arXiv subjects

Erik Nielsen

Publications and source records attributed to Erik Nielsen.

At least 19 recordsLinked to original sources

arXiv2026

Breakeven demonstration of quantum low-density parity-check codes

High-rate quantum low-density parity-check (qLDPC) codes are a leading candidate for fault-tolerant quantum computing. They feature higher encoding rates than planar alternatives such as the surface code, but their implementation often entails significant hardware hurdles like the need for long-range couplers. We leverage the flexibility of a trapped-ion quantum computer to demonstrate nine quantum error-correcting codes with starkly different qubit connectivity requirements on a single device without any hardware reconfiguration. These experiments span three families of quantum error-correcting codes: qLDPC codes, topological codes, and concatenated codes. With a qLDPC code encoding 4 logical qubits into 18 physical qubits, we achieve a logical error rate up to $9\times$ better than a previous demonstration of a similar code on superconducting solid-state qubits. Moreover, our implementation exhibits breakeven performance, with some instances achieving qubit lifetimes comparable to or slightly exceeding that of our trapped-ion qubits. We use a novel implementation of the optical-metastable-ground (OMG) architecture for addressable mid-circuit measurement and reset, which enables us to perform these experiments without any ion transport or dedicated coolant ions, requirements that typically consume a large fraction of the runtime or ion count of trapped-ion quantum computers.

quant-ph↗

arXiv2026

From Layers to Submodules: Rethinking Granularity in Replacement-Based LLM Compression

Post-training compression of Large Language Models (LLMs) removes entire architectural components, either deleting them or replacing them with fitted modules. Existing replacement-based methods share two design constraints: full-layer granularity and contiguous selection. We argue that this is overly restrictive: in fact, redundancy in pretrained transformers is not confined to contiguous regions, nor does it evenly distribute between Attention and FeedForward outputs, implying that different strategies best approximate different submodule types and that removable components need not cluster within contiguous depth ranges. Based on this intuition, we introduce SubFit (Submodule-level Fitted residual replacement), which compresses LLMs at the submodule level: Attention and FeedForward submodules are selected non-contiguously, and each receives its own lightweight fitted residual bypass. SubFit operates post-training and requires only calibration data. Across ten LLMs (five base, five instruction-tuned), five sparsity levels from 12.5% to 37.5%, and four replacement-based baselines, SubFit achieves the best aggregate perplexity-accuracy trade-off across the evaluated sparsity levels, with larger gains under aggressive compression. At 25% sparsity, it retains 84.6% of dense downstream accuracy and incurs 2.42x perplexity degradation, against 81.6% and 4.34x for the strongest baselines, while delivering measurable inference speedup and KV-cache savings. Code is available at https://github.com/eliacunegatti/SubFit.

cs.CL↗

arXiv2026

A gauge-invariant theory of small Markovian errors in quantum gate sets

Noisy logic operations on a quantum computational register -- e.g., one or more qubits -- can be described by transfer matrices (a.k.a. CPTP maps or superoperators) that act linearly on the density matrix representing the register's quantum state. Collectively, these operations form a gate set. Gate sets have a gauge freedom; many gate sets that appear different actually predict the same experimental outcomes. A property of a gate set can be observable (and thus physically relevant) only if it is gauge-invariant. Unfortunately, no good gauge-invariant parameterizations of gate sets are known. We introduce the next best thing, a perturbative gauge-invariant parameterization of small Markovian errors in gate sets. We construct vector spaces of properties that are first-order gauge-invariant (FOGI). We show how to construct and understand FOGI properties, how to use them as coordinates to parameterize gate sets without gauge freedom, and how to extract approximately gauge-invariant error metrics.

quant-ph↗

arXiv2026

Hallucination as an Anomaly: Dynamic Intervention via Probabilistic Circuits

One of the most critical challenges in Large Language Models is their tendency to hallucinate, i.e., produce factually incorrect responses. Existing approaches show promising results in terms of hallucination correction, but still suffer from a main limitation: they apply corrections indiscriminately to every token, corrupting also the originally correct generations. To overcome this drawback, we propose PCNET, a Probabilistic Circuit trained as a tractable density estimator over the LLM residual stream. The method detects hallucinations as geometric anomalies on the factual manifold, which is done via exact Negative Log-Likelihood computation, hence without the need for sampling, external verifiers, or weight modifications, as in existing techniques. To demonstrate its effectiveness, we exploit PCNET as a dynamic gate that distinguishes hallucinated from factual hidden states at each decoding step. This triggers our second main contribution, PC-LDCD (Probabilistic Circuit Latent Density Contrastive Decoding), only when the latent geometry deviates from factual regions, while leaving correct generations untouched. Across four LLMs, ranging from 1B to 8B models, and four benchmarks covering conversational reasoning, knowledge-intensive QA, reading comprehension, and truthfulness, PCNET achieves near-perfect hallucination detection across CoQA, SQuAD v2.0, and TriviaQA, with AUROC reaching up to 99%. Moreover, PC-LDCD obtains the highest True+Info, MC2, and MC3 scores on TruthfulQA in three out of four models, in comparison with state-of-the-art baselines, while reducing the mean corruption rate to 53.7% and achieving a preservation rate of 79.3%. Our proposed method is publicly available on GitHub.

cs.CL↗

arXiv2026

Establishing trust in quantum computations

Quantum computing hardware has grown sufficiently complex that it often can no longer be simulated by classical computers, but its computational power remains limited by errors. These errors corrupt the results of quantum algorithms, and it is no longer always feasible to use classical simulations to directly check the correctness of quantum computations. Without practical methods for quantifying the accuracy with which a quantum algorithm has been executed, it is difficult to establish trust in the results of a quantum computation. Here we solve this problem, by introducing a simple and efficient technique for measuring the fidelity with which an as-built quantum computer can execute an algorithm. Our technique converts the algorithm's quantum circuits into a set of closely related ``mirror circuits'' whose success rates can be efficiently measured. It enables measuring the fidelity of quantum algorithm executions both in the near-term, with algorithms run on hundreds or thousands of physical qubits, and into the future, with algorithms run on logical qubits protected by quantum error correction.

quant-ph↗

arXiv2024

Benchmarking a trapped-ion quantum computer with 30 qubits

Quantum computers are rapidly becoming more capable, with dramatic increases in both qubit count and quality. Among different hardware approaches, trapped-ion quantum processors are a leading technology for quantum computing, with established high-fidelity operations and architectures with promising scaling. Here, we demonstrate and thoroughly benchmark the IonQ Forte system: configured as a single-chain 30-qubit trapped-ion quantum computer with all-to-all operations. We assess the performance of our quantum computer operation at the component level via direct randomized benchmarking (DRB) across all 30 choose 2 = 435 gate pairs. We then show the results of application-oriented benchmarks and show that the system passes the suite of algorithmic qubit (AQ) benchmarks up to #AQ 29. Finally, we use our component-level benchmarking to build a system-level model to predict the application benchmarking data through direct simulation. While we find that the system-level model correlates with the experiment in predicting application circuit performance, we note quantitative discrepancies indicating significant out-of-model errors, leading to higher predicted performance than what is observed. This highlights that as quantum computers move toward larger and higher-quality devices, characterization becomes more challenging, suggesting future work required to push performance further.

quant-ph↗

arXiv2023

Demonstrating scalable randomized benchmarking of universal gate sets

Randomized benchmarking (RB) protocols are the most widely used methods for assessing the performance of quantum gates. However, the existing RB methods either do not scale to many qubits or cannot benchmark a universal gate set. Here, we introduce and demonstrate a technique for scalable RB of many universal and continuously parameterized gate sets, using a class of circuits called randomized mirror circuits. Our technique can be applied to a gate set containing an entangling Clifford gate and the set of arbitrary single-qubit gates, as well as gate sets containing controlled rotations about the Pauli axes. We use our technique to benchmark universal gate sets on four qubits of the Advanced Quantum Testbed, including a gate set containing a controlled-S gate and its inverse, and we investigate how the observed error rate is impacted by the inclusion of non-Clifford gates. Finally, we demonstrate that our technique scales to many qubits with experiments on a 27-qubit IBM Q processor. We use our technique to quantify the impact of crosstalk on this 27-qubit device, and we find that it contributes approximately 2/3 of the total error per gate in random many-qubit circuit layers.

quant-ph↗

arXiv2023

Two-Qubit Gate Set Tomography with Fewer Circuits

Gate set tomography (GST) is a self-consistent and highly accurate method for the tomographic reconstruction of a quantum information processor's quantum logic operations, including gates, state preparations, and measurements. However, GST's experimental cost grows exponentially with qubit number. For characterizing even just two qubits, a standard GST experiment may have tens of thousands of circuits, making it prohibitively expensive for platforms. We show that, because GST experiments are massively overcomplete, many circuits can be discarded. This dramatically reduces GST's experimental cost while still maintaining GST's Heisenberg-like scaling in accuracy. We show how to exploit the structure of GST circuits to determine which ones are superfluous. We confirm the efficacy of the resulting experiment designs both through numerical simulations and via the Fisher information for said designs. We also explore the impact of these techniques on the prospects of three-qubit GST.

quant-ph↗

arXiv2023

Extremum seeking control of quantum gates

To be useful for quantum computation, gate operations must be maintained at high fidelities over long periods of time. In addition to decoherence, slow drifts in control hardware leads to inaccurate gates, causing the quality of operation of as-built quantum computers to vary over time. Here, we demonstrate a data-driven approach to stabilized control, combining extremum-seeking control (ESC) with direct randomized benchmarking (DRB) to stabilize two-qubit gates under unknown control parameter fluctuations. As a case study, we consider these control strategies in the context of a trapped ion quantum computer using physically-realistic simulation. We then experimentally demonstrate this control strategy on a state-of-the-art, commercial trapped-ion quantum computer.

quant-ph↗

arXiv2022

Scalable randomized benchmarking of quantum computers using mirror circuits

The performance of quantum gates is often assessed using some form of randomized benchmarking. However, the existing methods become infeasible for more than approximately five qubits. Here we show how to use a simple and customizable class of circuits -- randomized mirror circuits -- to perform scalable, robust, and flexible randomized benchmarking of Clifford gates. We show that this technique approximately estimates the infidelity of an average many-qubit logic layer, and we use simulations of up to 225 qubits with physically realistic error rates in the range 0.1-1% to demonstrate its scalability. We then use up to 16 physical qubits of a cloud quantum computing platform to demonstrate that our technique can reveal and quantify crosstalk errors in many-qubit circuits.

quant-ph↗

arXiv2022

Precision tomography of a three-qubit donor quantum processor in silicon

Nuclear spins were among the first physical platforms to be considered for quantum information processing, because of their exceptional quantum coherence and atomic-scale footprint. However, their full potential for quantum computing has not yet been realized, due to the lack of methods to link nuclear qubits within a scalable device combined with multi-qubit operations with sufficient fidelity to sustain fault-tolerant quantum computation. Here we demonstrate universal quantum logic operations using a pair of ion-implanted 31P donor nuclei in a silicon nanoelectronic device. A nuclear two-qubit controlled-Z gate is obtained by imparting a geometric phase to a shared electron spin, and used to prepare entangled Bell states with fidelities up to 94.2(2.7)%. The quantum operations are precisely characterised using gate set tomography (GST), yielding one-qubit average gate fidelities up to 99.95(2)%, two-qubit average gate fidelity of 99.37(11)% and two-qubit preparation/measurement fidelities of 98.95(4)%. These three metrics indicate that nuclear spins in silicon are approaching the performance demanded in fault-tolerant quantum processors. We then demonstrate entanglement between the two nuclei and the shared electron by producing a Greenberger-Horne-Zeilinger three-qubit state with 92.5(1.0)% fidelity. Since electron spin qubits in semiconductors can be further coupled to other electrons or physically shuttled across different locations, these results establish a viable route for scalable quantum information processing using donor nuclear and electron spins.

quant-ph↗

arXiv2022

Measuring the Capabilities of Quantum Computers

A quantum computer has now solved a specialized problem believed to be intractable for supercomputers, suggesting that quantum processors may soon outperform supercomputers on scientifically important problems. But flaws in each quantum processor limit its capability by causing errors in quantum programs, and it is currently difficult to predict what programs a particular processor can successfully run. We introduce techniques that can efficiently test the capabilities of any programmable quantum computer, and we apply them to twelve processors. Our experiments show that current hardware suffers complex errors that cause structured programs to fail up to an order of magnitude earlier - as measured by program size - than disordered ones. As a result, standard error metrics inferred from random disordered program behavior do not accurately predict performance of useful programs. Our methods provide efficient, reliable, and scalable benchmarks that can be targeted to predict quantum computer performance on real-world problems.

quant-ph↗

arXiv2021

Gate Set Tomography

Gate set tomography (GST) is a protocol for detailed, predictive characterization of logic operations (gates) on quantum computing processors. Early versions of GST emerged around 2012-13, and since then it has been refined, demonstrated, and used in a large number of experiments. This paper presents the foundations of GST in comprehensive detail. The most important feature of GST, compared to older state and process tomography protocols, is that it is calibration-free. GST does not rely on pre-calibrated state preparations and measurements. Instead, it characterizes all the operations in a gate set simultaneously and self-consistently, relative to each other. Long sequence GST can estimate gates with very high precision and efficiency, achieving Heisenberg scaling in regimes of practical interest. In this paper, we cover GST's intellectual history, the techniques and experiments used to achieve its intended purpose, data analysis, gauge freedom and fixing, error bars, and the interpretation of gauge-fixed estimates of gate sets. Our focus is fundamental mathematical aspects of GST, rather than implementation details, but we touch on some of the foundational algorithmic tricks used in the pyGSTi implementation.

quant-ph↗

arXiv2021

Experimental Characterization of Crosstalk Errors with Simultaneous Gate Set Tomography

Crosstalk is a leading source of failure in multiqubit quantum information processors. It can arise from a wide range of disparate physical phenomena, and can introduce subtle correlations in the errors experienced by a device. Several hardware characterization protocols are able to detect the presence of crosstalk, but few provide sufficient information to distinguish various crosstalk errors from one another. In this article we describe how gate set tomography, a protocol for detailed characterization of quantum operations, can be used to identify and characterize crosstalk errors in quantum information processors. We demonstrate our methods on a two-qubit trapped-ion processor and a two-qubit subsystem of a superconducting transmon processor.

quant-ph↗

arXiv2021

Characterizing mid-circuit measurements on a superconducting qubit using gate set tomography

Measurements that occur within the internal layers of a quantum circuit -- mid-circuit measurements -- are an important quantum computing primitive, most notably for quantum error correction. Mid-circuit measurements have both classical and quantum outputs, so they can be subject to error modes that do not exist for measurements that terminate quantum circuits. Here we show how to characterize mid-circuit measurements, modelled by quantum instruments, using a technique that we call quantum instrument linear gate set tomography (QILGST). We then apply this technique to characterize a dispersive measurement on a superconducting transmon qubit within a multiqubit system. By varying the delay time between the measurement pulse and subsequent gates, we explore the impact of residual cavity photon population on measurement error. QILGST can resolve different error modes and quantify the total error from a measurement; in our experiment, for delay times above 1000 ns we measured a total error rate (i.e., half diamond distance) of $ε_{\diamond} = 8.1 \pm 1.4 \%$, a readout fidelity of $97.0 \pm 0.3\%$, and output quantum state fidelities of $96.7 \pm 0.6\%$ and $93.7 \pm 0.7\%$ when measuring $0$ and $1$, respectively.

quant-ph↗

arXiv2021

Efficient flexible characterization of quantum processors with nested error models

We present a simple and powerful technique for finding a good error model for a quantum processor. The technique iteratively tests a nested sequence of models against data obtained from the processor, and keeps track of the best-fit model and its wildcard error (a quantification of the unmodeled error) at each step. Each best-fit model, along with a quantification of its unmodeled error, constitute a characterization of the processor. We explain how quantum processor models can be compared with experimental data and to each other. We demonstrate the technique by using it to characterize a simulated noisy 2-qubit processor.

quant-ph↗

arXiv2021

A taxonomy of small Markovian errors

Errors in quantum logic gates are usually modeled by quantum process matrices (CPTP maps). But process matrices can be opaque, and unwieldy. We show how to transform a gate's process matrix into an error generator that represents the same information more usefully. We construct a basis of simple and physically intuitive elementary error generators, classify them, and show how to represent any gate's error generator as a mixture of elementary error generators with various rates. Finally, we show how to build a large variety of reduced models for gate errors by combining elementary error generators and/or entire subsectors of generator space. We conclude with a few examples of reduced models, including one with just $9N^2$ parameters that describes almost all commonly predicted errors on an N-qubit processor.

quant-ph↗

arXiv2020

Wildcard error: Quantifying unmodeled errors in quantum processors

Error models for quantum computing processors describe their deviation from ideal behavior and predict the consequences in applications. But those processors' experimental behavior -- the observed outcome statistics of quantum circuits -- are rarely consistent with error models, even in characterization experiments like randomized benchmarking (RB) or gate set tomography (GST), where the error model was specifically extracted from the data in question. We show how to resolve these inconsistencies, and quantify the rate of unmodeled errors, by augmenting error models with a parameterized wildcard error model. Adding wildcard error to an error model relaxes and weakens its predictions in a controlled way. The amount of wildcard error required to restore consistency with data quantifies how much unmodeled error was observed, in a way that facilitates direct comparison to standard gate error rates. Using both simulated and experimental data, we show how to use wildcard error to reconcile error models derived from RB and GST experiments with inconsistent data, to capture non-Markovianity, and to quantify all of a processor's observed error.

quant-ph↗