pith. sign in

arxiv: 2604.11534 · v1 · submitted 2026-04-13 · 🪐 quant-ph

Quantum circuit optimization for arbitrary high-dimensional bipartite quantum computation

Gui-Long Jiang , Hai-Rui Wei This is my paper

Pith reviewed 2026-05-10 15:59 UTC · model grok-4.3

classification 🪐 quant-ph
keywords quantum circuitshigh-dimensional quantum computationquNit-quMit gatescontrolled-increment gatescircuit synthesisuniversal gate setsquantum gate decomposition
0
0 comments X

The pith

Controlled-increment gates plus local operations form a universal set that realizes any quNit-quMit gate with an O(n²) upper bound on CINC count.

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

The paper develops an explicit decomposition that builds the circuit for an arbitrary unitary acting on one n-level system and one m-level system from controlled-increment gates and single-qudit local gates. This construction proves the two families together are universal for bipartite high-dimensional quantum computation. The resulting bound of O(n²) CINC gates improves the best previously known count, and the special case of a controlled quNit-quMit unitary collapses to exactly two CINC gates rather than 2n. A reader would care because fewer non-local gates directly lower the experimental cost of running high-dimensional algorithms on physical hardware.

Core claim

We exhibit a synthesis procedure showing that controlled-increment gates together with local gates are universal for any bipartite quNit-quMit unitary. The procedure produces a circuit using at most O(n²) controlled-increment gates for a general gate and only two controlled-increment gates when the target is itself a controlled quNit-quMit operation.

What carries the argument

The synthesis scheme that systematically decomposes an arbitrary quNit-quMit unitary into a sequence of controlled-increment (CINC) gates interleaved with local single-qudit operations.

If this is right

  • Any unitary on an n-by-m qudit pair admits a circuit of quadratic CINC complexity.
  • Controlled operations between high-dimensional systems become especially cheap, requiring only two CINC gates.
  • High-dimensional quantum algorithms gain a concrete universal gate library whose non-local cost scales quadratically rather than linearly in dimension.
  • The previous linear-factor overhead for controlled high-dimensional gates is eliminated.

Where Pith is reading between the lines

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

  • If CINC gates prove easier to implement than generic two-qudit gates in ion-trap or superconducting platforms, the quadratic bound could translate into measurable resource savings.
  • The same decomposition pattern might extend to multi-partite high-dimensional systems with comparable scaling.
  • Certain high-dimensional quantum algorithms whose bottlenecks involve controlled operations could see their resource estimates revised downward.

Load-bearing premise

Controlled-increment gates and local gates can be realized physically with acceptable overhead and the decomposition sequence incurs no hidden non-local costs for arbitrary dimensions n and m.

What would settle it

An explicit example for small n and m whose minimal CINC count exceeds O(n²), or a concrete physical platform where the proposed CINC sequence cannot be executed without additional entangling operations beyond those counted.

Figures

Figures reproduced from arXiv: 2604.11534 by Gui-Long Jiang, Hai-Rui Wei.

Figure 1
Figure 1. Figure 1: Quantum circuit symbols. (a) Controlled unitary gate with the control system state [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: (b) shows an equivalent quantum circuit of Ck(e iDm) based on equation (19). Here Ck(X† m) can be implemented by Ck(Xm) and local operations from figure 1(c). Hence, combining with figure 2(a), local operations and two Ck(Xm) are sufficient to implement Ck(U) [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Equivalent quantum circuit for the uniformly controlled unitary gate [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: (a) Equivalent quantum circuit for the uniformly controlled [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Quantum circuit for the non-local operation [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Quantum circuit for general unitary operations on [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Experimental setup of controlled increment gates for [PITH_FULL_IMAGE:figures/full_fig_p012_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Code for calculating CINC gate counts in Wolfram Mathematica. [PITH_FULL_IMAGE:figures/full_fig_p014_8.png] view at source ↗
read the original abstract

Implementation of high-dimensional (HD) quantum gates shows very promising perspectives for HD quantum computation. A bipartite quantum system with arbitrary dimensions $n$ and $m$ is termed a quNit-quMit. Here we propose a synthesis scheme to construct the quantum circuit for general quNit-quMit gates with controlled increment (CINC) gates and local gates. This shows that CINC gates combined with local gates form a universal gate set for HD quantum computation. An upper bound of $O(n^2)$ CINC gates is achieved for arbitrary quNit-quMit gate implementation in the proposed scheme, which is the best known result. Especially for the controlled quNit-quMit gates, our scheme requires only 2 CINC gates, whereas the previous scheme required $2n$.

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 proposes a synthesis scheme to implement arbitrary unitaries on bipartite quNit-quMit systems using controlled-increment (CINC) gates together with local gates. It asserts that this combination forms a universal gate set for high-dimensional quantum computation and derives a constructive upper bound of O(n²) CINC gates for any such unitary, with the special case that every controlled quNit-quMit gate can be realized with exactly two CINC gates (improving on a prior 2n bound).

Significance. If the explicit, uniform construction holds for every unitary, the result would supply the tightest published CINC-count upper bound for arbitrary high-dimensional bipartite operations and a particularly efficient controlled-gate decomposition. The constructive character of the bound and the claimed universality constitute the primary strengths.

major comments (2)
  1. [Main synthesis construction (section containing the general-gate theorem)] The general decomposition that is asserted to achieve the O(n²) CINC upper bound for arbitrary unitaries is not supplied with an explicit, uniform sequence or algorithm whose gate count can be verified to remain O(n²) independently of the entries of the target matrix U. Without this, the headline bound cannot be confirmed to apply to every unitary on the n-by-m space rather than to a dense subset.
  2. [Controlled-gate construction subsection] The claim that any controlled quNit-quMit gate requires only two CINC gates (abstract and controlled-gate subsection) rests on a two-gate sequence that must be shown to work uniformly for every possible controlled operation; the manuscript does not demonstrate that the same two CINC plus locals suffice when the controlled unitary is arbitrary rather than diagonal or otherwise special.
minor comments (2)
  1. A side-by-side comparison table of CINC counts versus prior schemes for representative (n,m) pairs would clarify the improvement.
  2. The notation quNit-quMit and the precise definition of the CINC gate should be stated explicitly in the preliminaries before the main claims.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments on our manuscript. We address the major comments point by point below. Where the presentation can be strengthened by additional explicit details, we will revise the manuscript accordingly.

read point-by-point responses

  1. Referee: [Main synthesis construction (section containing the general-gate theorem)] The general decomposition that is asserted to achieve the O(n²) CINC upper bound for arbitrary unitaries is not supplied with an explicit, uniform sequence or algorithm whose gate count can be verified to remain O(n²) independently of the entries of the target matrix U. Without this, the headline bound cannot be confirmed to apply to every unitary on the n-by-m space rather than to a dense subset.

    Authors: We appreciate the referee's emphasis on uniformity. The synthesis proceeds by first decomposing an arbitrary unitary on the n-by-m space into a product of at most O(n m) controlled unitaries (via a fixed-basis decomposition that holds for every matrix), followed by the two-CINC implementation of each controlled gate. Because the number of controlled gates in this decomposition is bounded by O(n m) independently of the specific entries of U, the total CINC count remains O(n²) for all unitaries (taking m ≤ n without loss of generality). To make the algorithm fully explicit and verifiable, we will insert a pseudocode description of the full procedure in the revised general-gate theorem section. revision: yes

  2. Referee: [Controlled-gate construction subsection] The claim that any controlled quNit-quMit gate requires only two CINC gates (abstract and controlled-gate subsection) rests on a two-gate sequence that must be shown to work uniformly for every possible controlled operation; the manuscript does not demonstrate that the same two CINC plus locals suffice when the controlled unitary is arbitrary rather than diagonal or otherwise special.

    Authors: We thank the referee for this observation. The two-CINC construction relies on using one CINC to align the control basis state and a second CINC (with appropriately chosen local unitaries on the target) to apply the arbitrary target unitary in the selected subspace; the same sequence works for any unitary on the target because the local gates are chosen to realize that unitary exactly when the control is active. We acknowledge that the current subsection presents the argument primarily through examples and does not contain a fully general proof for non-diagonal cases. We will revise the subsection to include an explicit general argument and circuit diagram that confirms the construction holds uniformly for every controlled unitary. revision: yes

Circularity Check

0 steps flagged

No significant circularity; bound follows from explicit construction

full rationale

The paper advances an explicit synthesis scheme that decomposes arbitrary unitaries on the n-by-m space into CINC gates plus local operations, directly yielding the O(n^2) CINC upper bound and the 2-CINC controlled case as consequences of the decomposition sequence itself. No self-definitional loops, fitted parameters renamed as predictions, or load-bearing self-citations are present; the universality statement and gate-count claims rest on the constructive algorithm rather than on any prior result by the same authors or on tautological re-labeling. The derivation is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on standard quantum-information assumptions about gate universality and decomposition; no new free parameters or invented entities are introduced.

axioms (1)
  • domain assumption CINC gates together with local gates form a universal gate set for high-dimensional quantum computation.
    Invoked to establish that the proposed scheme can implement any quNit-quMit gate.

pith-pipeline@v0.9.0 · 5421 in / 1179 out tokens · 47320 ms · 2026-05-10T15:59:09.615476+00:00 · methodology

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

58 extracted references · 58 canonical work pages · 1 internal anchor

  1. [1]

    Quantum circuit optimization for arbitrary high-dimensional bipartite quantum computation

    GCX and CINC gates to synthesize a generaln-qutrit gate [43]. However, this scheme uses two types of imprimitive gates. Forn-ququart (4-level) systems, Liet al.[38] proposed a scheme to construct a quantum circuit of generaln-ququart gates, where 5(42(n−1)−4n−1) CDNOT gates are required. The theoretical lower bound of [d2n−n(d2−1)−1]/[4(d−1)] GCX gates fo...

    work page internal anchor Pith review Pith/arXiv arXiv 2026

  2. [2]

    Ifn 1 = 1 andn 2 >1 (i.e.,n 2 = 2), it is only necessary to decomposeU 2, in whichW 1 =U 1 andW ′ 1 =V 1 =I m in equation (31)

    in equation (31), from equations (26) and (28), we have W1 0 0W 2 =   W11 0 0 0 0W 12 0 0 0 0W 21 0 0 0 0W 22   .(34) whereW 11 ∈U(⌊ n1 2 ⌋m),W 12 ∈U((n 1 − ⌊ n1 2 ⌋)m),W 21 ∈U(⌊ n2 2 ⌋m), andW 22 ∈U((n 2 − ⌊ n2 2 ⌋)m). Ifn 1 = 1 andn 2 >1 (i.e.,n 2 = 2), it is only necessary to decomposeU 2, in whichW 1 =U 1 andW ′ 1 =V 1 =I m in equation (31). For...

    work page

  3. [3]

    eZ1 0 0 eZ2 # ·exp(−iσ 12 x2 ⊗D (5) m )·

    Here PS 1 denotes a parity sorter, which transmits even modes{|0⟩,|2⟩, . . .}and reflects odd modes{|1⟩,|3⟩, . . .}. PS 2 denotes a second-order parity sorter, which transmits modes|2k×2⟩(k= 0,1,2, . . .) and reflects modes|(2k−1)×2⟩. If the input is an odd mode, then PS2 transmits the photon with a probability of 1 2 and reflects the photon with equal pr...

    work page

  4. [4]

    Dimension of the first system n=

    Based on equation (B11), we can get the quantum circuit ofX, as shown in figure 6. Appendix C: Algorithm for the CINC gate counts Given an arbitrary dimensionn, the Wolfram code to calculate the number of CINC required to accurately imple- ment a general quNit-quMit gate based on our scheme is shown in figure 8. In[ ]:= n=Input["Dimension of the first sy...

    work page

  5. [5]

    Adv.4eaat9304

    Hu X M, Guo Y, Liu B H, Huang Y F, Li C F and Guo G C 2018 Beating the channel capacity limit for superdense coding with entangled ququartsSci. Adv.4eaat9304

    work page 2018

  6. [6]

    Br¨ uß D and Macchiavello C 2002 Optimal Eavesdropping in Cryptography with Three-Dimensional Quantum StatesPhys. Rev. Lett.88127901

    work page 2002

  7. [7]

    Cerf N J, Bourennane M, Karlsson A and Gisin N 2002 Security of quantum Key distribution using d-level systemsPhys. Rev. Lett.88127902

    work page 2002

  8. [8]

    Bocharov A, Roetteler M and Svore K M 2017 Factoring with qutrits: Shor’s algorithm on ternary and metaplectic quantum architecturesPhys. Rev. A96012306 15

    work page 2017

  9. [9]

    Quantum Technol.31900074

    Lu H H, Hu Z, Alshaykh M S, Moore A J, Wang Y, Imany P, Weiner A M and Kais S 2019 Quantum phase estimation with time-frequency qudits in a single photonAdv. Quantum Technol.31900074

    work page 2019

  10. [10]

    Saha A, Majumdar R, Saha D, Chakrabarti A and Sur-Kolay S 2022 Asymptotically improved circuit for ad-ary Grover’s algorithm with advanced decomposition of then-qudit Toffoli gatePhys. Rev. A105062453

    work page 2022

  11. [11]

    Lanyon B P, Weinhold T J, Langford N K, O’Brien J L, Resch K J, Gilchrist A and White A G 2008 Manipulating Biphotonic QutritsPhys. Rev. Lett.100060504

    work page 2008

  12. [12]

    Paesani S, Bulmer J F F, Jones A E, Santagati R and Laing A 2021 Scheme for Universal High-Dimensional Quantum Computation with Linear OpticsPhys. Rev. Lett.126230504

    work page 2021

  13. [13]

    Commun.101678

    Soltamov V A, Kasper C, Poshakinskiy A V, Anisimov A N, Mokhov E N, Sperlich A, Tarasenko S A, Baranov P G, Astakhov G V and Dyakonov V 2019 Excitation and coherent control of spin qudit modes in silicon carbide at room temperatureNat. Commun.101678

    work page 2019

  14. [14]

    Commun.142242

    Hrmo P, Wilhelm B, Gerster L, van Mourik M W, Huber M, Blatt R, Schindler P, Monz T and Ringbauer M 2023 Native qudit entanglement in a trapped ion quantum processorNat. Commun.142242

    work page 2023

  15. [15]

    Leupold F M, Malinowski M, Zhang C, Negnevitsky V, Cabello A, Alonso J and Home J P 2018 Sustained State- Independent Quantum Contextual Correlations from a Single IonPhys. Rev. Lett.120180401

    work page 2018

  16. [16]

    Morvan A, Ramasesh V V, Blok M S, Kreikebaum J M, O’Brien K, Chen L, Mitchell B K, Naik R K, Santiago D I and Siddiqi I 2021 Qutrit Randomized BenchmarkingPhys. Rev. Lett.126210504

    work page 2021

  17. [17]

    Cervera-Lierta A, Krenn M, Aspuru-Guzik A and Galda A 2022 Experimental High-Dimensional Greenberger-Horne- Zeilinger Entanglement with Superconducting Transmon QutritsPhys. Rev. Appl.17024062

    work page 2022

  18. [18]

    Luo Ket al2023 Experimental Realization of Two Qutrits Gate with Tunable Coupling in Superconducting CircuitsPhys. Rev. Lett.130030603

    work page

  19. [19]

    Ding Y, Bacco D, Dalgaard K, Cai X, Zhou X, Rottwitt K and Oxenløwe L K 2017 High-dimensional quantum key distribution based on multicore fiber using silicon photonic integrated circuitsnpj Quantum Inform.325

    work page 2017

  20. [20]

    Doda M, Huber M, Murta G, Pivoluska M, Plesch M and Vlachou C 2021 Quantum Key Distribution Overcoming Extreme Noise: Simultaneous Subspace Coding Using High-Dimensional EntanglementPhys. Rev. Appl.15034003

    work page 2021

  21. [21]

    Bulla Let al2023 Distribution of genuine high-dimensional entanglement over 10.2 km of noisy metropolitan atmosphere Phys. Rev. A107L050402

    work page

  22. [22]

    Luo Y Het al2019 Quantum Teleportation in High DimensionsPhys. Rev. Lett.123070505

    work page

  23. [23]

    Hu X Met al2020 Experimental High-Dimensional Quantum TeleportationPhys. Rev. Lett.125230501

    work page

  24. [24]

    Adv.8 eabn9783

    Zhang Het al2022 Resource-efficient high-dimensional subspace teleportation with a quantum autoencoderSci. Adv.8 eabn9783

    work page

  25. [25]

    Bechmann-Pasquinucci H and Peres A 2000 Quantum Cryptography with 3-State SystemsPhys. Rev. Lett.853313

    work page 2000

  26. [26]

    Campbell E T 2014 Enhanced fault-tolerant quantum computing in d-level systemsPhys. Rev. Lett.113230501

    work page 2014

  27. [27]

    Krishna A and Tillich J P 2019 Towards Low Overhead Magic State DistillationPhys. Rev. Lett.123070507

    work page 2019

  28. [28]

    Imany P, Jaramillo-Villegas J A, Alshaykh M S, Lukens J M, Odele O D, Moore A J, Leaird D E, Qi M and Weiner A M 2019 High-dimensional optical quantum logic in large operational spacesnpj Quantum Inform.559

    work page 2019

  29. [29]

    Gao X, Erhard M, Zeilinger A and Krenn M 2020 Computer-Inspired Concept for High-Dimensional Multipartite Quantum GatesPhys. Rev. Lett.125050501

    work page 2020

  30. [30]

    Daboul J, Wang X and Sanders B C 2003 Quantum gates on hybrid quditsJ. Phys. A: Math. Gen.362525

    work page 2003

  31. [31]

    Klimov A B, Guzm´ an R, Retamal J C and Saavedra C 2003 Qutrit quantum computer with trapped ionsPhys. Rev. A 67062313

    work page 2003

  32. [32]

    Howard M and Vala J 2012 Qudit versions of the qubitπ/8 gatePhys. Rev. A86022316

    work page 2012

  33. [33]

    Babazadeh A, Erhard M, Wang F, Malik M, Nouroozi R, Krenn M and Zeilinger A 2017 High-Dimensional Single-Photon Quantum Gates: Concepts and ExperimentsPhys. Rev. Lett.119180510

    work page 2017

  34. [34]

    Gao X, Krenn M, Kysela J and Zeilinger A 2019 Arbitraryd-dimensional PauliXgates of a flying quditPhys. Rev. A99 023825

    work page 2019

  35. [35]

    Technol.7015016

    Wang Y, Ru S, Wang F, Zhang P and Li F 2022 Experimental demonstration of efficient high-dimensional quantum gates with orbital angular momentumQuantum Sci. Technol.7015016

    work page 2022

  36. [36]

    Su Q P, Zhang Y, Bin L and Yang C P 2022 Hybrid controlled-sum gate with one superconducting qutrit and one cat-state qutrit and application in hybrid entangled state preparationPhys. Rev. A105042434

    work page 2022

  37. [37]

    Meng Z, Liu W Q, Song B W, Wang X Y, Zhang A N and Yin Z Q 2024 Experimental realization of high-dimensional quantum gates with ultrahigh fidelity and efficiencyPhys. Rev. A109022612

    work page 2024

  38. [38]

    Phys.5134

    Lanyon B P, Barbieri M, Almeida M P, Jennewein T, Ralph T C, Resch K J, Pryde G J, O’Brien J L, Gilchrist A and White A G 2009 Simplifying quantum logic using higher-dimensional Hilbert spacesNat. Phys.5134

    work page 2009

  39. [39]

    Fedorov A, Steffen L, Baur M, da Silva M P and Wallraff A 2012 Implementation of a Toffoli gate with superconducting circuitsNature (London)481170

    work page 2012

  40. [40]

    Liu W Q, Wei H R and Kwek L C 2020 Low-cost Fredkin gate with auxiliary spacePhys. Rev. Appl.14054057

    work page 2020

  41. [41]

    Gao X, Appel P, Friis N, Ringbauer M and Huber M 2023 On the role of entanglement in qudit-based circuit compression Quantum71141

    work page 2023

  42. [42]

    Li W D, Gu Y J, Liu K, Lee Y H and Zhang Y Z 2013 Efficient universal quantum computation with auxiliary Hilbert spacePhys. Rev. A88034303

    work page 2013

  43. [43]

    Barenco A, Bennett C H, Cleve R, DiVincenzo D P, Margolus N, Shor P, Sleator T, Smolin J A and Weinfurter H 1995 Elementary gates for quantum computationPhys. Rev. A523457

    work page 1995

  44. [44]

    Brylinski J L and Brylinski R 2001 Universal quantum gates arXiv:quant-ph/0108062 16

    work page arXiv 2001

  45. [45]

    Di Y M and Wei H R 2013 Synthesis of multivalued quantum logic circuits by elementary gatesPhys. Rev. A87012325

    work page 2013

  46. [46]

    Comput.6436

    Brennen G K, Bullock S S and O’Leary D P 2006 Efficient circuits for exact-universal computations with quditsQuantum Inf. Comput.6436

    work page 2006

  47. [47]

    Quantum Technol.72400033

    Jiang G L, Liu W Q and Wei H R 2024 Optimal Quantum Circuits for General Multi-Qutrit Quantum ComputationAdv. Quantum Technol.72400033

    work page 2024

  48. [48]

    Di Y M and Wei H R 2015 Optimal synthesis of multivalued quantum circuitsPhys. Rev. A92062317

    work page 2015

  49. [49]

    Comput.9423

    Nakajima Y, Kawano Y, Sekigawa H, Nakanishi M, Yamashita S and Nakashima Y 2009 Synthesis of quantum circuits for d-level systems by using cosine-sine decompositionQuantum Inf. Comput.9423

    work page 2009

  50. [50]

    Bullock S S, O’Leary D P and Brennen G K 2005 Asymptotically Optimal Quantum Circuits ford-Level SystemsPhys. Rev. Lett.94230502

    work page 2005

  51. [51]

    Mansky M B, Castillo S L, Puigvert V R and Linnhoff-Popien C 2023 Near-optimal quantum circuit construction via Cartan decompositionPhys. Rev. A108052607

    work page 2023

  52. [52]

    Shende V V, Markov I L and Bullock S S 2004 Minimal universal two-qubit controlled-NOT-based circuitsPhys. Rev. A 69062321

    work page 2004

  53. [53]

    CAD251000

    Shende V V, Bullock S S and Markov I L 2006 Synthesis of Quantum-Logic CircuitsIEEE Trans. CAD251000

    work page 2006

  54. [54]

    Reck M, Zeilinger A, Bernstein H J and Bertani P 1994 Experimental realization of any discrete unitary operatorPhys. Rev. Lett.7358

    work page 1994

  55. [55]

    Clements W R, Humphreys P C, Metcalf B J, Kolthammer W S and Walmsley I A 2016 Optimal design for universal multiport interferometersOptica31460

    work page 2016

  56. [56]

    Meng H 2022 Deterministic linear-optical quantum control gates utilizing path and polarization degrees of freedomPhys. Rev. A105032607

    work page 2022

  57. [57]

    Paige C C and Wei M 1994 History and generality of the CS decompositionLinear Algebra Appl.208-209303

    work page 1994

  58. [58]

    Brennen G K, O’Leary D P and Bullock S S 2005 Criteria for exact qudit universalityPhys. Rev. A71052318

    work page 2005