arxiv: 2604.11963 · v1 · submitted 2026-04-13 · 🪐 quant-ph

Recognition: unknown

The Rotation Gap Is Not An Error: Ternary Structure in IBM Quantum Hardware

Selina Stenberg

Authors on Pith no claims yet

Pith reviewed 2026-05-10 15:37 UTC · model grok-4.3

classification 🪐 quant-ph

keywords quantum error correctionsyndrome statisticsFano factorternary transitionsIBM quantum hardwareregime classifiersurface codelogical error rate

0 comments

The pith

IBM quantum hardware produces structured ternary transitions that standard QEC wrongly corrects, increasing logical errors.

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

Quantum error correction has assumed every syndrome activation signals a random error needing correction. Data from 756 runs on three IBM Eagle processors instead show sub-Poissonian syndrome counts with Fano factor 0.856, independent of code distance, pointing to cooperative transitions rather than noise. The authors introduce a regime classifier decoder that flags these ternary states and abstains from correction. On hardware-calibrated models the classifier lowers logical error rates by 7-19 percent at unit detection depth through selective abstention, while a Google Willow control run shows the opposite super-Poissonian behavior. The work indicates that maximum correction can actively erase valid quantum information when error structure is platform-specific.

Core claim

The hardware exhibits sub-Poissonian syndrome statistics (F = 0.856) indicating that a fraction of syndrome events are structured cooperative transitions rather than random noise. A regime classifier decoder distinguishes binary errors requiring correction from ternary transitions that should be left uncorrected, achieving 7-19 percent lower logical error rates by correct abstention on 75-98 percent of the latter.

What carries the argument

The regime classifier decoder, which identifies ternary transitions from syndrome patterns and selectively abstains from correction on them.

If this is right

On mixed binary/ternary error models the classifier reduces logical error rates by 7-19 percent at static detection depth across all cell sizes.
The classifier correctly identifies 75-98 percent of ternary transitions and leaves them uncorrected, while standard decoders miscorrect them.
Sub-Poissonian statistics show zero dependence on code distance, consistent with cooperative transitions rather than independent noise.
A cross-platform control on Google's Willow processor yields super-Poissonian statistics and positive spatial correlation, confirming the effect is tied to IBM circuit asymmetry.

Where Pith is reading between the lines

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

QEC decoders may need to be calibrated to each hardware platform's specific error correlations rather than using universal correction rules.
Circuit designs that minimize P-gate asymmetry could reduce the occurrence of ternary transitions and simplify decoding.
Similar statistical tests on other quantum processors might reveal additional non-binary error regimes that benefit from abstention-based decoding.

Load-bearing premise

The sub-Poissonian syndrome statistics arise from ternary transitions inherent to the hardware rather than from measurement effects or other correlations.

What would settle it

Measuring the Fano factor on IBM hardware using surface-code circuits that lack the P-gate asymmetry, or repeating the runs on an IBM device without that gate asymmetry, to test whether the Fano factor remains below one.

Figures

Figures reproduced from arXiv: 2604.11963 by Selina Stenberg.

**Figure 2.** Figure 2: FIG. 2. Syndrome burst scaling on IBM Eagle r3. Mean burst count versus code distance with [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗

**Figure 3.** Figure 3: FIG. 3. Temporal decomposition of Google Willow syndrome statistics. (a) Fano factor versus [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗

**Figure 4.** Figure 4: FIG. 4. Cross-platform comparison of syndrome statistics: IBM Eagle r3 (756 QEC runs, heavy [PITH_FULL_IMAGE:figures/full_fig_p012_4.png] view at source ↗

**Figure 5.** Figure 5: FIG. 5. The miscorrection mechanism. Standard decoders (top, red) treat all syndrome activations [PITH_FULL_IMAGE:figures/full_fig_p016_5.png] view at source ↗

**Figure 6.** Figure 6: FIG. 6. Regime classifier decoder performance. (a) Logical error rate improvement across eight [PITH_FULL_IMAGE:figures/full_fig_p019_6.png] view at source ↗

read the original abstract

Quantum error correction assumes that all syndrome activations represent errors requiring correction. We present evidence from 756 QEC runs across three IBM Eagle r3 processors that this assumption is wrong. The hardware exhibits sub-Poissonian syndrome statistics (Fano factor F = 0.856, t = -131 against Poisson, zero dependence on code distance), indicating that a fraction of syndrome events are not random noise but structured cooperative transitions. We introduce a regime classifier decoder that distinguishes binary errors (which should be corrected) from ternary transitions (which should not). On a mixed binary/ternary error model calibrated to IBM hardware statistics, the classifier reduces logical error rates by 7-19% at static detection depth (tau = 1) across all cell sizes, with statistical significance p < 0.05 in 7 of 8 test conditions (p < 0.0001 in all four tau = 1 conditions). The improvement mechanism is selective abstention: the classifier correctly identifies 75-98% of ternary transitions and leaves them uncorrected (75-81% at tau = 1, 88-98% at tau = 5), whereas a standard decoder miscorrects them, introducing errors that would not otherwise exist. A cross-platform control on Google's 105-qubit Willow processor (420 experiments, d = 3, 5, 7) shows the opposite: super-Poissonian statistics (F = 2.42), super-linear burst scaling, and positive spatial correlation -- confirming that the sub-Poissonian signal is absent from standard surface-code circuits that lack the P-gate asymmetry. The result demonstrates that standard QEC actively destroys quantum information by correcting valid ternary states, and that less correction produces better performance when the hardware has cooperative error structure.

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.

Desk Editor's Note private letter to a colleague

The paper shows IBM hardware with sub-Poissonian syndrome counts and builds a decoder that abstains on some events for 7-19% gains, but the link to ternary transitions is not yet nailed down.

read the letter

Hi, the main point is that this work reports clear sub-Poissonian syndrome statistics from 756 runs on three IBM Eagle r3 processors, with a Fano factor of 0.856 and a very strong t-statistic against Poisson. They argue this means some syndrome events are structured ternary transitions rather than random errors, and they introduce a regime classifier decoder that leaves those alone. On a model fitted to the IBM data the abstention approach cuts logical error rates by 7-19% at fixed detection depth, with the Google Willow control showing the opposite super-Poissonian pattern and confirming the effect is IBM-specific to their circuit asymmetry. That cross-platform contrast and the raw scale of the runs are the strongest parts of the paper. The statistics themselves look reproducible from the numbers given, and the idea of selective abstention is a practical angle worth exploring for hardware-aware decoding. The soft spot is the causal step from sub-Poissonian variance to distinct ternary states that must not be corrected. Sub-Poissonian counts can also come from negative correlations among ordinary binary errors, leakage, or readout effects without any rotation gap. The classifier is trained and scored only on a mixed binary/ternary model calibrated to the same observed statistics, so the reported gains are tied to that fitting. No independent check such as pulse tomography or randomized benchmarking on the abstained subset is described to confirm those events are actually valid ternary states. This paper is for people working on decoder design for real superconducting hardware, especially IBM users who want to test whether less correction can help on their devices. A reader focused on hardware-specific error models will get concrete numbers and a new decoder strategy to try. It has enough experimental content and a testable claim to deserve peer review, even though the interpretation needs more direct evidence before the gains can be taken as settled.

Referee Report

2 major / 2 minor

Summary. The manuscript claims that quantum error correction incorrectly assumes all syndrome activations are errors requiring correction. Evidence from 756 QEC runs on three IBM Eagle r3 processors shows sub-Poissonian syndrome statistics (Fano factor F=0.856, t=-131 vs. Poisson, independent of code distance), interpreted as indicating structured ternary transitions. A regime classifier decoder is proposed to distinguish binary errors (correct) from ternary transitions (abstain), yielding 7-19% logical error rate reduction on a calibrated mixed error model at tau=1 (p<0.05 in most conditions). A Google Willow control (420 runs) exhibits super-Poissonian statistics (F=2.42), supporting hardware specificity due to P-gate asymmetry.

Significance. If the central interpretation holds, the work has notable significance for QEC by demonstrating that standard decoders can introduce errors through over-correction of valid states in hardware with cooperative transitions, potentially enabling abstention-based improvements. The scale of experiments, statistical rigor (Fano factor, t-test, distance independence), and cross-platform contrast are strengths. However, the result's impact depends on confirming the ternary mechanism over alternatives, which would affect decoder design for IBM-like systems.

major comments (2)

[Abstract] Abstract: The headline interpretation that F=0.856 directly evidences 'structured cooperative transitions' that are ternary and should be left uncorrected is not uniquely supported by the data. Sub-Poissonian statistics are compatible with alternative mechanisms (negative temporal correlations in binary errors, measurement leakage, or finite-shot bias) that do not require a distinct ternary regime or abstention strategy; no distinguishing test (e.g., pulse tomography on the identified subset) is described.
[Decoder evaluation] Decoder evaluation (results on mixed model): The 7-19% improvement at tau=1 is assessed exclusively on a mixed binary/ternary error model whose parameters are calibrated to the same IBM syndrome statistics used to detect the sub-Poissonian signal. This creates circularity in the performance claim, as the classifier is trained and evaluated on synthetic data derived from the observations it seeks to explain; the Google control uses dissimilar circuits and does not isolate the claimed P-gate cause.

minor comments (2)

[Abstract] The abstract reports classifier accuracy (75-98% identification of ternary events) but provides no details on the classifier architecture, input features, or training procedure, hindering assessment of its generality.
[Methods] No mention of raw data availability, exact syndrome extraction circuit parameters, or full experimental protocol (e.g., how the 756 runs were distributed across processors and distances), which would be needed for independent reproduction.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive and detailed comments. We address each major point below, providing clarifications on the interpretation of our results and indicating revisions to improve the manuscript.

read point-by-point responses

Referee: The headline interpretation that F=0.856 directly evidences 'structured cooperative transitions' that are ternary and should be left uncorrected is not uniquely supported by the data. Sub-Poissonian statistics are compatible with alternative mechanisms (negative temporal correlations in binary errors, measurement leakage, or finite-shot bias) that do not require a distinct ternary regime or abstention strategy; no distinguishing test (e.g., pulse tomography on the identified subset) is described.

Authors: We acknowledge that sub-Poissonian Fano factors are not uniquely diagnostic of ternary transitions and could arise from negative temporal correlations among binary errors, leakage, or finite-shot effects. However, the observed F=0.856 is independent of code distance across d=3,5,7, which is inconsistent with leakage or depth-dependent biases that would typically strengthen with larger circuits. The Google Willow control (standard surface-code circuits, F=2.42, super-linear burst scaling) provides a hardware-specific contrast supporting that the sub-Poissonian signal is tied to IBM's P-gate asymmetry rather than a generic statistical artifact. While we have not performed pulse tomography on the low-syndrome subset, the regime classifier's selective abstention yields consistent 7-19% logical-error reduction under the calibrated model, demonstrating practical value even if the microscopic mechanism requires further confirmation. We will revise the abstract and add a dedicated limitations paragraph discussing alternative mechanisms and the value of future tomography experiments. revision: partial
Referee: The 7-19% improvement at tau=1 is assessed exclusively on a mixed binary/ternary error model whose parameters are calibrated to the same IBM syndrome statistics used to detect the sub-Poissonian signal. This creates circularity in the performance claim, as the classifier is trained and evaluated on synthetic data derived from the observations it seeks to explain; the Google control uses dissimilar circuits and does not isolate the claimed P-gate cause.

Authors: The mixed model parameters are fitted directly to the empirical IBM syndrome counts to embed a ternary component whose statistics reproduce the observed sub-Poissonian behavior. The reported improvement is not circular because it quantifies the benefit of abstention versus mandatory correction under that hypothesis; if the sub-Poissonian signal were produced solely by binary correlations without a distinct regime, the classifier would not improve performance. The Google Willow experiments, although using standard surface-code circuits, serve as a negative control showing the opposite (super-Poissonian) statistics precisely when the P-gate asymmetry is absent, thereby isolating the effect to IBM hardware rather than circuit topology. We will expand the methods and results sections to explicitly describe the calibration procedure, include sensitivity checks on model parameters, and clarify the role of the cross-platform control. revision: partial

Circularity Check

1 steps flagged

Logical-error improvement demonstrated only on synthetic model calibrated to the same hardware statistics used for the Fano-factor claim

specific steps

fitted input called prediction [Abstract (performance paragraph)]
"On a mixed binary/ternary error model calibrated to IBM hardware statistics, the classifier reduces logical error rates by 7-19% at static detection depth (tau = 1) across all cell sizes, with statistical significance p < 0.05 in 7 of 8 test conditions (p < 0.0001 in all four tau = 1 conditions)."

The error model is explicitly calibrated to the IBM hardware statistics that include the observed sub-Poissonian Fano factor. The 7-19% reduction is therefore computed inside a synthetic ensemble whose statistics were fitted to match the input data; the 'improvement' is a within-calibration comparison rather than an out-of-sample prediction.

full rationale

The paper measures a genuine sub-Poissonian Fano factor (F=0.856) from 756 real IBM runs; that datum is external. However, the headline performance result (7-19% reduction) is obtained by training and evaluating the regime classifier exclusively on a mixed binary/ternary error model whose parameters were calibrated to reproduce those same IBM statistics. Consequently the reported gain is a within-model comparison rather than an independent prediction or hardware verification. The Google Willow control supplies an external contrast but does not test the classifier on real IBM data. This produces partial circularity of the 'fitted input called prediction' type without rendering the entire derivation self-referential.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 1 invented entities

The central claim depends on postulating ternary transitions to explain statistics and using a calibrated error model for decoder validation, adding burden beyond standard QEC assumptions.

free parameters (1)

mixed binary/ternary error model calibration parameters
The decoder performance is tested on a model fitted to match the observed IBM hardware statistics.

axioms (1)

domain assumption Syndrome statistics can be used to classify events into binary errors requiring correction and ternary transitions that should be left alone
This underpins the regime classifier decoder and the claim that standard QEC destroys information.

invented entities (1)

ternary transitions no independent evidence
purpose: To explain sub-Poissonian syndrome statistics as cooperative valid quantum state changes rather than errors
New entity postulated based on deviation from Poisson; no independent experimental confirmation beyond syndrome counts.

pith-pipeline@v0.9.0 · 5626 in / 1520 out tokens · 121900 ms · 2026-05-10T15:37:49.131790+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

31 extracted references · 14 canonical work pages · 1 internal anchor

[1]

Missing a binary error (same as any decoder)
[2]

Miscorrecting a ternary transition (unique to the paradigm) The regime classifier eliminates the second failure mode by identifying ternary transitions and leaving them alone. FIG. 5. The miscorrection mechanism. Standard decoders (top, red) treat all syndrome activations as errors and correct every flagged node. When 14.4% of activations are ternary tran...
[3]

Sub-Poissonian syndrome statistics (F= 0.856), measured across 756 runs on three processors, with zero dependence on code distance
[4]

A mixed error model in which 14.4% of syndrome events are ternary transitions rather than binary errors, calibrated to the hardware-measured Fano factor
[5]

A regime classifier decoder that reduces logical error rates by 7–19% through selective abstention—correctly identifying ternary transitions 75–98% of the time and leaving them uncorrected (75–81% atτ= 1, 88–98% atτ= 5). The result demonstrates a principle that inverts the standard logic of quantum error correction: when the hardware has cooperative struc...
[6]

The Identity The Merkabit architecture has 5 gates{S, R, T, P, F}and each cell on the Eisenstein lattice has 6 coordination directions and 7 nodes. The architectural prediction for the strong coupling constant at leading order is: αs = 5 6×7 = 5 42 ≈0.11905 (A1) This can be decomposed asα s = (5/6)/7 =F ideal/Ncell, where 5/6 is the ideal Fano factor (5 g...
[7]

Theibm brisbaneprocessor (lowest noise, best calibration) gives the closest match at 0.4%

Evidence from This Paper’s Data The Fano factors measured in Section II B at code distanced= 7 (matching the cell size) are: •ibm brisbane:F(d=7) = 0.8303,F/7 = 0.1186 (deviation from 5/42: 0.4%) •ibm kyoto:F(d=7) = 0.8584,F/7 = 0.1226 (deviation from 5/42: 3.0%) 24 •ibm osaka:F(d=7) = 0.8360,F/7 = 0.1194 (deviation from 5/42: 0.3%) The mean atd= 7 isF= 0...
[8]

The largerbrisbanedeviation is consistent with its higher noise floor noted in§A 2

Cross-check: Adjacent Correlation The relationshipF≈1−2×(adjacent correlation) holds to better than 1% across all three processors (ibm brisbane: predicted 0.840 vs actual 0.846, 0.7%;ibm kyoto: 0.870 vs 0.871, 0.1%;ibm osaka: 0.848 vs 0.849, 0.1%). The largerbrisbanedeviation is consistent with its higher noise floor noted in§A 2. This confirms that the ...
[10]

DAQEC-Benchmark: Drift-Aware Quantum Error Correction Dataset with IBM Hardware Validation,

A. Ashuraliyev, “DAQEC-Benchmark: Drift-Aware Quantum Error Correction Dataset with IBM Hardware Validation,”Zenodo, DOI: 10.5281/zenodo.17881116 (2025). Dataset: 756 QEC runs acrossibm brisbane,ibm kyoto,ibm osaka(127-qubit Eagle r3), 14 days contin- uous operation

work page doi:10.5281/zenodo.17881116 2025
[11]

Photon antibunching in resonance fluorescence,

H. J. Kimble, M. Dagenais, and L. Mandel, “Photon antibunching in resonance fluorescence,” Phys. Rev. Lett.39, 691 (1977)

1977
[12]

α= 4/3 in Driven Coherent Systems Near Cooperative Threshold,

S. Stenberg, “α= 4/3 in Driven Coherent Systems Near Cooperative Threshold,”Zenodo, 10.5281/zenodo.18980026 (2026)

work page doi:10.5281/zenodo.18980026 2026
[13]

A Single Geometric Constant Generates the Fine Structure Hierarchy,

S. Stenberg, “A Single Geometric Constant Generates the Fine Structure Hierarchy,”Zenodo, 10.5281/zenodo.18981288 (2026)

work page doi:10.5281/zenodo.18981288 2026
[14]

Resilient quantum computation: error models and thresholds,

E. Knill, R. Laflamme, and W. H. Zurek, “Resilient quantum computation: error models and thresholds,” Proc. R. Soc. Lond. A454, 365–384 (1998)

1998
[15]

Surface codes: Towards practical large-scale quantum computation,

A. G. Fowler, M. Mariantoni, J. M. Martinis, and A. N. Cleland, “Surface codes: Towards practical large-scale quantum computation,” Phys. Rev. A86, 032324 (2012)

2012
[16]

Suppressing quantum errors by scaling a surface code logical qubit,

Google Quantum AI, “Suppressing quantum errors by scaling a surface code logical qubit,” Nature614, 676–681 (2023)

2023
[17]

Evidence for the utility of quantum computing before fault tolerance,

Y. Kimet al., “Evidence for the utility of quantum computing before fault tolerance,” Nature 618, 500–505 (2023). IBM Eagle r3 (ibm kyiv), 127 fixed-frequency transmon qubits, heavy- hex connectivity, medianT 1 = 288µs,T 2 = 127µs

2023
[18]

Pengyu, Data for ”achieving optimal-distance atom-loss correction via pauli envelope”, 10.5281/zen- odo.19339056 (2026)

S. Stenberg, “Geometric Operator on the Eisenstein Lattice,”Zenodo, 10.5281/zen- odo.19075162 (2026). Construction of the Eisenstein lattice embedding for hexagonal quantum hardware connectivity

work page doi:10.5281/zen- 2026
[19]

The Rotation Gap Is Flat: Two-Scale Error Correction, a Structural Constant of Quantum Architecture, and the Migration Path to Fault-Tolerant Computation,

S. Stenberg, “The Rotation Gap Is Flat: Two-Scale Error Correction, a Structural Constant of Quantum Architecture, and the Migration Path to Fault-Tolerant Computation,”Zenodo, 10.5281/zenodo.19417293 (2026)

work page doi:10.5281/zenodo.19417293 2026
[20]

Ionization yield of radiations. II. The fluctuations of the number of ions,

U. Fano, “Ionization yield of radiations. II. The fluctuations of the number of ions,” Phys. 29 Rev.72, 26 (1947). Original derivation of the Fano factorF= Var(n)/⟨n⟩as a measure of sub-Poissonian statistics

1947
[21]

Theorie des elektrischen R¨ uckstandes in der Leidener Flasche,

R. Kohlrausch, “Theorie des elektrischen R¨ uckstandes in der Leidener Flasche,” Ann. Phys. 167, 179–214 (1854); G. Williams and D. C. Watts, Trans. Faraday Soc.66, 80–85 (1970). The KWW (Kohlrausch–Williams–Watts) stretched exponentialϕ(t) = exp(−(t/τ) α)

1970
[22]

The Yang–Mills Mass Gap as Spectral Resonance, Algebraic Connection Be- tween the Eisenstein Torus, the Coxeter Numberh(E 6) = 12, and ∆ = 1/24,

S. Stenberg, “The Yang–Mills Mass Gap as Spectral Resonance, Algebraic Connection Be- tween the Eisenstein Torus, the Coxeter Numberh(E 6) = 12, and ∆ = 1/24,”Zenodo, 10.5281/zenodo.19330363 (2026)

work page doi:10.5281/zenodo.19330363 2026
[23]

Stabilizer Codes and Quantum Error Correction

D. Gottesman, “Stabilizer codes and quantum error correction,” Ph.D. thesis, Caltech (1997). arXiv:quant-ph/9705052

work page internal anchor Pith review arXiv 1997
[24]

Scheme for reducing decoherence in quantum computer memory,

P. W. Shor, “Scheme for reducing decoherence in quantum computer memory,” Phys. Rev. A 52, R2493 (1995)

1995
[25]

The Standard Model Gauge Group from PSL(2,7) — SU(3)×SU(2)×U(1) as Representation Theory of the Three-Stratum Decomposition of GL(3,F 2),

S. Stenberg, “The Standard Model Gauge Group from PSL(2,7) — SU(3)×SU(2)×U(1) as Representation Theory of the Three-Stratum Decomposition of GL(3,F 2),”Zenodo, 10.5281/zenodo.19159718 (2026)

work page doi:10.5281/zenodo.19159718 2026
[26]

The Merkabit Architecture and the Klein Quartic: Cyclotomic Unification of the Fine Structure Constant, the Riemann Zeros, and the Most Symmetric Riemann Surface,

S. Stenberg, “The Merkabit Architecture and the Klein Quartic: Cyclotomic Unification of the Fine Structure Constant, the Riemann Zeros, and the Most Symmetric Riemann Surface,” Zenodo, 10.5281/zenodo.19066587 (2026)

work page doi:10.5281/zenodo.19066587 2026
[27]

Quantum error correction below the surface code threshold,

Google Quantum AI, “Quantum error correction below the surface code threshold,” Nature 638, 920–926 (2025). Data:Zenodo, DOI: 10.5281/zenodo.13273331

work page doi:10.5281/zenodo.13273331 2025
[28]

The Merkabit — A Ternary Computational Unit on the Eisenstein Lattice,

S. Stenberg, “The Merkabit — A Ternary Computational Unit on the Eisenstein Lattice,” Zenodo, 10.5281/zenodo.18925475 (v4, March 2026)

work page doi:10.5281/zenodo.18925475 2026
[29]

The abstractions leak: a day with IBM quantum hardware,

T. H. Hetland, “The abstractions leak: a day with IBM quantum hardware,”wiki.totto. org/blog/2026/04/06/(2026)

2026
[30]

The P Gate Is Native: Hardware Confirmation of the Dual- Spinor Merkabit on IBM Quantum,

S. Stenberg and T. H. Hetland, “The P Gate Is Native: Hardware Confirmation of the Dual- Spinor Merkabit on IBM Quantum,”Zenodo, 10.5281/zenodo.19484743 (2026)

work page doi:10.5281/zenodo.19484743 2026
[31]

Four of Five: Berry Phase, Quasi-Period, and the Fano Gap on IBM Eagle r3,

S. Stenberg and T. H. Hetland, “Four of Five: Berry Phase, Quasi-Period, and the Fano Gap on IBM Eagle r3,”Zenodo, 10.5281/zenodo.19502830 (2026)

work page doi:10.5281/zenodo.19502830 2026
[32]

The Merkabit Is Geometric: Cross-Architecture Hardware Validation, Corrected Willow Interpretation, and a Pre-Registered Prediction for Square-Grid Quantum Processors,

S. Stenberg and T. H. Hetland, “The Merkabit Is Geometric: Cross-Architecture Hardware Validation, Corrected Willow Interpretation, and a Pre-Registered Prediction for Square-Grid Quantum Processors,”Zenodo, 10.5281/zenodo.19554030 (2026). 30

work page doi:10.5281/zenodo.19554030 2026