gate set tomography

Quantum device analog signal calibration is complex, time-consuming, and lacks systematic statistical evaluation methods. To address this, we propose SPAQ, the first framework to introduce statistical model checking (SMC) into quantum calibration. SPAQ enables probabilistic modeling and formal verification of DAG-structured calibration workflows, capturing critical properties—including parameter thresholds, implicit dependencies, and lower bounds on fault occurrence times. By integrating parameter sensitivity analysis with system availability assessment, SPAQ automates the optimization of calibration strategies. Experimental evaluation demonstrates that SPAQ effectively identifies dominant failure-inducing parameters and latent dependencies, significantly improving calibration efficiency and system availability for analog quantum systems. SPAQ establishes a verifiable, scalable statistical analysis paradigm for quantum hardware calibration.

This work addresses the lack of systematic experimental tracking in current quantum software development, which hinders effective monitoring of hardware noise, software evolution, and error sources. It introduces, for the first time, a holistic experimental tracking methodology tailored to quantum characteristics, proposing an end-to-end tracking framework that integrates error mitigation techniques with quantum reservoir computing. Validated through a chaotic time series prediction case study, the framework enables fully reproducible tracking of quantum experiments, accurately identifies critical error sources, and aggregates marginal gains across the workflow. The approach offers a generalizable methodological foundation for advancing quantum software engineering practices.

Addressing the unreliability of quantum measurements under depolarizing noise—dominant in the NISQ era—this paper proposes a statistically driven quantum error mitigation (QEM) method to accurately estimate the most probable noise-free output from noisy measurement samples. The method introduces an innovative two-stage “filtering + EM” framework: first, a heuristic filtering stage explicitly isolates and suppresses non-informative depolarizing noise; second, expectation-maximization (EM) is applied to the denoised data to enhance both interpretability and scalability of maximum-likelihood estimation. Small-scale experiments using Qiskit demonstrate that the approach significantly outperforms existing statistical QEM techniques. Further validation on synthetic datasets confirms its scalability to systems with ~100 qubits. By unifying theoretical rigor with engineering practicality, this work establishes a novel paradigm for error mitigation in intermediate-scale noisy quantum computation.

This work identifies and empirically validates a novel power-side-channel attack paradigm targeting quantum computer controllers: reconstructing gate-level quantum circuits—and even the original quantum algorithms—from a single power-consumption trace of real-device control pulses. We formally define and implement the first single-trace power analysis attack against quantum hardware, proposing two new reconstruction methods: (i) channel-separated brute-force reconstruction and (ii) full-power single-trace reconstruction based on mixed-integer linear programming (MILP). Leveraging algebraic analysis, per-channel power measurements, quantum pulse modeling, and simulation, we demonstrate high-fidelity circuit recovery across 32 real-world benchmark circuits. Our results underscore the urgent need for side-channel countermeasures in quantum hardware and provide critical empirical evidence and methodological foundations for designing quantum-secure architectures.

This work introduces the paradigm of *agnostic process tomography*: given query access to an unknown quantum channel Φ—without assuming Φ belongs to any prespecified model class—we select, from a given concept class ℂ, the channel that best approximates Φ. Our core techniques include Pauli spectrum analysis, superoperator spectral estimation, ancilla-enhanced state-tomography transfer, and efficient query sampling. We establish, for the first time, an agnostic learning transfer framework from quantum states to quantum processes, applicable to broad learnable classes including Pauli channels, quantum juntas, and QAC⁰ circuits. We design polynomial-time agnostic learning algorithms for Clifford circuits and circuits with few T-gates. Furthermore, we provide a theoretical characterization of sufficient conditions for agnostic learnability of quantum processes. These results advance foundational tools for quantum machine learning and error mitigation.

This work addresses the challenges in verifying and porting quantum programs, which stem from imperative gate sequencing, probabilistic outputs, and backend dependencies. To overcome these limitations, the authors propose a novel quantum domain-specific language (QDSL) featuring a separation-of-concerns architecture that decouples intent from execution, an introspectable intent-level intermediate representation (IR), and an assertion-guided mechanism for automatic execution-mode inference. This approach enables pre-execution circuit-structure validation, automatically selects the optimal execution modality without user intervention, and incorporates structured logging with endianness standardization. Experimental results demonstrate that IR generation incurs sub-millisecond overhead, the total variation distance-based fault detector achieves a true positive rate of 1.0 under fault injection, and compiled outputs exhibit numerical consistency across both PennyLane and Qiskit backends.

This work addresses the challenge of characterizing quantum device dynamics, particularly non-diagonal dissipative processes, which existing methods struggle to capture effectively due to their reliance on prior noise models, ancillary qubits, or complex control sequences. The authors propose an efficient reconstruction scheme requiring only product Pauli initial state preparation, a single uninterrupted evolution, and product Pauli measurements. Notably, the method identifies the support of a sparse Lindbladian without assuming locality. Leveraging compressed sensing principles, it robustly reconstructs all coefficients of the Hamiltonian and jump operators with $\tilde{O}(\Gamma^2 M_0^2 / \varepsilon^4)$ experimental repetitions and $\tilde{O}(\Gamma M_0^2 / \varepsilon^2)$ total evolution time to achieve accuracy $\varepsilon$, while providing theoretical robustness guarantees against calibrated initialization and measurement errors.

Existing model synchronization approaches struggle to accommodate the superposition and entanglement inherent in quantum systems, thereby failing to ensure consistency across heterogeneous models from multiple domains. To address this challenge, this work proposes QSysMM—the first dedicated model management framework tailored for engineered quantum systems—which establishes a unified digital single source of truth by harmonizing quantum engineering models along four dimensions: ontology, abstraction, composition, and exposure. Built upon the SysML v2 technology stack, we introduce QSysML, a companion modeling language that integrates model-driven engineering with quantum information science to preserve quantum semantics throughout model transformations. This framework provides a comprehensive foundation for the construction, verification, and maintenance of complex quantum systems through robust model synchronization and management.

This work presents the first integration of large language model (LLM) agents into nitrogen-vacancy (NV) center–based quantum sensing, establishing an end-to-end autonomous experimental workflow that fully automates the sequence from individual NV center selection and frequency calibration to T₂* measurement and CPMG sequence validation. The approach combines persistent experimental logging, deterministic hardware control, photoluminescence-detected magnetic resonance (pODMR) analysis, signal modeling, and quantitative evaluation tools, and introduces two offline benchmarks to independently assess scientific reasoning capabilities. Experimental results demonstrate that the agent efficiently executes complex quantum sensing tasks, and incorporating expected signal computation substantially reduces false-positive rates under high reasoning demands, thereby validating the effectiveness of synergistically combining LLMs with deterministic code for autonomous scientific discovery.

This study addresses the integrity challenges faced by quantum circuits in the NISQ era, including compilation-induced transformations, hardware constraints, and potential malicious tampering. Existing approaches fall short due to their reliance solely on structural or behavioral analysis, limiting comprehensive integrity assessment. To overcome this, the work proposes the first three-tier evaluation framework that jointly characterizes circuit integrity across structural, behavioral, and gate-interaction dimensions through a Structural Integrity Score (SIS), an Operational Integrity Score (OIS), and an Interaction Graph Semantic logic Score (IGS). The method innovatively employs Jensen–Shannon divergence to quantify behavioral deviations, constructs an interaction graph to capture pre-execution dependencies, and validates efficacy via controlled anomaly injection. Experiments demonstrate that, in structurally ambiguous cases, OIS and IGS achieve detection rates of 93.85% and 72.58%, respectively, significantly outperforming single-metric approaches.