Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

An Independent Reproduction and Stress Test of Logical Virtual Error Cancellation: Partial Replication of "Measure Now, Mitigate Later" (arXiv:2610.12400)

Amit Brahmbhatt

Published Oct 18, 2026• 0 citations
Publisher

Abstract

This report presents an independent, simulation-only reproduction of the logical virtual error cancellation (VEC) protocol of Shtanko, Roy, Kumar and Minev, "Measure Now, Mitigate Later: Virtual Error Cancellation for Logical Quantum Circuits" (arXiv:2610.12400). VEC reweights stored syndrome records by signed decoder-disagreement sector weights so that residual logical errors cancel in expectation. All results use an exact batched coset-probability evaluator for the code-capacity rotated surface code, validated against an independent library to ~10−15. No quantum hardware was used. Main results Paper element Our result Verdict Fig. 2, d = 3, 33 tie-invariant points maximum discrepancy 0.12% from digitized curves agrees (diagnostic) Fig. 2, d = 7, p = 0.05, 8 prespecified targets 7 of 8 match; wX = −0.4076 vs −0.42 partial Fig. 2e, d = 3 four-sector overhead 5.10–8.39 across tie rules; paper 5.76 convention-dependent Fig. 2e and Fig. 3c, p ≤ 0.01 rare sector forecast empty in 107 records (probability 0.94–0.999) forecast refusal; author convention required Fig. 3a noise learning, p = 0.005 median rate error 0.356% vs reported 0.33% consistent Fig. 3c residual bias, p = 0.022 budget scaling matches (exponent −0.47 vs −0.50); magnitude about 10× lower (3–5× lower under a post hoc population variant) outside band (conditional); remaining gap unexplained Extension: partition design distance-5 improvement failed distance-7 validation (median overhead ratio 1.13) closed, negative Findings beyond reproduction Support bound: the decoder-disagreement share obeys πdis ≤ 2pL, which explains why literal empirical calibration of the rare sectors is very likely undefined at the paper's lowest error rates. Unstated conventions: maximum-likelihood tie-breaking, empty-sector handling, and how the residual in Eq. S.270 is formed each change or may explain reported numbers. Section 2 of the report lists them with the specific questions for the authors. Sensitivity and trade-offs: the residual is highly sensitive to a coherent change in the rate inverse (14.5×). Calibrating on observed syndromes gives a 2–3× smaller learning-induced residual than population calibration (post hoc), but needs every sector populated. Cost of universality: for a single known observable, simpler corrections beat VEC; its advantage is correcting all observables at once. Method and transparency All confirmatory studies used protocols frozen before their outcomes were computed. Refusals, misses, a post hoc test that failed its frozen prediction, and errors corrected during the work are all reported. Files vec-report-v1.0.pdf — the report (31 pages). vec-replication-archive-v1.zip — credential-free source snapshots, frozen protocols, sufficient statistics and replay scripts. Large posterior tables are listed by SHA-256. SHA256SUMS.txt — checksum manifest for the archive contents. report-facts.json — full study-archive hashes, freeze records and environment details. The text is licensed CC-BY-4.0; the code in the archive is MIT-licensed.

DOI
10.5281/zenodo.23445193

Related papers

Measure Now, Mitigate Later: Virtual Error Cancellation for Logical Quantum Circuits

Oles Shtanko, Saswata Roy, Shashwat Kumar +1 more

Early fault-tolerant quantum algorithms are a key step in scalable quantum computing, but are ultimately constrained by residual, uncorrectable logical errors. Logical quantum error mitigation can eliminate this bias, however its standard requirements include prior noise characterization, more complex experiments, and substantial sampling overhead. Although no mitigation technique can simultaneously overcome all three constraints, we demonstrate that the first two---noise learning and need for modified circuits---can be completely bypassed for logical circuits by leveraging only syndrome records. Specifically, we introduce virtual error cancellation: a method that, unlike previous bias-free approaches, operates entirely in classical post-processing on universal quantum circuits. We develop and numerically test end-to-end protocols that reduce the sampling overhead beyond fundamental limitations of existing syndrome-aware mitigation techniques, including syndrome-based postselection. In particular, we propose a scheme that pairs the low-latency decoder used in the experiment with a high-complexity decoder run in post-processing. In numerical simulations of a distance-7 rotated surface code, we show that our full-stack dual-decoder algorithm exhibits an error that diminishes as the number of samples increases, capable of reaching more than three orders of magnitude of suppression compared to the decoded logical error rate in an experimentally achievable setting. These results establish syndrome records as a resource for correcting universal computational estimates after quantum execution.

quant-pharXiv:2610.12400•Oct 9, 2026•1 citation

Cancellation of Correlated Readout Errors in Measurement-Based Quantum Wires: Exact Channels, Finite-Memory Crossover, and Length-Reversal Boundaries

Jun Tsuzurugi

This record contains the revision 9 preprint and its verification and figure-reproduction materials. We determine when correlated readout-label errors cancel in a fixed measurement-based quantum wire and how finite correlation and independent noise determine its optimal length. Quantum operations and X measurements are ideal. A stationary binary Markov component with marginal 0 < p < 1/2 and persistence 0 ≤ c ≤ 1, combined with independent flips of probability 0 ≤ r < 1/2, yields an exact output Pauli channel through two-dimensional transfer matrices. The channel acts on arbitrary inputs, including reference-entangled inputs, and its entanglement infidelity equals its half-diamond distance from the identity. Common-mode errors cancel at lengths divisible by four when c = 1 and r = 0. Every fixed c < 1 instead gives a completely depolarizing long-wire limit; two residue-class crossover functions resolve the singular limit. Fidelity decreases within each even residue class, so its maximum over all positive even lengths is attained at two or four steps. Exact boundaries select the maximizer, and an effective depolarizing penalty quantifies the overhead tolerance of the four-step advantage. Contents. The ZIP includes the latest 19-page manuscript PDF, its authoritative LaTeX source, the three manuscript figures, the unchanged ESM_1.zip containing Python verification and figure-generation code, numerical data, saved execution records and provenance, plus licensing information, a BibTeX citation, and SHA-256 checksums. Scientific files are unchanged from the supplied revision 9 materials. The packaging did not rerun calculations or recompile the manuscript. The earlier reference-only PDF and submission-preparation working records are excluded. Licenses. MIT applies to Python source code, including code inside ESM_1.zip. CC BY 4.0 applies to the manuscript, LaTeX source, figures, numerical data, documentation, and other non-code research materials. These licenses apply to the respective material categories; see the README and license files for scope. Funding. No funding or grants were received for this study. Competing interests. The author has no relevant financial or non-financial interests to disclose. AI-assisted preparation. ChatGPT (OpenAI) was used to assist with mathematical analysis, code development, figure preparation, English-language editing, LaTeX submission preparation, and organization of supplementary files. The author carefully reviewed the mathematical reasoning, code, figures, and manuscript text, and takes full responsibility for the scientific content and the final manuscript.

•Nov 10, 2026•0 citations

Zeno-Enhanced Probabilistic Error Cancellation with Quantum Error Detection Codes

Yi Yuan, Yuanchen Zhao, Dong E. Liu

Probabilistic error cancellation (PEC) is unbiased but suffers exponential sampling overhead set by noise-weighted circuit volume, whereas quantum error-detecting codes (QEDCs) remove many physical faults by stabilizer post-selection but leave an undetectable logical residue. We exploit this complementarity by using post-selection to map physical noise to a weaker accepted logical channel, and then applying PEC only to the residual channel. The resulting feedback-free QED+PEC scheme interleaves Clifford logical blocks, stabilizer measurements, post-selection, and probabilistic cancellation on accepted trajectories, without real-time decoding or active recovery. A key complication is that post-selection correlates accepted fault branches through stabilizer-commutation constraints, so the sparse Pauli-Lindblad factorization underlying bare PEC no longer applies directly. We therefore construct the inverse channel perturbatively: for fixed order $K$, only accepted fault branches up to order $K$ are retained, reducing preprocessing from $2^m$ branches to $O(m^K)$ per block. The order-$K$ protocol cancels the normalized post-selected channel through degree $K$, leaving a per-block error $O(W^{K+1})$ that accumulates at most linearly. For logical GHZ-state preparation with the $[[n,n-2,2]]$ Iceberg code under circuit-level depolarizing noise and ideal stabilizer measurements, first-order QED+PEC reaches $n=200$ physical qubits and lowers sampling overhead by three to four orders of magnitude relative to standard PEC while maintaining $F\simeq0.956$. Syndrome-noise tests show that readout-only flips mainly increase post-selection cost, whereas noisy GHZ-assisted global stabilizer extraction can remove the advantage. This identifies a discrete-Zeno trade-off: cheap detection reshapes the effective channel PEC must invert, rather than simply adding overhead.

quant-pharXiv:2605.12149•Quantum Science and Technology•May 13, 2026•5 citations

HRR - Selective Logical-Sector Rescue for QLDPC Decoding

Kyuchul Lee, cording.ai

HRR-HYBRID Under Stronger Validation: A Reproducible Correction and Negative Result for Selective Logical-Sector Rescue Abstract Version 1 reported that HRR-HYBRID improved several quantum low-density parity-check decoding baselines on asynthetic fixture. A subsequent audit identified design weaknesses: the check matrix was not a demonstratedquantum code family, errors were planted rather than sampled from the declared channel, train and test recordscould share exact error patterns, the random control had a smaller candidate budget, paired significance was notreported, and hybrid latency was not measured. We redesigned the experiment around three verifiedhypergraph-product CSS codes, separated fitting, validation, and testing by code size, used independent Bernoulli Xerrors, matched candidate budgets, and added paired statistical tests and end-to-end reference timing. On the held-out [[50,2,5]] code, HRR improved 7, tied 6, and worsened 7 of 20 decoder-by-error-rate cells.BeliefFind at physical error rate 0.10 showed an individual reduction in logical error rate from 0.280273 to 0.268066(48 fixes, 23 breaks; exact McNemar p=0.004065), but this did not survive a conservative correction across 20examined cells. Matched random candidates were better than HRR in two nearby BeliefFind cells. The sequentialreference hybrid was approximately 77-189 times slower than the corresponding baseline. These data do not supportthe v1 characterization of HRR-HYBRID as a generally effective, decoder-agnostic, deployment-oriented rescuelayer. They support continued study of a narrower, decoder-specific high-noise hypothesis. We publish the negativeresult, executable code, fixture, raw logs, dependency locks, and validation scripts as a correction to the publicrecord.

•Jun 22, 2026•0 citations

Co-Designing Error Mitigation and Error Detection for Logical Qubits

Rohan S. Kumar, Takahiro Tsunoda, Sophia H. Xue +3 more

Near-term quantum workloads demand error management, yet the two lightest-weight techniques, Quantum Error Detection (QED) and Probabilistic Error Cancellation (PEC), have complementary cost profiles whose joint architectural design space remains unexplored. QED encodes logical qubits and discards error-flagged runs, filtering noise with low qubit overhead but leaving residual errors; PEC can correct these in software, but at exponential cost in noise strength. If QED efficiently reduces per-gate noise, PEC's cost savings can outweigh QED's discard overhead; realizing this, however, requires solving two system-level design challenges. First, the \textit{QED interval} -- how often detection cycles are inserted -- is a tunable architectural parameter governing the cost-accuracy tradeoff. We derive an efficiency condition and show that the canonical one-cycle-per-gate frequency does not achieve break-even in any code we evaluate, while optimized intervals on high-rate Iceberg codes do. Second, we discover that naive PEC+QED integration \textit{degrades} accuracy below the QED-only baseline. The root cause is a transient error profile in the first detection cycle that corrupts PEC's noise model. We develop \textit{steady-state extraction}, a co-designed characterization protocol that isolates steady-state error behavior, reducing estimation bias by up to $10.2\times$. On a $[[6,4,2]]$ Iceberg code running QAOA ($p{=}4$--$8$) with a fixed shot budget, PEC+QED achieves $2$--$11\times$ lower absolute error and up to $31\times$ lower MSE versus PEC on physical qubits, with per-interval savings compounding over interval depth.

quant-pharXiv:2604.19871•Apr 23, 2026•7 citations

Quantum Error Correction on Error-mitigated Physical Qubits

Minjun Jeon, Zhenyu Cai

We present a general framework for applying linear quantum error mitigation (QEM) techniques directly to physical qubits within a logical qubit to suppress logical errors. By exploiting the linearity of quantum error correction (QEC), we demonstrate that any linear QEM method$\unicode{x2014}$including probabilistic error cancellation (PEC), zero-noise extrapolation (ZNE), and symmetry verification$\unicode{x2014}$can be integrated into the physical layer without requiring modifications to the subsequent QEC decoder. Applying this framework to memory experiments using PEC, we analytically prove and numerically verify that the leading-order contribution to the logical error can be removed, increasing the effective code distance by 2. Our simulations on repetition and rotated surface codes show that a distance-3 code with physical-level PEC achieves logical error rates lower than or similar to a distance-5 unmitigated code while using 40% and 64% fewer qubits, respectively. These results establish physical-level QEM as a widely compatible and resource-efficient strategy for enhancing logical performance in early fault-tolerant architectures.

quant-pharXiv:2601.18384•Jan 27, 2026•17 citations
This report presents an independent, simulation-only reproduction of the logical virtual error cancellation (VEC) protocol of Shtanko, Roy, Kumar and Minev, "Measure Now, Mitigate Later: Virtual Error Cancellation for Logical Quantum Circuits" (arXiv:2610.12400). VEC reweights stored syndrome records by signed decoder-disagreement sector weights so that residual logical errors cancel in expectation. All results use an exact batched coset-probability evaluator for the code-capacity rotated surface code, validated against an independent library to ~10−15. No quantum hardware was used. Main results Paper element Our result Verdict Fig. 2, d = 3, 33 tie-invariant points maximum discrepancy 0.12% from digitized curves agrees (diagnostic) Fig. 2, d = 7, p = 0.05, 8 prespecified targets 7 of 8 match; wX = −0.4076 vs −0.42 partial Fig. 2e, d = 3 four-sector overhead 5.10–8.39 across tie rules; paper 5.76 convention-dependent Fig. 2e and Fig. 3c, p ≤ 0.01 rare sector forecast empty in 107 records (probability 0.94–0.999) forecast refusal; author convention required Fig. 3a noise learning, p = 0.005 median rate error 0.356% vs reported 0.33% consistent Fig. 3c residual bias, p = 0.022 budget scaling matches (exponent −0.47 vs −0.50); magnitude about 10× lower (3–5× lower under a post hoc population variant) outside band (conditional); remaining gap unexplained Extension: partition design distance-5 improvement failed distance-7 validation (median overhead ratio 1.13) closed, negative Findings beyond reproduction Support bound: the decoder-disagreement share obeys πdis ≤ 2pL, which explains why literal empirical calibration of the rare sectors is very likely undefined at the paper's lowest error rates. Unstated conventions: maximum-likelihood tie-breaking, empty-sector handling, and how the residual in Eq. S.270 is formed each change or may explain reported numbers. Section 2 of the report lists them with the specific questions for the authors. Sensitivity and trade-offs: the residual is highly sensitive to a coherent change in the rate inverse (14.5×). Calibrating on observed syndromes gives a 2–3× smaller learning-induced residual than population calibration (post hoc), but needs every sector populated. Cost of universality: for a single known observable, simpler corrections beat VEC; its advantage is correcting all observables at once. Method and transparency All confirmatory studies used protocols frozen before their outcomes were computed. Refusals, misses, a post hoc test that failed its frozen prediction, and errors corrected during the work are all reported. Files vec-report-v1.0.pdf — the report (31 pages). vec-replication-archive-v1.zip — credential-free source snapshots, frozen protocols, sufficient statistics and replay scripts. Large posterior tables are listed by SHA-256. SHA256SUMS.txt — checksum manifest for the archive contents. report-facts.json — full study-archive hashes, freeze records and environment details. The text is licensed CC-BY-4.0; the code in the archive is MIT-licensed.
Early fault-tolerant quantum algorithms are a key step in scalable quantum computing, but are ultimately constrained by residual, uncorrectable logical errors. Logical quantum error mitigation can eliminate this bias, however its standard requirements include prior noise characterization, more complex experiments, and substantial sampling overhead. Although no mitigation technique can simultaneously overcome all three constraints, we demonstrate that the first two---noise learning and need for modified circuits---can be completely bypassed for logical circuits by leveraging only syndrome records. Specifically, we introduce virtual error cancellation: a method that, unlike previous bias-free approaches, operates entirely in classical post-processing on universal quantum circuits. We develop and numerically test end-to-end protocols that reduce the sampling overhead beyond fundamental limitations of existing syndrome-aware mitigation techniques, including syndrome-based postselection. In particular, we propose a scheme that pairs the low-latency decoder used in the experiment with a high-complexity decoder run in post-processing. In numerical simulations of a distance-7 rotated surface code, we show that our full-stack dual-decoder algorithm exhibits an error that diminishes as the number of samples increases, capable of reaching more than three orders of magnitude of suppression compared to the decoded logical error rate in an experimentally achievable setting. These results establish syndrome records as a resource for correcting universal computational estimates after quantum execution.
This record contains the revision 9 preprint and its verification and figure-reproduction materials. We determine when correlated readout-label errors cancel in a fixed measurement-based quantum wire and how finite correlation and independent noise determine its optimal length. Quantum operations and X measurements are ideal. A stationary binary Markov component with marginal 0 < p < 1/2 and persistence 0 ≤ c ≤ 1, combined with independent flips of probability 0 ≤ r < 1/2, yields an exact output Pauli channel through two-dimensional transfer matrices. The channel acts on arbitrary inputs, including reference-entangled inputs, and its entanglement infidelity equals its half-diamond distance from the identity. Common-mode errors cancel at lengths divisible by four when c = 1 and r = 0. Every fixed c < 1 instead gives a completely depolarizing long-wire limit; two residue-class crossover functions resolve the singular limit. Fidelity decreases within each even residue class, so its maximum over all positive even lengths is attained at two or four steps. Exact boundaries select the maximizer, and an effective depolarizing penalty quantifies the overhead tolerance of the four-step advantage. Contents. The ZIP includes the latest 19-page manuscript PDF, its authoritative LaTeX source, the three manuscript figures, the unchanged ESM_1.zip containing Python verification and figure-generation code, numerical data, saved execution records and provenance, plus licensing information, a BibTeX citation, and SHA-256 checksums. Scientific files are unchanged from the supplied revision 9 materials. The packaging did not rerun calculations or recompile the manuscript. The earlier reference-only PDF and submission-preparation working records are excluded. Licenses. MIT applies to Python source code, including code inside ESM_1.zip. CC BY 4.0 applies to the manuscript, LaTeX source, figures, numerical data, documentation, and other non-code research materials. These licenses apply to the respective material categories; see the README and license files for scope. Funding. No funding or grants were received for this study. Competing interests. The author has no relevant financial or non-financial interests to disclose. AI-assisted preparation. ChatGPT (OpenAI) was used to assist with mathematical analysis, code development, figure preparation, English-language editing, LaTeX submission preparation, and organization of supplementary files. The author carefully reviewed the mathematical reasoning, code, figures, and manuscript text, and takes full responsibility for the scientific content and the final manuscript.
Probabilistic error cancellation (PEC) is unbiased but suffers exponential sampling overhead set by noise-weighted circuit volume, whereas quantum error-detecting codes (QEDCs) remove many physical faults by stabilizer post-selection but leave an undetectable logical residue. We exploit this complementarity by using post-selection to map physical noise to a weaker accepted logical channel, and then applying PEC only to the residual channel. The resulting feedback-free QED+PEC scheme interleaves Clifford logical blocks, stabilizer measurements, post-selection, and probabilistic cancellation on accepted trajectories, without real-time decoding or active recovery. A key complication is that post-selection correlates accepted fault branches through stabilizer-commutation constraints, so the sparse Pauli-Lindblad factorization underlying bare PEC no longer applies directly. We therefore construct the inverse channel perturbatively: for fixed order KKK, only accepted fault branches up to order KKK are retained, reducing preprocessing from 2m2^m2m branches to O(mK)O(m^K)O(mK) per block. The order-KKK protocol cancels the normalized post-selected channel through degree KKK, leaving a per-block error O(WK+1)O(W^{K+1})O(WK+1) that accumulates at most linearly. For logical GHZ-state preparation with the [[n,n−2,2]][[n,n-2,2]][[n,n−2,2]] Iceberg code under circuit-level depolarizing noise and ideal stabilizer measurements, first-order QED+PEC reaches n=200n=200n=200 physical qubits and lowers sampling overhead by three to four orders of magnitude relative to standard PEC while maintaining F≃0.956F\simeq0.956F≃0.956. Syndrome-noise tests show that readout-only flips mainly increase post-selection cost, whereas noisy GHZ-assisted global stabilizer extraction can remove the advantage. This identifies a discrete-Zeno trade-off: cheap detection reshapes the effective channel PEC must invert, rather than simply adding overhead.
HRR-HYBRID Under Stronger Validation: A Reproducible Correction and Negative Result for Selective Logical-Sector Rescue Abstract Version 1 reported that HRR-HYBRID improved several quantum low-density parity-check decoding baselines on asynthetic fixture. A subsequent audit identified design weaknesses: the check matrix was not a demonstratedquantum code family, errors were planted rather than sampled from the declared channel, train and test recordscould share exact error patterns, the random control had a smaller candidate budget, paired significance was notreported, and hybrid latency was not measured. We redesigned the experiment around three verifiedhypergraph-product CSS codes, separated fitting, validation, and testing by code size, used independent Bernoulli Xerrors, matched candidate budgets, and added paired statistical tests and end-to-end reference timing. On the held-out [[50,2,5]] code, HRR improved 7, tied 6, and worsened 7 of 20 decoder-by-error-rate cells.BeliefFind at physical error rate 0.10 showed an individual reduction in logical error rate from 0.280273 to 0.268066(48 fixes, 23 breaks; exact McNemar p=0.004065), but this did not survive a conservative correction across 20examined cells. Matched random candidates were better than HRR in two nearby BeliefFind cells. The sequentialreference hybrid was approximately 77-189 times slower than the corresponding baseline. These data do not supportthe v1 characterization of HRR-HYBRID as a generally effective, decoder-agnostic, deployment-oriented rescuelayer. They support continued study of a narrower, decoder-specific high-noise hypothesis. We publish the negativeresult, executable code, fixture, raw logs, dependency locks, and validation scripts as a correction to the publicrecord.
Near-term quantum workloads demand error management, yet the two lightest-weight techniques, Quantum Error Detection (QED) and Probabilistic Error Cancellation (PEC), have complementary cost profiles whose joint architectural design space remains unexplored. QED encodes logical qubits and discards error-flagged runs, filtering noise with low qubit overhead but leaving residual errors; PEC can correct these in software, but at exponential cost in noise strength. If QED efficiently reduces per-gate noise, PEC's cost savings can outweigh QED's discard overhead; realizing this, however, requires solving two system-level design challenges. First, the \textit{QED interval} -- how often detection cycles are inserted -- is a tunable architectural parameter governing the cost-accuracy tradeoff. We derive an efficiency condition and show that the canonical one-cycle-per-gate frequency does not achieve break-even in any code we evaluate, while optimized intervals on high-rate Iceberg codes do. Second, we discover that naive PEC+QED integration \textit{degrades} accuracy below the QED-only baseline. The root cause is a transient error profile in the first detection cycle that corrupts PEC's noise model. We develop \textit{steady-state extraction}, a co-designed characterization protocol that isolates steady-state error behavior, reducing estimation bias by up to 10.2×10.2\times10.2×. On a [[6,4,2]][[6,4,2]][[6,4,2]] Iceberg code running QAOA (p=4p{=}4p=4--888) with a fixed shot budget, PEC+QED achieves 222--11×11\times11× lower absolute error and up to 31×31\times31× lower MSE versus PEC on physical qubits, with per-interval savings compounding over interval depth.
We present a general framework for applying linear quantum error mitigation (QEM) techniques directly to physical qubits within a logical qubit to suppress logical errors. By exploiting the linearity of quantum error correction (QEC), we demonstrate that any linear QEM method$\unicode{x2014}$including probabilistic error cancellation (PEC), zero-noise extrapolation (ZNE), and symmetry verification$\unicode{x2014}$can be integrated into the physical layer without requiring modifications to the subsequent QEC decoder. Applying this framework to memory experiments using PEC, we analytically prove and numerically verify that the leading-order contribution to the logical error can be removed, increasing the effective code distance by 2. Our simulations on repetition and rotated surface codes show that a distance-3 code with physical-level PEC achieves logical error rates lower than or similar to a distance-5 unmitigated code while using 40% and 64% fewer qubits, respectively. These results establish physical-level QEM as a widely compatible and resource-efficient strategy for enhancing logical performance in early fault-tolerant architectures.