right

  • Larger multi-period unit commitment instances become solvable on near-term quantum hardware.
  • The leader-follower split cleanly separates discrete scheduling from continuous economic dispatch.
  • Operational constraints such as ramping limits and reserve requirements remain satisfied in the final schedules.
  • Competitive total operating costs are achieved on both small and large test instances.

Where Pith is reading between the lines

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

  • The same correlation-encoding idea could be applied to other mixed-integer problems that separate binary decisions from continuous variables.
  • Error mitigation techniques tailored to the correlation structure might further improve solution quality on noisy hardware.
  • Embedding the method inside a rolling-horizon framework could address real-time operations where forecasts update frequently.

Load-bearing premise

Once the quantum optimizer fixes the on/off decisions, the classical solver can always produce a feasible and near-optimal power dispatch without backtracking or infeasibility.

What would settle it

A test case in which the on/off schedule returned by the quantum optimizer leaves the classical power-dispatch subproblem infeasible or markedly suboptimal.

Figures

Figures reproduced from arXiv: 2605.17145 by Ilya Safro, Kien X. Nguyen, Xiaoyuan Liu.

Figure 1
Figure 1. Figure 1: Overview of our heuristic-assisted differentiable bilevel optimization on a 12-variable problem, which is encoded into 4 qubits. The quantum circuit [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Ablation study on the number of ansatz layers over the range [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Ablation study on the number of optimization steps over the range [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 5
Figure 5. Figure 5: Ablation study on the number of optimization steps over the range [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
read the original abstract

Unit commitment is an important optimization problem in power system operations, classified as NP-hard. This paper presents a hybrid quantum-classical method for the unit commitment problem with time-dependent constraints, where decisions must be made about which generators to turn on/off and how much power they should produce over a planning horizon. We use a hybrid quantum-classical optimization procedure to determine the on/off schedules of the generating units and the corresponding power dispatch that satisfies operational constraints such as load balance, generator limits, ramping, and reserve requirements. We frame the optimization loop as a leader-follower structure, where the quantum optimizer leads to give the on/off decisions, and the classical optimizer follows to produce the power level schedule. Leveraging Pauli-Correlation Encoding, our method scales to horizon-wide unit commitment schedules by encoding the binary variables with far fewer qubits. By combining these components, the method can handle multi-period settings while using far fewer qubits than straightforward quantum encodings that allocate one qubit per decision variable as in prior approaches. We evaluate the approach on both small- and large-scale instances, up to 312 binary variables, and show that it reliably produces feasible schedules with competitive operating costs.

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.

Referee Report

2 major / 2 minor

Summary. The manuscript presents a hybrid quantum-classical method for the unit commitment problem. It frames the task as a leader-follower optimization in which a quantum optimizer using Pauli-Correlation Encoding supplies binary on/off decisions for generators over a multi-period horizon, while a classical optimizer solves the continuous power-dispatch subproblem. The encoding is claimed to represent the binary variables with substantially fewer qubits than one-qubit-per-variable schemes, enabling solution of instances with up to 312 binary variables while satisfying load-balance, ramping, reserve, and minimum up/down-time constraints and yielding competitive operating costs.

Significance. If the feasibility guarantees and qubit-reduction claims are substantiated, the work would provide a concrete demonstration that hybrid quantum methods can address a practically relevant mixed-integer scheduling problem at scales beyond current quantum hardware limits. The explicit leader-follower decomposition and the Pauli-Correlation Encoding constitute the main technical contributions; reproducible code or machine-checked proofs of constraint preservation would further strengthen the result.

major comments (2)
  1. [Optimization loop / Leader-follower structure] Leader-follower decomposition (abstract and optimization-loop description): The scaling claim rests on the premise that every binary vector returned by the quantum optimizer admits a feasible classical dispatch satisfying all time-coupling constraints. The manuscript provides no explicit mechanism, penalty term, or post-processing step that enforces ramp limits, minimum up/down times, or reserve coupling inside the Pauli-Correlation Encoding or the quantum objective; if these constraints are enforced only in the follower LP, infeasible on/off schedules can arise, necessitating unaccounted backtracking that would erase the reported qubit advantage for horizon-wide instances.
  2. [Numerical results] Evaluation on large instances (results section): The abstract asserts that feasible schedules with competitive costs are obtained for instances up to 312 binary variables, yet no baseline comparisons (e.g., classical MIP solvers or prior quantum encodings), error bars, or quantitative verification of constraint satisfaction (e.g., maximum violation of ramp or reserve constraints) are reported. Without these data it is impossible to judge whether the method reliably outperforms or even matches established approaches.
minor comments (2)
  1. The abstract and introduction repeatedly use the phrase 'far fewer qubits' without a side-by-side qubit-count table comparing the Pauli-Correlation Encoding to standard binary and one-hot encodings for the same horizon length.
  2. [Pauli-Correlation Encoding] Notation for the correlation operators and the mapping from Pauli strings to binary decisions should be introduced with an explicit equation in the methods section to allow independent reproduction.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the thorough review and constructive comments on our manuscript. We address each major comment in turn below.

read point-by-point responses

  1. Referee: [Optimization loop / Leader-follower structure] Leader-follower decomposition (abstract and optimization-loop description): The scaling claim rests on the premise that every binary vector returned by the quantum optimizer admits a feasible classical dispatch satisfying all time-coupling constraints. The manuscript provides no explicit mechanism, penalty term, or post-processing step that enforces ramp limits, minimum up/down times, or reserve coupling inside the Pauli-Correlation Encoding or the quantum objective; if these constraints are enforced only in the follower LP, infeasible on/off schedules can arise, necessitating unaccounted backtracking that would erase the reported qubit advantage for horizon-wide instances.

    Authors: We thank the referee for highlighting this critical point. The manuscript describes a leader-follower structure in which the quantum optimizer proposes on/off schedules via Pauli-Correlation Encoding and the classical LP solves the dispatch subproblem. However, we acknowledge that the text does not provide an explicit description of how time-coupling constraints (ramping, minimum up/down times, reserve coupling) are prevented from producing infeasible proposals or how any resulting backtracking is managed. We will revise the optimization-loop section to include a precise account of the full procedure, any penalty or feedback terms used to guide the quantum optimizer, and an assessment of the computational overhead for the reported instance sizes. revision: yes

  2. Referee: [Numerical results] Evaluation on large instances (results section): The abstract asserts that feasible schedules with competitive costs are obtained for instances up to 312 binary variables, yet no baseline comparisons (e.g., classical MIP solvers or prior quantum encodings), error bars, or quantitative verification of constraint satisfaction (e.g., maximum violation of ramp or reserve constraints) are reported. Without these data it is impossible to judge whether the method reliably outperforms or even matches established approaches.

    Authors: We agree that the numerical results section would be strengthened by additional quantitative information. In the revised manuscript we will add direct comparisons with classical MIP solvers on the same test instances, report performance statistics with error bars obtained from multiple independent runs of the quantum optimizer, and include explicit metrics (e.g., maximum observed violation of ramp and reserve constraints) to document constraint satisfaction. These additions will enable a clearer evaluation of the method against established baselines. revision: yes

Circularity Check

0 steps flagged

No significant circularity; encoding and hybrid structure are independent contributions

full rationale

The paper presents Pauli-Correlation Encoding as a new technical method to represent binary on/off decisions with fewer qubits than one-qubit-per-variable encodings. The leader-follower hybrid loop is described as quantum supplying binaries and classical solving continuous dispatch, with evaluation on instances up to 312 variables producing feasible schedules. No equations, fitted parameters, or self-citation chains are shown that reduce any claimed scaling or performance result to a quantity defined by the method itself. The derivation remains self-contained against external benchmarks and does not rely on renaming known results or smuggling ansatzes via prior self-work.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The central claim rests on the validity of the leader-follower decomposition and the correctness of the Pauli-Correlation Encoding for preserving feasibility; both are domain and algorithmic assumptions without external verification supplied in the abstract.

axioms (1)
  • domain assumption Binary on/off decisions can be separated from continuous power dispatch without loss of global optimality in the presence of the listed operational constraints.
    The leader-follower structure in the abstract presupposes this separability.
invented entities (1)
  • Pauli-Correlation Encoding no independent evidence
    purpose: Represent multiple binary variables using correlations among Pauli operators to reduce qubit count.
    Presented as the key innovation enabling scaling; no independent evidence or external reference is given in the abstract.

pith-pipeline@v0.9.0 · 5732 in / 1385 out tokens · 66289 ms · 2026-05-20T14:34:47.033342+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

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

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.

Reference graph

Works this paper leans on

41 extracted references · 41 canonical work pages · 1 internal anchor

  1. [1]

    A comparative study of hybrid quantum–classical algo- rithms for the deterministic unit commitment problem,

    S. E. Doku, “A comparative study of hybrid quantum–classical algo- rithms for the deterministic unit commitment problem,” 2025

    work page 2025

  2. [2]

    Optimization based methods for unit commitment: Lagrangian relaxation versus general mixed integer programming,

    X. Guan, Q. Zhai, and A. Papalexopoulos, “Optimization based methods for unit commitment: Lagrangian relaxation versus general mixed integer programming,” in2003 IEEE power engineering society general meeting (IEEE cat. no. 03CH37491), vol. 2. IEEE, 2003, pp. 1095–1100

    work page 2003

  3. [3]

    On the complexity of the unit commitment problem,

    P. Bendotti, P. Fouilhoux, and C. Rottner, “On the complexity of the unit commitment problem,”Annals of Operations Research, vol. 274, no. 1, pp. 119–130, 2019

    work page 2019

  4. [4]

    On mixed-integer pro- gramming formulations for the unit commitment problem,

    B. Knueven, J. Ostrowski, and J.-P. Watson, “On mixed-integer pro- gramming formulations for the unit commitment problem,”INFORMS Journal on Computing, vol. 32, no. 4, pp. 857–876, 2020

    work page 2020

  5. [5]

    A computationally efficient mixed-integer linear formulation for the thermal unit commitment problem,

    M. Carri ´on and J. M. Arroyo, “A computationally efficient mixed-integer linear formulation for the thermal unit commitment problem,”IEEE Transactions on power systems, vol. 21, no. 3, pp. 1371–1378, 2006

    work page 2006

  6. [6]

    A practical mixed integer linear programming based approach for unit commitment,

    G. W. Chang, Y . Tsai, C. Lai, and J. Chung, “A practical mixed integer linear programming based approach for unit commitment,” inIEEE Power Engineering Society General Meeting, 2004.IEEE, 2004, pp. 221–225

    work page 2004

  7. [7]

    Price-based unit commitment: A case of lagrangian relaxation versus mixed integer programming,

    T. Li and M. Shahidehpour, “Price-based unit commitment: A case of lagrangian relaxation versus mixed integer programming,”IEEE transactions on power systems, vol. 20, no. 4, pp. 2015–2025, 2005

    work page 2015

  8. [8]

    Tight mixed integer linear programming formulations for the unit commitment problem,

    J. Ostrowski, M. F. Anjos, and A. Vannelli, “Tight mixed integer linear programming formulations for the unit commitment problem,”IEEE transactions on power systems, vol. 27, no. 1, pp. 39–46, 2011

    work page 2011

  9. [9]

    A comparison between mixed-integer linear programming and dynamic programming with state prediction as novelty for solving unit commitment,

    D. Putz, D. Schwabeneder, H. Auer, and B. Fina, “A comparison between mixed-integer linear programming and dynamic programming with state prediction as novelty for solving unit commitment,”International Journal of Electrical Power & Energy Systems, vol. 125, p. 106426, 2021

    work page 2021

  10. [10]

    Quantum computing for finance,

    D. Herman, C. Googin, X. Liu, Y . Sun, A. Galda, I. Safro, M. Pistoia, and Y . Alexeev, “Quantum computing for finance,”Nature Reviews Physics, vol. 5, no. 8, pp. 450–465, 2023

    work page 2023

  11. [11]

    Network community detection on small quantum computers,

    R. Shaydulin, H. Ushijima-Mwesigwa, I. Safro, S. Mniszewski, and Y . Alexeev, “Network community detection on small quantum computers,”Advanced Quantum Technologies, vol. 2, no. 9, p. 1900029,

    work page

  12. [12]

    Available: https://dx.doi.org/10.1002/qute.201900029

    [Online]. Available: https://dx.doi.org/10.1002/qute.201900029

    work page doi:10.1002/qute.201900029

  13. [13]

    Recent advances in quantum computing for drug discovery and development,

    G. Kumar, S. Yadav, A. Mukherjee, V . Hassija, and M. Guizani, “Recent advances in quantum computing for drug discovery and development,” IEEE Access, vol. 12, pp. 64 491–64 509, 2024

    work page 2024

  14. [14]

    A hybrid classical-quantum approach to highly constrained unit commitment problems,

    B. Salgado, A. Sequeira, and L. P. Santos, “A hybrid classical-quantum approach to highly constrained unit commitment problems,”arXiv preprint arXiv:2412.11312, 2024

    work page arXiv 2024

  15. [15]

    A new hybrid quantum-classical algorithm for solving the unit commitment problem,

    W. Aboumrad, P. R. V . Marthi, S. Debnath, M. Roetteler, and E. Epifanovsky, “A new hybrid quantum-classical algorithm for solving the unit commitment problem,” 2025. [Online]. Available: https://arxiv.org/abs/2505.00145

    work page arXiv 2025

  16. [16]

    Adapting quantum approximation optimization algorithm (qaoa) for unit commitment,

    S. Koretsky, P. Gokhale, J. M. Baker, J. Viszlai, H. Zheng, N. Gurung, R. Burg, E. A. Paaso, A. Khodaei, R. Eskandarpour, and F. T. Chong, “Adapting quantum approximation optimization algorithm (qaoa) for unit commitment,” 2021. [Online]. Available: https://arxiv.org/abs/2110.12624

    work page arXiv 2021

  17. [17]

    A new method for unit commitment with ramping constraints,

    W. Fan, X. Guan, and Q. Zhai, “A new method for unit commitment with ramping constraints,”Electric Power Systems Research, vol. 62, no. 3, pp. 215–224, 2002

    work page 2002

  18. [18]

    Towards large-scale quantum optimization solvers with few qubits,

    M. Sciorilli, L. Borges, T. L. Patti, D. Garc ´ıa-Mart´ın, G. Camilo, A. Anandkumar, and L. Aolita, “Towards large-scale quantum optimization solvers with few qubits,”Nature Communications, vol. 16, no. 1, Jan. 2025. [Online]. Available: http://dx.doi.org/10.1038/ s41467-024-55346-z

    work page 2025

  19. [19]

    Large-scale portfolio optimization using pauli correlation encoding,

    V . P. Soloviev and M. Krompiec, “Large-scale portfolio optimization using pauli correlation encoding,”arXiv preprint arXiv:2511.21305, 2025

    work page arXiv 2025

  20. [20]

    Pauli correla- tion encoding for budget-constrained optimization,

    J. Pad ´ın-Mart´ınez, V . P. Soloviev, A. Borrallo-Rentero, A. Rodr ´ıguez- Otero, R. Alfonso-Rodr ´ıguez, and M. Krompiec, “Pauli correla- tion encoding for budget-constrained optimization,”arXiv preprint arXiv:2602.17479, 2026

    work page arXiv 2026

  21. [21]

    Warm-starting pce for traveling salesman problem,

    R. S. do Carmo, R. G. dos Reis, S. F. F Silva, L. G. E Arruda, F. F. Fanchiniet al., “Warm-starting pce for traveling salesman problem,” Brazilian Journal of Physics, vol. 56, no. 1, p. 49, 2026

    work page 2026

  22. [22]

    An overview of bilevel opti- mization,

    B. Colson, P. Marcotte, and G. Savard, “An overview of bilevel opti- mization,”Annals of Operations Research, vol. 153, no. 1, pp. 235–256, 2007

    work page 2007

  23. [23]

    Dempe,F oundations of Bilevel Programming

    S. Dempe,F oundations of Bilevel Programming. Springer, 2002

    work page 2002

  24. [24]

    J. F. Bard,Practical Bilevel Optimization: Algorithms and Applications. Springer, 1998

    work page 1998

  25. [25]

    , author Banjac, G

    B. Stellato, G. Banjac, P. Goulart, A. Bemporad, and S. Boyd, “OSQP: an operator splitting solver for quadratic programs,”Mathematical Programming Computation, vol. 12, no. 4, pp. 637–672, 2020. [Online]. Available: https://doi.org/10.1007/s12532-020-00179-2

    work page doi:10.1007/s12532-020-00179-2 2020

  26. [27]

    J. F. Bonnans and A. Shapiro,Perturbation analysis of optimization problems. Springer Science & Business Media, 2013

    work page 2013

  27. [28]

    Optnet: Differentiable optimization as a layer in neural networks,

    B. Amos and J. Z. Kolter, “Optnet: Differentiable optimization as a layer in neural networks,” inInternational conference on machine learning. PMLR, 2017, pp. 136–145

    work page 2017

  28. [29]

    Differentiable convex optimization layers,

    A. Agrawal, B. Amos, S. Barratt, S. Boyd, S. Diamond, and J. Z. Kolter, “Differentiable convex optimization layers,”Advances in neural information processing systems, vol. 32, 2019

    work page 2019

  29. [30]

    Quantum circuit learning

    K. Mitarai, M. Negoro, M. Kitagawa, and K. Fujii, “Quantum circuit learning,”Physical Review A, vol. 98, no. 3, Sep. 2018. [Online]. Available: http://dx.doi.org/10.1103/PhysRevA.98.032309

    work page doi:10.1103/physreva.98.032309 2018

  30. [31]

    Gradients of parameterized quantum gates using the parameter-shift rule and gate decomposition

    G. E. Crooks, “Gradients of parameterized quantum gates using the parameter-shift rule and gate decomposition,”arXiv preprint arXiv:1905.13311, 2019

    work page internal anchor Pith review Pith/arXiv arXiv 1905

  31. [32]

    Evaluating analytic gradients on quantum hardware,

    M. Schuld, V . Bergholm, C. Gogolin, J. Izaac, and N. Killoran, “Evaluating analytic gradients on quantum hardware,”Phys. Rev. A, vol. 99, p. 032331, Mar 2019. [Online]. Available: https: //link.aps.org/doi/10.1103/PhysRevA.99.032331

    work page doi:10.1103/physreva.99.032331 2019

  32. [33]

    Pauli correlation encoding to reduce maxcut requirements

    “Pauli correlation encoding to reduce maxcut requirements.” [Online]. Available: https://quantum.cloud.ibm.com/docs/en/tutorials/ pauli-correlation-encoding-for-qaoa

    work page

  33. [34]

    Quantum vision transformers

    E. A. Cherrat, I. Kerenidis, N. Mathur, J. Landman, M. Strahm, and Y . Y . Li, “Quantum vision transformers,”Quantum, vol. 8, no. arXiv: 2209.08167, p. 1265, 2024

    work page arXiv 2024

  34. [35]

    Efficientsu2

    “Efficientsu2.” [Online]. Available: https://quantum.cloud.ibm.com/docs/ en/api/qiskit/qiskit.circuit.library.EfficientSU2

    work page

  35. [36]

    Ibm ilog cplex optimization studio v22.1,

    IBM, “Ibm ilog cplex optimization studio v22.1,”International Business Machines Corporation, 2022

    work page 2022

  36. [37]

    Mlqaoa: Graph learning accelerated hybrid quantum-classical multilevel qaoa,

    B. Bach, J. Falla, and I. Safro, “Mlqaoa: Graph learning accelerated hybrid quantum-classical multilevel qaoa,” in2024 IEEE International Conference on Quantum Computing and Engineering (QCE), vol. 01, 2024, pp. 1–12

    work page 2024

  37. [38]

    Hybrid quantum- classical multilevel approach for maximum cuts on graphs,

    A. Angone, X. Liu, R. Shaydulin, and I. Safro, “Hybrid quantum- classical multilevel approach for maximum cuts on graphs,” in2023 IEEE High Performance Extreme Computing Conference (HPEC). IEEE, 2023, pp. 1–7

    work page 2023

  38. [39]

    A multilevel approach for solving large-scale qubo problems with noisy hybrid quantum approximate optimization,

    F. B. Maciejewski, B. G. Bach, M. Dupont, P. A. Lott, B. Sundar, D. E. B. Neira, I. Safro, and D. Venturelli, “A multilevel approach for solving large-scale qubo problems with noisy hybrid quantum approximate optimization,” in2024 IEEE High Performance Extreme Computing Conference (HPEC). IEEE, 2024, pp. 1–10

    work page 2024

  39. [40]

    Graph representation learning for parameter transferability in quantum approximate optimiza- tion algorithm,

    J. Falla, Q. Langfitt, Y . Alexeev, and I. Safro, “Graph representation learning for parameter transferability in quantum approximate optimiza- tion algorithm,”Quantum Machine Intelligence, vol. 6, no. 2, p. 46, 2024

    work page 2024

  40. [41]

    Similarity-based parameter transferability in the quantum approximate optimization algorithm,

    A. Galda, E. Gupta, J. Falla, X. Liu, D. Lykov, Y . Alexeev, and I. Safro, “Similarity-based parameter transferability in the quantum approximate optimization algorithm,”Frontiers in Quantum Science and Technology, vol. 2, 2023. [Online]. Available: https: //www.frontiersin.org/articles/10.3389/frqst.2023.1200975

    work page doi:10.3389/frqst.2023.1200975 2023

  41. [42]

    Qaoa-gpt: Efficient generation of adaptive and regular quantum approximate optimization algorithm circuits,

    I. Tyagin, M. H. Farag, K. Sherbert, K. Shirali, Y . Alexeev, and I. Safro, “QAOA-GPT: Efficient Generation of Adaptive and Regular Quantum Approximate Optimization Algorithm Circuits,”accepted in IEEE Quan- tum Computing and Engineering, arXiv preprint arXiv:2504.16350, 2025

    work page arXiv 2025