How Measuring a Quantum Bit Hides What It Really Is

In the paper's words

Engineered qubits are certified through a closed calibration chain in which drive pulses, readout discriminators, and tomographic validation mutually calibrate one another. This self‑referential loop raises the question of whether the physical referent of the qubit label (collective circuit mode, electronic transition, or spin) is determined by the loop itself or remains underdetermined. We formalize the calibration hierarchy as an iterated map, derive its contraction condition, and compute the number of iterations required to reach the readout noise floor.

The abstract, verbatim; every line below is quoted from the text.

What it claims

that the standard calibration loop cannot distinguish a two‑level abstraction from a three‑level carrier—would be falsified if any of the following observations were made:

Numbers it reports

1. By assuming a typical trapped‑ion readout fidelity of $99.991\%$ and a leakage probability of $|c|^{2}=0.01$, we obtain a per‑shot error of $9\times10^{-5}$, a minimum of $11$ calibration iterations, and a shot budget of $N\ge2500$ (one‑sigma) required to resolve the ontologies once a leakage‑resolving observable is available.

   from Abstract

2. We model the calibration hierarchy as an iterated map and derive its contraction condition and convergence rate in closed form, computing the iteration count needed to reach a stated readout noise floor (Sections 3.1 and 4.3).

   from 1. Introduction

3. We anchor the analysis quantitatively by assuming a trapped‑ion readout fidelity of $99.991\%$, deriving the per‑shot error scale, the calibration iteration count, and the shot budget that would break the degeneracy once a leakage‑resolving observable exists (Section 4).

   from 1. Introduction

4. Average readout fidelity $f_{\mathrm{ro}}=99.991\% = 0.99991$ for a trapped‑ion qubit, over $N_{\mathrm{trials}}=10^{6}$ trials.

   from 4. Analysis

5. Adaptive‑measurement fidelity $f_{\mathrm{ad}}=99.99\% = 0.99999$.

   from 4. Analysis

6. Loop gain–sensitivity product $g\,s=0.5$, giving contraction factor $|1-g\,s|=0.5$; initial calibration error $e_{0}=0.1$ (drive‑parameter units).

   from 4. Analysis

How it is done

1. Using a two‑level abstraction and a three‑level anharmonic ladder with leakage amplitude $|c|$, we prove that, under any measurement chain confined to the computational subspace, the two ontologies produce identical outcome distributions (total‑variation distance $0$).

   from Abstract

2. Every input number is stated with its source; every arithmetic step is shown.

   from 4. Analysis

3. To resolve $\Delta p=|c|^{2}=0.01$ at one‑sigma significance, use the conservative binomial standard error bound $\sqrt{1/(4N)}\le0.01$: $$ N\ge\frac{1}{4\times(0.01)^{2}}=\frac{1}{4\times10^{-4}}=2500\text{ shots}. $$ For a three‑sigma requirement: $$ 3\sqrt{\frac{1}{4N}}\le0.01 \;\Longrightarrow\; N\ge\frac{9}{4\times10^{-4}}=22500\text{ shots}. $$ Noise‑floor margin: $$ \frac{|c|^{2}}{\varepsilon_{\mathrm{ro}}} =\frac{0.01}{9\times10^{-5}}\approx111. $$ Thus a leakage‑resolving extension of a high‑fidelity readout would see the three‑level ontology clearly, provided the extension exists.

   from 4.5 Shot budget and noise‑floor margin (projection)

4. Operating point $\Omega/\Delta=0.1$ gives the adiabatic bound $$ \left(\frac{\Omega}{2\Delta}\right)^{2} =\left(\frac{0.1}{2}\right)^{2} =(0.05)^{2}=2.5\times10^{-3}. $$ Hence $|c|^{2}\lesssim2.5\times10^{-3}$, a factor four below the assumed $0.01$.

   from 4.6 Weak‑drive consistency check (projection)

What would falsify it

1. The argument that differing platforms impose a consistency constraint is qualitative; a quantitative model would be needed to turn it into a falsifiable prediction.

   from Limitations

2. The central claim—that the standard calibration loop cannot distinguish a two‑level abstraction from a three‑level carrier—would be falsified if any of the following observations were made: A measurement chain confined to the computational subspace (i.e., no explicit $|2\rangle$ resolution) yields outcome statistics that differ from the predictions of Model A by a total‑variation distance exceeding the statistical noise floor set by $\varepsilon_{\mathrm{ro}}$.

   from Failure modes and falsification

3. By assuming a typical trapped‑ion readout fidelity of $99.991\%$ and a leakage probability of $|c|^{2}=0.01$, we obtain a per‑shot error of $9\times10^{-5}$, a minimum of $11$ calibration iterations, and a shot budget of $N\ge2500$ (one‑sigma) required to resolve the ontologies once a leakage‑resolving observable is available.

   from Abstract

Where the claim stops

1. The per‑shot error $\varepsilon_{\mathrm{ro}}$ is derived from an assumed fidelity $f_{\mathrm{ro}}=99.991\%$ (Input 1).

   from Limitations

2. The shot‑budget estimates assume independent Bernoulli trials and use the conservative $p\approx1/2$ bound; correlated noise or systematic errors would alter required $N$.

   from Limitations

3. By assuming a typical trapped‑ion readout fidelity of $99.991\%$ and a leakage probability of $|c|^{2}=0.01$, we obtain a per‑shot error of $9\times10^{-5}$, a minimum of $11$ calibration iterations, and a shot budget of $N\ge2500$ (one‑sigma) required to resolve the ontologies once a leakage‑resolving observable is available.

   from Abstract

Where to read

1. Abstract188 words · 1 min
2. 1. Introduction441 words · 2 min
3. 2. Background and Related Work545 words · 2 min
4. 3. Methods521 words · 2 min
5. 4. Analysis385 words · 2 min
6. 5. Results148 words · 1 min
7. 6. Discussion329 words · 1 min
8. 7. Practical Implications and Engineering Pathways295 words · 1 min
9. 8. Conclusion140 words · 1 min
10. Version History41 words · 1 min
11. Provenance and AI Involvement225 words · 1 min