Pith. sign in

REVIEW 3 major objections 4 minor 33 references

Analysis of heralded higher-fidelity two-qubit entangling gates with self-correction

T0 review · 3 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read A PT-symmetric pair of buffer-atom-mediated CZ gates can cancel first-order errors and, after projecting the buffer atom, herald a CZ gate whose error is reduced to $10^{-4}$--$10^{-6}$.

desk verdict Genuinely new heralded-gate construction, but the headline error claim rests on an unverified antisymmetry condition; worth refereeing. read the letter →

arxiv 2501.14220 v1 pith:EMJORT5K submitted 2025-01-24 quant-ph cs.ARphysics.atom-ph

Yuan Sun This is my paper · ORCID
classification quant-phcs.ARphysics.atom-ph PACS 32.80.Qk03.67.Lx42.50.-p33.80.Rv
keywords RydbergblockadeCZgateheraldedquantumself-correctionPTsymmetrydual-railbuffer-atom-mediatederrormitigation
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

This paper proposes a heralded, probabilistic CZ gate for neutral-atom Rydberg systems. The gate runs two buffer-atom-mediated gates in superposition, one the PT-reversal of the other, so that first-order errors from common experimental imperfections enter with opposite signs. After a final $\pi/2$ pulse on the buffer atom, projecting onto $|0_b\rangle$ cancels those errors and yields the ideal CZ transformation, while projection onto $|1_b\rangle$ flags failure. Numerical searches produce waveforms satisfying this condition, and simulated errors from Rabi-frequency, detuning, and blockade-strength changes drop to the $10^{-4}$--$10^{-6}$ level, conditioned on post-selecting away spontaneous emission. The result would effectively add a hardware-level error-mitigation layer on top of existing Rydberg gates.

What carries the argument

The dual-rail BAM gate: two simultaneous buffer-atom-mediated CZ gates acting on the same two qubits, one associated with buffer state $|0_b\rangle$ and the other with $|1_b\rangle$, with Rabi frequencies $\Omega_0,\Omega_1$ and opposite detunings $\Delta_1$ and $-\Delta_1$ so that the pair is PT-related. The load-bearing identity is Eq. (1), the exact anti-symmetry of first-order deviations between the two rails; it is what converts the $|0_b\rangle$ branch into the corrected CZ and the $|1_b\rangle$ branch into a pure error signal. The second local $\pi/2$ pulse plus projective readout of the buffer atom is the heralding and self-correction mechanism.

What would settle it

Apply a common Rabi-frequency scaling error $\varepsilon$ to both rails with the Fig. 2 waveforms and measure the $|1_b\rangle$ branch amplitude; if the first-order sum $\delta_{u0}+\delta_{u1}$ is nonzero, the heralded fidelity degrades linearly in $\varepsilon$ instead of quadratically, showing Eq. (1) fails.

Watch

Extended reading notes

Core claim

The central claim is that a pair of dual-rail buffer-atom-mediated (BAM) CZ gates, related by a PT transformation (time reversal $t\to -t$ plus inversion of the Rydberg amplitude $Y\to -Y$), obey the anti-symmetry condition $\delta_{u0}=-\delta_{u1}$, $\delta_{v0}=-\delta_{v1}$, $\delta_{w0}=-\delta_{w1}$, $\delta_{z0}=-\delta_{z1}$ for first-order errors. Under that condition the post-projection $|0_b\rangle$ branch carries the ideal CZ operation and all first-order deviations are pushed into the $|1_b\rangle$ branch, so conditioning on $|0_b\rangle$ yields a gate whose error is second order in each perturbation. The author demonstrates practical waveforms for both the case of no qubit-qubit interaction and the case of ideal Rydberg blockade among all three atoms, and reports numerical simulations showing the heralded gate error at $10^{-4}$--$10^{-6}$ for Rabi-frequency, detuning, and blockade-strength errors, with success probability set by the raw BAM fidelity.

Load-bearing premise

The load-bearing premise is that every relevant error source produces exactly opposite first-order deviations in the two rails (so the $|0_b\rangle$ branch is clean), and that only no-spontaneous-emission events are kept.

Editorial extensions

If this is right

  • A CZ gate with error suppressed to roughly the square of the first-order error, i.e. $10^{-4}$--$10^{-6}$, becomes available for neutral-atom Rydberg platforms.
  • The success probability is approximately the raw BAM gate fidelity, so the scheme trades a probabilistic success for a large fidelity gain.
  • The construction works with off-resonant buffer driving and resonant qubit driving, and the symmetry argument extends to two-photon and three-photon ground-Rydberg transitions.
  • A supplementary $\pi$-phase dressing of the Rydberg level pushes residual first-order population leakage into the heralded-failure branch, so the final gate error remains quadratic.
  • Errors split into two categories: those that only reduce success probability and those that also reduce conditioned fidelity, guiding which experimental noise sources matter most.

Reading between the lines

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

  • If the anti-symmetry condition can be certified for each error channel, the same dual-rail projection could serve as a generic error-mitigation layer for other gate types and qubit platforms, not just neutral-atom CZ gates.
  • The burden of reaching $10^{-6}$ conditioned fidelity shifts to buffer-atom readout: a false herald at even the $10^{-4}$ level would directly corrupt the accepted events, so the method is only as good as the measurement.
  • The method assumes no spontaneous emission during the gate; in a real experiment the heralding cannot distinguish an error event from a lost photon, so the stated fidelity ceiling is an upper bound under near-infinite-coherence conditions.
  • A natural testable extension is to measure the $|1_b\rangle$ branch amplitude as an in-situ error signal; its magnitude should be first order in each perturbation while the $|0_b\rangle$ branch error is second order, giving a direct experimental check of Eq. (1).
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 4 minor

Summary. The paper proposes a heralded, probabilistic two-qubit CZ gate for neutral atoms built from two buffer-atom-mediated (BAM) gates in a dual-rail configuration. A buffer atom is prepared in a superposition of |0b⟩ and |1b⟩; the two components undergo BAM gates with PT-reversed detunings; a second π/2 pulse and buffer measurement herald success. If the first-order deviations of the two rails are opposite for each computational basis state (Eq. (1)), the deviations cancel in the |0b⟩ branch to first order, leaving errors that are second order. The paper presents numerically searched waveforms for two interaction geometries and simulations for Rabi-frequency, detuning, and Rydberg-blockade errors, and it anticipates conditional gate errors of 10^-4 to 10^-6, post-selected on no spontaneous emission.

Significance. If the anti-symmetry condition Eq. (1) is actually satisfied by realistic waveforms, the dual-rail self-correction scheme is an appealing mechanism: it converts first-order phase errors into second-order errors in the heralded branch while preserving the CZ operation. The algebraic cancellation logic is clear, and the use of PT symmetry to generate opposite detunings for the two rails is a genuine design idea. However, the paper does not supply a direct verification of Eq. (1) for the numerically searched waveforms, and the quantitative error claim relies on post-selection on no spontaneous emission and on a buffer-readout false-herald rate that is not part of the reported error budget. The conceptual contribution is therefore interesting but remains a proposal with preliminary numerics rather than a fully validated protocol.

major comments (3)
  1. [§3 (Eq. (1) and following)]
  2. [Post-selection and readout discussion]
  3. [Figs. 3–6 and quantitative claims]
minor comments (4)
  1. [Figure numbering]
  2. [Notation]
  3. [Typos]
  4. [Reference [28]]

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the anti-symmetry condition (Eq. 1) is a stated design requirement, and the reported fidelity gains are numerical predictions from explicit waveforms rather than fits to the target error.

full rationale

The paper's derivation is not circular. The central algebra after Eq. (1) shows that if the two dual-rail BAM gates have first-order deviations satisfying δu0=-δu1, δv0=-δv1, etc., then the |0b> branch after projection is ideal to first order. Equation (1) is presented as a design condition to be achieved by waveform search, not as a definition of the target fidelity; the reported gate error is then obtained by explicit numerical simulation of the listed waveforms in Figs. 3, 4, and 6. The PT symmetry in Eqs. (2)-(3) is a genuine external symmetry of the Schrödinger equation and provides an independent rationale for seeking opposite-detuning waveform pairs. Citations to the author's prior BAM-gate work [18,19] establish a building block, but the dual-rail construction, the PT mapping, and the performance simulations are contained in this paper; the self-citations are not load-bearing for the 10^-4 to 10^-6 claim. The main weakness is that the paper asserts, without displaying the verification, that the numerically searched waveforms satisfy Eq. (1) for common-mode Rabi-scale and detuning errors; if that anti-symmetry fails, the heralded improvement vanishes. That is an unproven assumption and a correctness risk, but it is not circularity because the conclusion is not assumed in the premise.

Assumptions & free parameters 7 free parameters · 6 assumptions · 0 invented entities

The central result rests on a designed waveform search (18 to 27 real coefficients), a two-level model, a simplified dipole-dipole interaction, first-order perturbation theory, and an explicit post-selection excluding spontaneous emission. The PT symmetry is a genuine external mathematical property, but the cancellation condition Eq. (1) is asserted for the searched waveforms rather than demonstrated in the text.

free parameters (7)
  • Omega_1 coefficients, Fig. 2 waveform = [129.82, -33.36, -11.16, 5.33, -17.69, -1.62, -8.79, 4.57, -2.18]
    Numerically searched Rabi-frequency waveform for the first BAM gate; chosen to make the dual-rail CZ gate work.
  • Omega_2 coefficients, Fig. 2 waveform = [97.16, -16.78, -33.32, -11.58, 15.95, 5.78, -9.72, 2.92, -1.82]
    Rabi-frequency waveform for the second BAM gate, found by numerical search.
  • Delta_1 coefficients, Fig. 2 waveform = [-66.80, 3.86, -63.90, 6.18, -46.78, 79.58, -2.47, -5.85, -8.46]
    Detuning waveform for the buffer atom; the PT-reversed version uses -Delta_1 for the second rail.
  • Omega_1 coefficients, Fig. 5 waveform = [57.46, -17.45, 11.37, -19.97, 2.62, -5.30]
    Waveform for the ideal-blockade three-body configuration, found by numerical search.
  • Omega_2 coefficients, Fig. 5 waveform = [69.00, -34.79, 3.38, -3.22, -1.85, 1.98]
    Second rail Rabi waveform for the ideal-blockade case.
  • Delta_1 coefficients, Fig. 5 waveform = [-33.12, 40.94, 9.90, -41.01, 22.85, -31.37]
    Buffer detuning waveform for the ideal-blockade case.
  • Gate time tau = 0.25 microseconds
    Chosen reference time that fixes the frequency scale of the Fourier waveforms.
assumptions (6)
  • domain assumption Two-level model of ground-Rydberg transition with Hamiltonian Eq. (2), states X=ground and Y=Rydberg, real symmetric Omega(t) and Delta(t).
    Used throughout for buffer and qubit atoms; ignores multi-level structure and intermediate states in two-photon transitions. Stated in the text as 'the simplest case of optically driving a two-level atom'.
  • standard math Time-reversal and parity (PT) transformation leaves the Schrodinger evolution solvable with opposite detuning, Eq. (3).
    The derivation of Eq. (3) is a mathematical property of the two-level equations with X even and Y odd; the paper uses it to justify the opposite-detuning dual-rail construction.
  • domain assumption Rydberg dipole-dipole interaction model: B=2*pi*100 MHz for buffer-qubit pairs, no qubit-qubit interaction in the Fig. 2 case, or ideal blockade between all atoms in the Fig. 5 case.
    The exact atomic interaction potential and geometry are simplified; the text acknowledges 'these calculations depend on the exact model of Rydberg dipole-dipole interaction'.
  • domain assumption First-order perturbation response: the deviations delta_u0 etc. are small and the gate responses to adverse effects are first-order phase deviations.
    The cancellation argument is first-order; the paper cites its own ref [22] for this behavior without proving it here for each error source.
  • ad hoc to paper Post-selection on the absence of spontaneous emission.
    Explicitly stated: 'we can regard the main results of this paper as based on the post-selection of spontaneous emissions do not occur'; spontaneous emission is omitted from the analysis.
  • domain assumption Existence of two Rydberg channels |r'> and |r''> (or one level with different magnetic substates) for the dual-rail buffer coupling.
    The dual-rail requires two distinguishable Rydberg interaction channels; the paper states this can be realized by Forster resonance structures or polarization degrees of freedom.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Analysis of heralded higher-fidelity two-qubit entangling gates with self-correction." pith.science (2026). https://pith.science/paper/EMJORT5K

@misc{pith2026250114220,
  author       = {Pith},
  title        = {Pith review of: Analysis of heralded higher-fidelity two-qubit entangling gates with self-correction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/EMJORT5K}},
  note         = {Machine review of arXiv:2501.14220}
}
abstract

For the quantum error correction (QEC) and noisy intermediate-scale quantum (NISQ) algorithms to function with high efficiency, the raw fidelity of quantum logic gates on physical qubits needs to satisfy strict requirement. The neutral atom quantum computing equipped with Rydberg blockade gates has made impressive progress recently, which makes it worthwhile to explore its potential in the two-qubit entangling gates, including Controlled-PHASE gate and in particular the CZ gate. Provided the quantum coherence is well preserved, improving the fidelity of Rydberg blockade gates calls for special mechanisms to deal with adverse effects caused by realistic experimental conditions. Here the heralded very-high-fidelity Rydberg blockade Controlled-PHASE gate is designed to address these issues, which contains self-correction and projection as the key steps. This trailblazing method can be built on the basis of the previously established buffer-atom-mediated gate, and a special form of symmetry under PT transformation plays a crucial role in the process. We further analyze the performance with respect to a few typical sources of imperfections. This procedure can also be regarded as quantum hardware error correction or mitigation. While this paper by itself does not cover every single subtle issue and still contains many over-simplifications, we find it reasonable to anticipate very-high-fidelity two-qubit quantum logic gate operated in the sense of heralded but probabilistic, whose gate error can reduce to the level of $10^{-4}$--$10^{-6}$ or even lower with reasonably high possibilities.

Figures

Figures reproduced from arXiv: 2501.14220 by the authors.

Figure 1
Figure 1. FIG. 1. (Color online) The (a) linkage structure and (b) ba [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. (Color online) Sample waveforms of BAM gate with [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 4
Figure 4. FIG. 4. (Color online) Numerical simulation of the heralded [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figures from the paper (3 more)
Figure 3
Figure 3. Figure 3: FIG. 3. (Color online) Numerical simulation of the heralded [PITH_FULL_IMAGE:figures/full_fig_p004_3.png]
Figure 6
Figure 6. Figure 6: FIG. 6. (Color online) Numerical simulation of the heralded [PITH_FULL_IMAGE:figures/full_fig_p005_6.png]
Figure 5
Figure 5. Figure 5: FIG. 5. (Color online) Sample waveforms of BAM gate under [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

33 extracted references · 24 canonical work pages

  1. [1]

    Suppression of higher frequency components and smooth on-off process are also implicitly included [21]

    MHz for a given reference time τ = 0 .25 µs which is also the gate time throughout the text. Suppression of higher frequency components and smooth on-off process are also implicitly included [21]. FIG. 2. (Color online) Sample waveforms of BAM gate with one buffer atom. (a) Waveforms of modulation. (b) Popula- tions of wave functions corresponding to (a)....

    work page

  2. [2]

    More specially, this feels like the linear superpo- sition of two buffer-atom-mediated gates as PT-reversal of each other and we will see the necessity of this later

    (II) Apply the special symmetric form of composite buffer- atom-mediated gate involving both |0b⟩, |1b⟩ as the buffer states. More specially, this feels like the linear superpo- sition of two buffer-atom-mediated gates as PT-reversal of each other and we will see the necessity of this later. (III) Apply a second local π 2 -pulse on the buffer atom and mea...

    work page

  3. [3]

    The consequence of step II amounts to two simultaneous BAM CZ gates: one with |0b⟩, Ω0 and the other one with |1b⟩, Ω1. This re- sults in |0b⟩ −ieiη|1b⟩ C00|00⟩ + C01|01⟩ + C10|10⟩ − C11|11⟩ / √ 2 and eiη marks the relative overall phase difference between the two BAM CZ gates which can be corrected for locally on the buffer atom. So far in the idealized ...

    work page

  4. [4]

    The appropriate second local π 2 -pulse will generate the state of |0b⟩ (2 + δu0 + δu1)C00|00⟩ + (2 + δv0 + δv1)C01|01⟩ + (2 + δw0 + δw1)C10|10⟩ −(2 + δz0 + δz1)C11|11⟩ /2 +i|1b⟩ (δu0 − δu1)C00|00⟩ + (δv0 − δv1)C01|01⟩+(δw0 −δw1)C10|10⟩− (δz0 −δz1)C11|11⟩ /2. Currently, the experimental techniques of quantum control can suppress the two-qubit gate error t...

    work page

  5. [5]

    Kitaev, Annals of Physics 303, 2 (2003)

    A. Kitaev, Annals of Physics 303, 2 (2003)

    2003

  6. [6]

    Morvan, B

    A. Morvan, B. Villalonga, X. Mi, et al., Nature 634, 328 (2024)

    work page 2024

  7. [7]

    Google Quantum AI and Collaborators, Nature (2024)

    work page 2024

  8. [8]

    J. Wu, Y. Liu, B. Zhang, X. Jin, Y. Wang, H. Wang, and X. Yang, National Science Review 5, 715 (2018)

    work page 2018

Show all 33 references
  1. [9]

    Dennis, A

    E. Dennis, A. Kitaev, A. Landahl, and J. Preskill, Jour- nal of Mathematical Physics 43, 4452 (2002)

    2002

  2. [10]

    Z. Fu, P. Xu, Y. Sun, Y.-Y. Liu, X.-D. He, X. Li, M. Liu, R.-B. Li, J. Wang, L. Liu, and M.-S. Zhan, Phys. Rev. A 105, 042430 (2022)

    2022

  3. [11]

    A. G. Fowler, M. Mariantoni, J. M. Martinis, and A. N. Cleland, Phys. Rev. A 86, 032324 (2012)

    2012

  4. [12]

    You and M

    L. You and M. S. Chapman, Phys. Rev. A 62, 052302 (2000)

    2000

  5. [13]

    Saffman, T

    M. Saffman, T. G. Walker, and K. Mølmer, Rev. Mod. Phys. 82, 2313 (2010)

    2010

  6. [14]

    Saffman, National Science Review 6, 24 (2018)

    M. Saffman, National Science Review 6, 24 (2018)

    2018

  7. [15]

    Y. Sun, P. Xu, P.-X. Chen, and L. Liu, Phys. Rev. Ap- plied 13, 024059 (2020). 7

    2020

  8. [16]

    S. J. Evered, D. Bluvstein, M. Kalinowski, S. Ebadi, T. Manovitz, H. Zhou, S. H. Li, A. A. Geim, T. T. Wang, N. Maskara, H. Levine, G. Semeghini, M. Greiner, V. Vuleti´ c, and M. D. Lukin, Nature622, 268 (2023)

    2023

  9. [17]

    Urban, T

    E. Urban, T. A. Johnson, T. Henage, L. Isenhower, D. D. Yavuz, T. G. Walker, and M. Saffman, Nature Physics5, 110 (2009)

    2009

  10. [18]

    and explained the concepts of both off-resonant [15] and resonant drivings [19]. These preparations provide necessary groundwork to instantiate the abstract con- cepts in the above derivations and outlines the proce- dures to practically compute necessary waveforms. We keep th...

  11. [19]

    Isenhower, E

    L. Isenhower, E. Urban, X. L. Zhang, A. T. Gill, T. Henage, T. A. Johnson, T. G. Walker, and M. Saffman, Phys. Rev. Lett. 104, 010503 (2010)

    2010

  12. [20]

    Pohl and P

    T. Pohl and P. R. Berman, Phys. Rev. Lett. 102, 013004 (2009)

    2009

  13. [21]

    Y. Liu, Y. Sun, Z. Fu, P. Xu, X. Wang, X. He, J. Wang, and M. Zhan, Phys. Rev. Appl. 15, 054020 (2021)

    2021

  14. [22]

    Pause, L

    L. Pause, L. Sturm, M. Mittenb¨ uhler, S. Amann, T. Preuschoff, D. Sch¨ affner, M. Schlosser, and G. Birkl, Optica 11, 222 (2024)

    2024

  15. [23]

    Sun, SCIENCE CHINA Physics, Mechanics & Astron- omy 67, 120311 (2024)

    Y. Sun, SCIENCE CHINA Physics, Mechanics & Astron- omy 67, 120311 (2024)

    2024

  16. [24]

    X. Fan, X. Wang, and Y. Sun, Fundamental Research 10.1016/j.fmre.2024.07.002 (2024)

    2024 doi

  17. [25]

    X. Wang, T. Sheng, and Y. Sun, arXiv:2411.09882

    arXiv

  18. [26]

    Sun, Phys

    Y. Sun, Phys. Rev. Appl. 20, L061002 (2023)

    2023

  19. [27]

    Sun, Opt

    Y. Sun, Opt. Express 31, 3114 (2023)

    2023

  20. [28]

    L. H. Pedersen, N. M. Møller, and K. Mølmer, Physics Letters A 367, 47 (2007)

    2007

  21. [29]

    Petrosyan, F

    D. Petrosyan, F. Motzoi, M. Saffman, and K. Mølmer, Phys. Rev. A 96, 042306 (2017)

    2017

  22. [30]

    Schine, A

    N. Schine, A. W. Young, W. J. Eckner, M. J. Martin, and A. M. Kaufman, Nature Physics 18, 1067 (2022)

    2022

  23. [31]

    Bornet, G

    G. Bornet, G. Emperauger, C. Chen, B. Ye, M. Block, M. Bintz, J. A. Boyd, D. Barredo, T. Comparin, F. Mezzacapo, T. Roscilde, T. Lahaye, N. Y. Yao, and A. Browaeys, Nature 621, 728 (2023)

    2023

  24. [32]

    C. L. Degen, F. Reinhard, and P. Cappellaro, Rev. Mod. Phys. 89, 035002 (2017)

    2017

  25. [33]

    C. N. Yang, Inter. J. Mod. Phys. A 18, 3263 (2003)

    2003

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.