Pith. sign in

REVIEW 4 major objections 6 minor 60 references

Quantum-circuit compilation restricted to unitarily equivalent circuits exhibits a genuine 2D Ising phase transition, with the equivalence constraint as the source of criticality.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

T0 review · deepseek-v4-flash

2026-08-04 01:04 UTC pith:QSACP4SQ

load-bearing objection Promising mapping of compilation to a spin model, but the 2D-Ising claim rests on post-selected annealing runs and three small sizes. the 4 major comments →

arxiv 2608.00189 v1 pith:QSACP4SQ submitted 2026-07-31 quant-ph cond-mat.stat-mech

Phase transitions in quantum-circuit compilation

classification quant-ph cond-mat.stat-mech MSC 81P6882B2082B27
keywords quantum-circuit compilationphase transitionIsing universalitysimulated annealingcrosstalkunitary equivalencespin-lattice mappingKolmogorov complexity
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper argues that compiling a quantum circuit into a hardware-efficient form is not just an engineering task but a thermodynamic process with genuine phase structure. For the paradigmatic case of a reversal permutation compiled into nearest-neighbor SWAP gates under an infidelity cost with crosstalk, it finds a second-order transition from a disordered ensemble of equivalent circuits to an antiferromagnetic brick-wall ordered phase whose critical exponents match the two-dimensional Ising universality class (T_c ≈ 1.01). It further shows that this order disappears when the unitary-equivalence constraint is lifted, making equivalence itself the source of criticality, and that stronger crosstalk produces higher-period Z_n ordered regimes. This matters because it predicts qualitative changes in the structure of compiled circuits and links compilation hardness to statistical mechanics.

Core claim

Within the equivalence class of circuits implementing the reversal permutation, the infidelity cost maps to a spin Hamiltonian on a lattice whose sites are qubit links at each time layer. The equivalence rules induce inter-layer couplings, so the constrained ensemble behaves as a two-dimensional interacting spin system. At crosstalk radius A ≈ 1, the low-temperature equilibrium is a brick-wall pattern of SWAPs, an antiferromagnetic spin state; finite-size scaling of the (π,π) structure factor gives T_c ≈ 1.01 ± 0.05, β ≈ 0.09 ± 0.03, ν ≈ 1.14 ± 0.16, consistent with the 2D Ising universality class. Removing the equivalence constraint destroys these phases; the paper therefore claims that equ

What carries the argument

The link-spin representation of circuits: each two-qubit gate on neighboring qubits becomes a spin-up site on a time-layer × qubit-link lattice, idle qubits require adjacent down spins, and local rewrite rules that preserve the circuit unitary become the updates of a Monte Carlo simulation. The infidelity cost is rewritten as H_infidelity, with a density term, a long-range (A/Δb)^6 crosstalk repulsion within each layer, and an empty-layer subtraction; the equivalence rules supply the effective inter-layer couplings that make the constrained ensemble two-dimensional. The order parameter is the static structure factor |S(π,π)|, analyzed with a finite-size scaling ansatz.

Load-bearing premise

The geometric simulated-annealing runs are treated as thermal-equilibrium samples at each nominal temperature, and post-selecting runs that reach the minimum-energy state is assumed not to bias the order parameter; if the chains do not thermalize, the quoted exponents describe the annealer's dynamics, not an equilibrium transition.

What would settle it

Run the same annealing protocol with a cooling schedule ten times slower (or with much longer runs at fixed temperature near T_c) and with an independent sampler such as parallel tempering; if the structure-factor curves and the fitted T_c, β, ν shift outside the quoted errors, the claim of an equilibrium Ising transition fails. Exact enumeration of the equivalence class for Nq=6 would also allow a direct computation of the true partition function and a check of whether the (π,π) susceptibility diverges at T_c.

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

If this is right

  • Compiled circuits for the reversal permutation at weak crosstalk will generically be brick-wall (antiferromagnetic) patterns, not arbitrary low-depth circuits, once temperature is lowered below T_c ≈ 1.01.
  • The transition's universality means small-register behavior extrapolates: the Ising data collapse implies critical exponents and T_c for larger qubit numbers.
  • Without the unitary-equivalence constraint, no phase transition occurs; the empty circuit is the trivial minimizer, so the constraint is doing the work.
  • Stronger crosstalk pushes the compiler into serial Z3-ordered schedules; by extension, progressively serial compiled circuits are expected at higher crosstalk.
  • Adding single-qubit gates breaks the antiferromagnetic order into clusters, with cluster count growing linearly in the number of T gates, so realistic circuits retain ordered domains but of finite size.

Where Pith is reading between the lines

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

  • A corollary not stated in the paper: near T_c the compiler should exhibit critical slowing down, so an annealing schedule that crosses this region quickly could systematically avoid equilibration; measuring wall-clock hardness versus temperature would be a direct test.
  • The statistical-mechanics mapping suggests a general principle for equivalence-constrained optimization: the constraint converts a trivial optimization (empty circuit) into a system with ordered phases; this could apply to other problems, such as permutation-based layout, where search is restricted to isomorphic instances.
  • The claimed 2D Ising criticality could be checked by exact transfer-matrix or partition-function computations for the smallest registers (Nq=6,8), without relying on annealed samples.
  • If the Z_n hierarchy holds, strong-crosstalk hardware should show predictable serialization in optimal compiled schedules, giving a quantitative resource-versus-error trade-off that compiler designers could target directly.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper represents quantum-circuit compilation as a statistical-mechanics problem: circuits are mapped to spin configurations on a time-link lattice, and the hardware infidelity cost (Eq. 1) becomes a spin Hamiltonian (Eq. 2). For reversal-permutation circuits compiled with nearest-neighbor SWAP gates under crosstalk, the authors report a second-order phase transition from a disordered ensemble to an antiferromagnetic brick-wall phase, claimed to be in the 2D Ising universality class with T_c≈1.01±0.05, β≈0.09±0.03, ν≈1.14±0.16. They also report a Z3-ordered regime at larger crosstalk, persistence of Z2 order for random permutations and augmented gate sets, and state that removing the unitary-equivalence constraint destroys the phases.

Significance. If the central claim is correct, the paper provides a novel and potentially fruitful connection between quantum-circuit compilation and statistical mechanics, with implications for the structure of compilation landscapes and for constrained optimization more generally. Strengths of the manuscript include the explicit derivation of the infidelity Hamiltonian in SM Sec. II, the use of a sound and complete equational theory to move within the equivalence class, and the availability of the Vulqano software and a data-availability statement, which support reproducibility. The notion that equivalence constraints generate emergent order is conceptually interesting and testable. However, the central numerical evidence for the 2D-Ising universality and the Z3 phase is not yet convincing, for the reasons detailed below.

major comments (4)
  1. [Main text, 'Ising phase transition'; SM Sec. III] The finite-size scaling is performed on simulated-annealing runs that are post-selected for convergence to the minimum-energy state. A single annealing trajectory is not a canonical sample at fixed T, and conditioning on eventual success biases the order parameter: near the putative T_c, the successful runs have above-average |S(π,π)|, so the averaged curve mixes equilibrium magnetization with the annealing success probability. The reported T_c, β, ν may thus describe the dynamical freezing of the annealer rather than a thermodynamic transition. No autocorrelation times, fixed-T equilibration checks, or replica-exchange comparisons are reported. The authors should either perform proper equilibrium sampling (e.g., parallel tempering or long fixed-T MCMC with thermalization diagnostics) or explicitly show that post-selection does not affect the scaling, e.g., by reporting the success proba
  2. [SM Sec. III] The finite-size scaling uses only three system sizes L=N_q−1=5,7,9, with N_t≈L(L+1)/2, so the aspect ratio changes with L. With three sizes and three fitted parameters (T_c, β, ν), the data collapse is only weak evidence, and the changing aspect ratio introduces uncontrolled corrections to the isotropic 2D scaling form. The additional moving-average smoothing over 100 points further smears the curves. To support the 2D-Ising universality claim, the analysis needs larger sizes (e.g., N_q=12,14) or an anisotropic scaling ansatz, and a quantitative measure of collapse quality beyond the minimum of the cost function.
  3. [Main text, 'Z3-ordered regime'; Fig. 3(c)] The claimed Z3-ordered regime is supported only by the structure factor at k=(±2π/3, 2π/3) for a single system size (N_q=8). Given the abstract's conclusion of successive Z_n-ordered phases, a single-size signal is insufficient. At minimum, the authors should show the Z3 peak for at least two register sizes and establish a temperature window where the order is stable.
  4. [Main text, 'Quantum circuits as 2D spin lattices'; Eq. (2)] The Hamiltonian in Eq. (2) contains only intra-layer terms; the claim that the equivalence rules 'induce effective inter-layer couplings' and that the constrained ensemble behaves as a 2D interacting spin system is not derived. This is the key physical assumption behind the 2D-Ising interpretation, but it is stated as a fact rather than as an ansatz or a derived effective theory. Without an explicit computation of the effective inter-layer couplings, the finite-size collapse cannot be regarded as independent confirmation of the 2D effective description. The authors should either derive the effective couplings or explicitly label the 2D description as a conjecture.
minor comments (6)
  1. [SM Sec. II, Eq. (S4)] The relation N_idle = N_q − 2N_G assumes non-overlapping gates in each layer. This follows from the spin representation but should be stated explicitly to avoid confusion.
  2. [Main text, Eq. (2)] The term −N_q i_Idle ∏_b(1−n_{t,b}) subtracts empty layers; its derivation is relegated to the SM. A one-sentence explanation in the main text would improve readability.
  3. [Main text, 'Random permutations' and SM Sec. IV] The Kendall tau distance is called 'Kolmogorov complexity.' Since Kolmogorov complexity has a specific algorithmic-information meaning, the authors should consider using a less overloaded term such as 'Kendall tau circuit complexity' to avoid confusion.
  4. [Fig. 3(b)] The individual annealing curves with moving-average smoothing are difficult to read. Adding the sample-averaged staggered magnetization with error bars would make the statement 'lower energies correspond to larger magnetization' more quantitative.
  5. [SM Sec. III (FSS)] The errors on T_c, β, and ν are quoted without derivation. State how they are obtained (e.g., spread over the top-100 collapsed solutions or a bootstrap procedure).
  6. [SM Sec. V, Fig. S4] The linear scaling of the number of antiferromagnetic clusters with the number of T gates is explicitly left for future work; this is acceptable, but the sentence 'On average, the number of clusters increases linearly...' should be phrased as a conjecture rather than a concluded finding.

Circularity Check

0 steps flagged

No significant circularity: the Ising critical parameters are fitted outputs of a self-contained numerical experiment, and the self-citations provide methodology rather than load-bearing assumptions.

full rationale

The derivation chain is a self-contained numerical experiment. The infidelity cost Eq. (1) is mapped to H_infidelity Eq. (2) via the link-spin mapping, and the equivalence-class moves are justified by an external equational theory (Refs. [32,33]) plus the prior ECSA framework (Ref. [46]). The central quantitative claim—T_c ≈ 1.01 ± 0.05, β ≈ 0.09 ± 0.03, ν ≈ 1.14 ± 0.16—is obtained by fitting a finite-size-scaling ansatz to simulated-annealing data in SM Sec. III, so these parameters are outputs, not inputs. No equation in the paper is equivalent by construction to the claimed 2D-Ising universality class, and no fitted parameter is renamed as a prediction. The self-citations (Refs. [31], [46], [58]) supply the optimization framework and software, but the phase-transition claim is not a restatement of them; the completeness of the rewrite rules is cited to independent external work. The control statement that removing the unitary-equivalence constraint makes the phases disappear is a direct consequence of i_G > i_Idle in Eq. (1), and it is not used to derive the criticality. The main methodological risk—SM Sec. III restricts averages to 'Monte Carlo samples converging to the minimum-energy state,' which could bias the order parameter—is an equilibration/sampling-validity concern, not a circular reduction. No circular steps are identified.

Axiom & Free-Parameter Ledger

3 free parameters · 5 axioms · 0 invented entities

The model rests on hand-chosen cost coefficients, a specific r^-6 crosstalk form, and the assumption that equivalence-rule-driven annealing reaches equilibrium. The most fragile link is the unproven assertion that rewrite rules turn the one-layer Hamiltonian into a 2D Ising system.

free parameters (3)
  • Gate and idle infidelity costs i_G, i_Idle = i_G = 10, i_Idle = 1
    Hand-chosen with i_G > i_Idle; all phase diagrams use this single ratio, with no scan over cost ratios.
  • Crosstalk exponent alpha in x = (A/d)^alpha = alpha = 6
    The r^-6 form is an assumed functional form inspired by neutral-atom platforms; ordering and universality may depend on it.
  • Annealing schedule length N_steps = varied by size (15, 28, 45 layers for N_q = 6, 8, 10)
    The cooling schedule is chosen per system size without convergence checks, and the inferred critical parameters depend on it.
axioms (5)
  • domain assumption The local equivalence rules E are a sound and complete rewrite system for the gate set used, so ECSA can reach all and only equivalent circuits.
    Completeness is asserted with citations [32,33], but no proof is given that the specific SWAP rules in Fig. 2 generate the full equivalence class.
  • domain assumption Geometric simulated-annealing runs sample the canonical Boltzmann distribution of H_infidelity at each nominal temperature.
    No thermalization or ergodicity diagnostics are reported; the annealing schedule Eq. (S2) is a cooling schedule, not an equilibrium sampler.
  • domain assumption Crosstalk between simultaneous two-qubit gates is (A/d)^6 and single-qubit crosstalk is negligible.
    This platform-inspired choice determines the Hamiltonian in Eq. (2); the phase diagram is computed for this model only.
  • ad hoc to paper Equivalence rules induce effective inter-layer couplings so the constrained ensemble behaves as a 2D interacting spin system.
    The explicit Hamiltonian contains no inter-layer terms; the claimed 2D/Ising behavior relies on this asserted effective coupling.
  • standard math For permutation circuits, Kolmogorov complexity equals the minimum number of SWAP gates (Kendall tau), and the reversal permutation is maximally complex.
    For adjacent transpositions the minimum swap count equals the inversion count; reversal maximizes this count.

pith-pipeline@v1.3.0-alltime-deepseek · 14453 in / 19500 out tokens · 209275 ms · 2026-08-04T01:04:03.987727+00:00 · methodology

0 comments
read the original abstract

Quantum-circuit compilation aims at finding an optimized realization of a target circuit under given constraints, e.g., the minimization of hardware-induced errors and the unitary-equivalence of the circuit. We connect the compilation process with the thermodynamics of a many-body spin system: circuit infidelity plays the role of the energy function and low-temperature states correspond to compiled circuits. In the paradigmatic case where crosstalk between parallel gates is present, we find a phase transition between a disordered phase and an antiferromagnetic brick-wall phase, compatible with the Ising universality class. At larger crosstalk, we observe a $\mathbb{Z}_3$-ordered regime, suggesting that increasingly serial compiled circuits are associated with emergent $\mathbb{Z}_n$-ordered phases. When the unitary-equivalence constraint is removed, these phases disappear, showing that the equivalence between circuits underlies the emergent criticality and constitutes a source of complexity in quantum-circuit compilation and, more generally, in equivalence-constrained optimization. Finally, we observe that the Kolmogorov complexity of the circuit enhances the emergence of ordered phases.

Figures

Figures reproduced from arXiv: 2608.00189 by Andrea De Girolamo, Davide Rattacaso, Ilaria Siloi, Simone Montangero, Simone Notarnicola.

Figure 1
Figure 1. Figure 1: FIG. 1: Phase diagram emerging from quantum-circuit [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2: (a) Equivalence rules for circuits with [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: a. In panel (b), we plot the staggered magnetization Ms = ⟨(−1)t+bσ(t,b)⟩|t,b at A = 1, showing that optimal, equivalent circuits exhibit opposite magnetic order. Each line shows a single annealing process for the same initial circuit, with its color indicating the final energy. Lower energies correspond to larger magnetization, confirming that the optimal circuit is an antiferromagnet. The emer￾gence of a… view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4: Normalized mean size of antiferromagnetic clus [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗

discussion (0)

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

Reference graph

Works this paper leans on

60 extracted references · 3 linked inside Pith

  1. [1]

    S. A. Cook, The complexity of theorem-proving proce- dures, inProceedings of the Third Annual ACM Sympo- sium on Theory of Computing, STOC ’71 (Association for Computing Machinery, New York, NY, USA, 1971) p. 151–158

  2. [2]

    Nemhauser and L

    G. Nemhauser and L. Wolsey,Integer and Combinatorial Optimization(John Wiley & Sons, Ltd, 1988)

  3. [3]

    R. M. Karp, Reducibility among combinatorial prob- lems, inComplexity of Computer Computations: Pro- ceedings of a symposium on the Complexity of Com- puter Computations, held March 20–22, 1972, at the IBM Thomas J. Watson Research Center, Yorktown Heights, New York, and sponsored by the Office of Naval Research, Mathematics Program, IBM World Trade Corpo...

  4. [4]

    Cheeseman, B

    P. Cheeseman, B. Kanefsky, and W. M. Taylor, Where the really hard problems are, inProceedings of the 12th International Joint Conference on Artificial Intelligence - Volume 1, IJCAI’91 (Morgan Kaufmann Publishers Inc., San Francisco, CA, USA, 1991) p. 331–337

  5. [5]

    Mitchell, B

    D. Mitchell, B. Selman, and H. Levesque, Hard and easy distributions of sat problems, inProceedings of the Tenth National Conference on Artificial Intelligence, AAAI’92 (AAAI Press, 1992) p. 459–465

  6. [6]

    Kirkpatrick and B

    S. Kirkpatrick and B. Selman, Critical behavior in the satisfiability of random boolean expressions, Science264, 1297 (1994)

  7. [7]

    Monasson, R

    R. Monasson, R. Zecchina, S. Kirkpatrick, B. Selman, and L. Troyansky, Determining computational complex- ity from characteristic ‘phase transitions’, Nature400, 133 (1999)

  8. [8]

    P. W. Anderson, Solving problems in finite time, Nature 400, 115 (1999)

  9. [9]

    C. P. Gomes and B. Selman, Satisfied with physics, Sci- ence297, 784 (2002)

  10. [10]

    M´ ezard, G

    M. M´ ezard, G. Parisi, and R. Zecchina, Analytic and al- gorithmic solution of random satisfiability problems, Sci- ence297, 812 (2002)

  11. [11]

    Onsager, Crystal statistics

    L. Onsager, Crystal statistics. i. a two-dimensional model with an order-disorder transition, Phys. Rev.65, 117 (1944)

  12. [12]

    L. P. Kadanoff, W. G¨ otze, D. Hamblen, R. Hecht, E. A. S. Lewis, V. V. Palciauskas, M. Rayl, J. Swift, D. Aspnes, and J. Kane, Static phenomena near critical points: The- ory and experiment, Rev. Mod. Phys.39, 395 (1967)

  13. [13]

    K. G. Wilson and J. Kogut, The renormalization group and theϵexpansion, Physics Reports12, 75 (1974)

  14. [14]

    Nishimori and G

    H. Nishimori and G. Ortiz,Elements of Phase Transi- tions and Critical Phenomena(Oxford University Press, 2010)

  15. [15]

    L. P. Kadanoff, Scaling laws for ising models nearT c, Physics Physique Fizika2, 263 (1966)

  16. [16]

    and M´ ezard, M., On the statistical me- chanics of optimization problems of the travelling sales- man type, J

    Vannimenus, J. and M´ ezard, M., On the statistical me- chanics of optimization problems of the travelling sales- man type, J. Physique Lett.45, 1145 (1984)

  17. [17]

    Fu and P

    Y. Fu and P. W. Anderson, Application of statistical me- chanics to np-complete problems in combinatorial opti- misation, Journal of Physics A: Mathematical and Gen- eral19, 1605 (1986)

  18. [18]

    Sherrington and S

    D. Sherrington and S. Kirkpatrick, Solvable model of a spin-glass, Phys. Rev. Lett.35, 1792 (1975)

  19. [19]

    Nishimori,Statistical Physics of Spin Glasses and In- formation Processing: An Introduction(Oxford Univer- sity Press, 2001)

    H. Nishimori,Statistical Physics of Spin Glasses and In- formation Processing: An Introduction(Oxford Univer- sity Press, 2001)

  20. [20]

    O. C. Martin, R. Monasson, and R. Zecchina, Statistical mechanics methods and phase transitions in optimization problems, Theoretical Computer Science265, 3 (2001), phase Transitions in Combinatorial Problems

  21. [21]

    Lucas, Ising formulations of many np problems, Fron- tiers in Physics2, 10.3389/fphy.2014.00005 (2014)

    A. Lucas, Ising formulations of many np problems, Fron- tiers in Physics2, 10.3389/fphy.2014.00005 (2014)

  22. [22]

    Kirkpatrick, C

    S. Kirkpatrick, C. D. Gelatt, and M. P. Vecchi, Optimiza- tion by simulated annealing, Science220, 671 (1983)

  23. [23]

    Kadowaki and H

    T. Kadowaki and H. Nishimori, Quantum annealing in the transverse ising model, Phys. Rev. E58, 5355 (1998)

  24. [24]

    Farhi, J

    E. Farhi, J. Goldstone, S. Gutmann, and M. Sipser, Quantum computation by adiabatic evolution (2000), arXiv:quant-ph/0001106 [quant-ph]

  25. [25]

    Farhi, J

    E. Farhi, J. Goldstone, S. Gutmann, J. Lapan, A. Lund- gren, and D. Preda, A quantum adiabatic evolution al- gorithm applied to random instances of an np-complete problem, Science292, 472 (2001)

  26. [26]

    G. E. Santoro, R. Martoˇ n´ ak, E. Tosatti, and R. Car, The- ory of quantum annealing of an ising spin glass, Science 295, 2427 (2002)

  27. [27]

    Albash and D

    T. Albash and D. A. Lidar, Adiabatic quantum compu- tation, Rev. Mod. Phys.90, 015002 (2018)

  28. [28]

    Hauke, H

    P. Hauke, H. G. Katzgraber, W. Lechner, H. Nishimori, and W. D. Oliver, Perspectives of quantum annealing: methods and implementations, Reports on Progress in Physics83, 054401 (2020)

  29. [29]

    Huet and D

    G. Huet and D. C. Oppen, Equations and rewrite rules: A survey, inFormal Language Theory, edited by R. V. BOOK (Academic Press, 1980) pp. 349–405

  30. [30]

    D. A. Plaisted, Equational reasoning and term rewrit- ing systems, inHandbook of Logic in Artificial Intelli- gence and Logic Programming (Vol. 1)(Oxford Univer- sity Press, Inc., USA, 1993) p. 274–364

  31. [31]

    Rattacaso, D

    D. Rattacaso, D. Jaschke, M. Ballarin, I. Siloi, and S. Montangero, Quantum algorithms for equational rea- soning, Science Advances12, eaec2736 (2026)

  32. [32]

    Cl´ ement, N

    A. Cl´ ement, N. Heurtel, S. Mansfield, S. Perdrix, and B. Valiron, A complete equational theory for quantum circuits, in2023 38th Annual ACM/IEEE Symposium on Logic in Computer Science (LICS)(2023) pp. 1–13

  33. [33]

    Cl´ ement, N

    A. Cl´ ement, N. Delorme, and S. Perdrix, Minimal equa- tional theories for quantum circuits, inProceedings of the 39th Annual ACM/IEEE Symposium on Logic in Com- puter Science, LICS ’24 (Association for Computing Ma- chinery, New York, NY, USA, 2024)

  34. [34]

    Blake, Simpler presentations for many fragments of quantum circuits (2026), arXiv:2602.09874 [quant-ph]

    C. Blake, Simpler presentations for many fragments of quantum circuits (2026), arXiv:2602.09874 [quant-ph]

  35. [35]

    Rattacaso, D

    D. Rattacaso, D. Jaschke, A. Trovato, I. Siloi, and S. Montangero, Quantum algorithms for compact poly- mer thermodynamics (2026), arXiv:2603.12334 [quant- ph]

  36. [36]

    F. T. Chong, D. Franklin, and M. Martonosi, Program- ming languages and compiler design for realistic quantum hardware, Nature549, 180 (2017)

  37. [37]

    Wille, D

    R. Wille, D. Große, D. M. Miller, and R. Drech- sler, Equivalence checking of reversible circuits, in2009 39th International Symposium on Multiple-Valued Logic (2009) pp. 324–330

  38. [38]

    Burgholzer and R

    L. Burgholzer and R. Wille, Advanced equivalence checking for quantum circuits, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Sys- 7 tems40, 1810–1824 (2021)

  39. [39]

    Peham, L

    T. Peham, L. Burgholzer, and R. Wille, Equivalence checking of quantum circuits with the zx-calculus, IEEE Journal on Emerging and Selected Topics in Circuits and Systems12, 662–675 (2022)

  40. [40]

    Sander, L

    A. Sander, L. Burgholzer, and R. Wille, Equivalence checking of quantum circuits via intermediary matrix product operator, Phys. Rev. Res.7, 023261 (2025)

  41. [41]

    Schmid, D

    L. Schmid, D. F. Locher, M. Rispler, S. Blatt, J. Zeiher, M. M¨ uller, and R. Wille, Computational capabilities and compiler development for neutral atom quantum proces- sors—connecting tool developers and hardware experts, Quantum Science and Technology9, 033001 (2024)

  42. [42]

    Sivarajah, S

    S. Sivarajah, S. Dilkes, A. Cowtan, W. Simmons, A. Edg- ington, and R. Duncan, t—ket〉: a retargetable compiler for nisq devices, Quantum Science and Technology6, 014003 (2020)

  43. [43]

    X. Zhou, S. Li, and Y. Feng, Quantum circuit transforma- tion based on simulated annealing and heuristic search, IEEE Transactions on Computer-Aided Design of Inte- grated Circuits and Systems39, 4683 (2020)

  44. [44]

    Maronese, L

    M. Maronese, L. Moro, L. Rocutto, and E. Prati, Quan- tum compiling, inQuantum Computing Environments, edited by S. S. Iyengar, M. Mastriani, and K. L. Kumar (Springer International Publishing, Cham, 2022) pp. 39– 74

  45. [45]

    J. v. d. Wetering, R. Yeung, T. Laakkonen, and A. Kissinger, Optimal compilation of parametrised quan- tum circuits, Quantum9, 1828 (2025)

  46. [46]

    Rattacaso, D

    D. Rattacaso, D. Jaschke, M. Ballarin, I. Siloi, and S. Montangero, Quantum circuit compilation with quan- tum computers, Phys. Rev. Res.7, 033268 (2025)

  47. [47]

    Bluvstein, S

    D. Bluvstein, S. J. Evered, A. A. Geim, S. H. Li, H. Zhou, T. Manovitz, S. Ebadi, M. Cain, M. Kali- nowski, D. Hangleiter, J. P. Bonilla Ataides, N. Maskara, I. Cong, X. Gao, P. Sales Rodriguez, T. Karolyshyn, G. Semeghini, M. J. Gullans, M. Greiner, V. Vuleti´ c, and M. D. Lukin, Logical quantum processor based on reconfigurable atom arrays, Nature626, 58 (2024)

  48. [48]

    P. Zhao, K. Linghu, Z. Li, P. Xu, R. Wang, G. Xue, Y. Jin, and H. Yu, Quantum crosstalk analysis for simul- taneous gate operations on superconducting qubits, PRX Quantum3, 020301 (2022)

  49. [49]

    J. J. Wesdorp, E. Hyypp¨ a, J. Andersson, J. Adam, R. Beriwal, V. Bergholm, S. Dahl, S. D. Fasciati, A. G. Friero, Z. Gao, D. Gusenkova, A. Guthrie, J. Heinsoo, T. Hiltunen, K. Holland, A. Hosseinkhani, S. Inel, J. Ikonen, S. W. Jolin, K. Juliusson, S.-G. Kim, A. Komlev, R. Kokkoniemi, O. Koskinen, J. Kylm¨ al¨ a, A. Landra, J. Lamprich, M. Lehmuskoski, N...

  50. [50]

    D. Nigg, M. M¨ uller, E. A. Martinez, P. Schindler, M. Hennrich, T. Monz, M. A. Martin-Delgado, and R. Blatt, Quantum computations on a topo- logically encoded qubit, Science345, 302 (2014), https://www.science.org/doi/pdf/10.1126/science.1253742

  51. [51]

    Parrado-Rodr ´ ıguez, C

    P. Parrado-Rodr ´ ıguez, C. Ryan-Anderson, A. Bermudez, and M. M¨ uller, Crosstalk Suppression for Fault-tolerant Quantum Error Correction with Trapped Ions, Quantum 5, 487 (2021)

  52. [52]

    C. Fang, Y. Wang, S. Huang, K. R. Brown, and J. Kim, Crosstalk suppression in individually addressed two- qubit gates in a trapped-ion quantum computer, Phys. Rev. Lett.129, 240504 (2022)

  53. [53]

    Gicev, B

    S. Gicev, B. Harper, H. Kang, M. Usman, and M. Sev- ior, Crosstalk in contemporary quantum devices (2026), arXiv:2605.26528 [quant-ph]

  54. [54]

    H. S. M. Coxeter and W. O. J. Moser,Generators and Relations for Discrete Groups, 4th ed., Ergebnisse der Mathematik und ihrer Grenzgebiete, Vol. 14 (Springer- Verlag, Berlin Heidelberg New York, 1980)

  55. [55]

    A. N. Kolmogorov, Three approaches to the quantitative definition of information, International Journal of Com- puter Mathematics2, 157 (1968)

  56. [56]

    M. G. Kendall, A new measure of rank correlation, Biometrika30, 81 (1938)

  57. [57]

    Gonzaga de Oliveira and A

    S. Gonzaga de Oliveira and A. de Abreu, A survey of heuristics for the minimum linear arrangement problem, Computer Science Review61, 100962 (2026)

  58. [58]

    Phase transitions in quantum-circuit compilation

    D. Rattacaso, M. Ballarin, D. Jaschke, I. Siloi, and S. Montangero, Vulqano: quantum-inspired compiler for quantum circuits (2024). 8 Supplemental Material for “Phase transitions in quantum-circuit compilation” CONTENTS I. Equivalence-Class Simulated Annealing 8 II. Derivation of the infidelity Hamiltonian in the link spin mapping 9 III. Finite-size scali...

  59. [59]

    Randomly choose a subset of spinsσ in sub on which an update rule σin sub ↔σ out sub ∈ Ecan be applied

  60. [60]

    Using the infidelity Hamiltonian, compute the energy of the updated configurationσ out and compare it to the energy ofσ in. 2.1. If the energy decreases, we accept the change, apply the rule, and move to the next step, using the updated configuration as the new input. 2.2. If the energy increases, we accept the change with probabilityp(k) =e − ∆E Tk , wit...