Benchmarking Error Mitigation: Artefactual Improvements in Zero-Noise Extrapolation

• URL: http://arxiv.org/abs/2607.09360v1
• Date: Fri, 10 Jul 2026 12:39:22 GMT
• Title: Benchmarking Error Mitigation: Artefactual Improvements in Zero-Noise Extrapolation
• Abstract summary: We expose a failure mode in Richardson Zero-Noise Extrapolation (ZNE)<n>We show that the extrapolation no longer reflects the underlying physics, but collapses into a fixed rescaling of a single noisy measurement.<n>This poses a rarely considered threat to the validity of many empirical evaluations in quantum computing.
• Score: 1.1683938179815823
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reliable benchmarking of Quantum Error Mitigation (QEM) requires distinguishing genuine improvements from artefacts of the post-processing arithmetic. In this paper, we expose a failure mode in Richardson Zero-Noise Extrapolation (ZNE), a widely used technique routinely (and often implicitly) relied upon in benchmarks and experiments. When noise amplification operates beyond usable signals - a regime that is quickly reached on current hardware for non-trivial circuits - we show that the extrapolation no longer reflects the underlying physics, but collapses into a fixed rescaling of a single noisy measurement, producing a bogus apparent improvement that is independent of noise amplification. This poses a rarely considered threat to the validity of many empirical evaluations in quantum computing. Measurements on real hardware (IQM Euro-Q-Exa) confirm this collapse with ordinary folding alone: as circuit depth erodes the signal, the reported estimate decouples from the truth and overshoots the ideal by up to 21%. We further introduce a matched-cost "garbage-folding" negative control that carries no usable signal yet reports a larger apparent improvement than genuine folding - showing that the magnitude of an improvement is not evidence of its correctness - alongside a zero-cost check flagging the artefact from data a benchmark already holds. We distil both into a short reporting checklist for ZNE benchmarks.

Related papers

• Robust Activation Map Rectification for Weakly Supervised Volumetric Segmentation: Temporal Coherence as a Free Lunch [31.156115062163412]
  Weakly supervised segmentation relies heavily on class activation maps (CAMs) to initially localize target regions.<n>Existing remedies typically introduce additional training stages or prototype learning, increasing computational cost and reducing robustness.<n>We propose a training-free prototype-free framework that rectifies unreliable CAMs by exploiting temporal and structural coherence in volumetric data as a free lunch.
  arXiv  Detail & Related papers  (2026-07-22T08:07:38Z)
• LiveFMBench: Unveiling the Power and Limits of Agentic Workflows in Specification Generation [75.05397479715576]
  Large language models (LLMs) and agents have shown promising progress, but their true capabilities and failure modes remain unclear.<n>We present the first systematic and contamination-aware study of LLM- and agent-based formal specification generation for C programs.
  arXiv  Detail & Related papers  (2026-05-02T11:31:33Z)
• From Early Encoding to Late Suppression: Interpreting LLMs on Character Counting Tasks [49.57538588967748]
  Large language models (LLMs) exhibit failures on elementary symbolic tasks such as character counting in a word, despite excelling on complex benchmarks.<n>We uncover a consistent phenomenon across modern architectures, including LLaMA, Qwen, and Gemma.<n>We show that symbolic reasoning failures in LLMs are not due to missing representations or insufficient scale, but arise from structured interference within the model's computation graph.
  arXiv  Detail & Related papers  (2026-04-01T11:40:12Z)
• Reducing Quantum Error Mitigation Bias Using Verifiable Benchmark Circuits [0.0]
  We present a simple, malleable and low-overhead approach for improving generic biased quantum error mitigation methods.<n>We achieve up to 15% fidelity improvements over standard QEM on 100-qubit circuits with up to 2000 entangling gates.
  arXiv  Detail & Related papers  (2026-03-10T20:51:30Z)
• Orders of magnitude runtime reduction in quantum error mitigation [0.0]
  We introduce a mitigation framework that combines virtual noise scaling with a layered mitigation architecture.<n>The proposed approach is compatible with dynamic circuits and can be seamlessly integrated with error detection and quantum error correction schemes.
  arXiv  Detail & Related papers  (2026-01-30T10:07:31Z)
• CARE What Fails: Contrastive Anchored-REflection for Verifiable Multimodal [84.71254539482369]
  Group-relative reinforcement learning with verifiable rewards (RLVR) often wastes the most informative data it already has the failures.<n>We present CARE, a failure-centric post-training framework for multimodal reasoning that turns errors into supervision.<n> CARE improves accuracy and training smoothness while explicitly increasing the share of learning signal that comes from failures.
  arXiv  Detail & Related papers  (2025-12-22T16:34:21Z)
• Reinforcement Learning with Verifiable yet Noisy Rewards under Imperfect Verifiers [90.50039419576807]
  Reinforcement Learning with Verifiable Rewards (RLVR) trains policies against automated verifiers to avoid costly human labeling.<n>To reduce vulnerability to verifier hacking, many RLVR systems collapse rewards to binary $0,1$ during training.<n>This choice carries a cost: it introduces textitfalse negatives (rejecting correct answers, FNs) and textitfalse positives (accepting incorrect ones, FPs)
  arXiv  Detail & Related papers  (2025-10-01T13:56:44Z)
• Reasoning or Memorization? Unreliable Results of Reinforcement Learning Due to Data Contamination [67.67725938962798]
  Pre-training on massive web-scale corpora leaves Qwen2.5 susceptible to data contamination in widely used benchmarks.<n>We introduce a generator that creates fully clean arithmetic problems of arbitrary length and difficulty, dubbed RandomCalculation.<n>We show that only accurate reward signals yield steady improvements that surpass the base model's performance boundary.
  arXiv  Detail & Related papers  (2025-07-14T17:55:15Z)
• Noise-Robust Estimation of Quantum Observables in Noisy Hardware [0.0]
  Noise-Robust Estimation is a noise-agnostic framework that systematically reduces estimation bias.<n>NRE exploits a bias-dispersion correlation uncovered in this work.<n>We experimentally validate NRE on an IQM superconducting quantum processor.
  arXiv  Detail & Related papers  (2025-03-09T17:18:16Z)
• Error Mitigation of BQP Computations using Measurement-Based Verification [0.7673339435080445]
  Utilising existing tools from quantum verification and measurement-based quantum computation, our protocol interleaves standard computation rounds alongside test rounds for noise sampling.<n>We introduce a post-selection technique called textitbasketing to address time-dependent noise and reduce overhead.
  arXiv  Detail & Related papers  (2023-06-07T11:35:21Z)
• Operational Markovianization in Randomized Benchmarking [7.680510419135912]
  We analytically show that error suppression techniques such as Dynamical Decoupling (DD) and Randomized Compiling (RC) can operationally Markovianize Randomized Benchmarking (RB) Our results show that simple and efficient error suppression methods can simultaneously tame non-Markovian noise and allow for standard and reliable gate quality estimation.
  arXiv  Detail & Related papers  (2023-05-08T13:37:18Z)
• Efficient and robust certification of genuine multipartite entanglement in noisy quantum error correction circuits [58.720142291102135]
  We introduce a conditional witnessing technique to certify genuine multipartite entanglement (GME) We prove that the detection of entanglement in a linear number of bipartitions by a number of measurements scales linearly, suffices to certify GME. We apply our method to the noisy readout of stabilizer operators of the distance-three topological color code and its flag-based fault-tolerant version.
  arXiv  Detail & Related papers  (2020-10-06T18:00:07Z)
• AQD: Towards Accurate Fully-Quantized Object Detection [94.06347866374927]
  We propose an Accurate Quantized object Detection solution, termed AQD, to get rid of floating-point computation. Our AQD achieves comparable or even better performance compared with the full-precision counterpart under extremely low-bit schemes.
  arXiv  Detail & Related papers  (2020-07-14T09:07:29Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.