The properties of quantum dot–based spin qubits have significant interdevice variability due to the unavoidable presence of various types of disorder in semiconductor nanostructures. A significant source of this variability is charge disorder at the semiconductor-oxide interface, which causes unpredictable, yet, as we show here, correlated fluctuations in such essential properties of quantum dots, such as their mutual tunnel couplings, and electronic confinement energies. This study presents a systematic approach to characterize and mitigate the effects of such disorder. We use finite-element modeling of a $\mathrm{Si}$/$\mathrm{Si}$-$\mathrm{Ge}$ double quantum dot to generate a large statistical ensemble of devices, simulating the impact of trapped interface charges. This work results in a predictive statistical model capable of generating realistic artificial data for training machine-learning algorithms. By applying principal component analysis to this dataset, we identify the dominant modes through which disorder affects the multidimensional parameter space of the device. Our findings show that the parameter variations are not arbitrary, but are concentrated along a few principal axes—i.e., there are significant correlations between many properties of the devices. We finally compare that against control modes generated by sweeping the gate voltages, revealing limitations of the plunger-only control. This work provides a framework for enhancing the controllability and tunability of spin qubit devices, by systematically addressing the nature of electrostatic disorder that leads to statistical correlations in properties of double quantum dots.

Obtaining accurate representations of the eigenstates of an array of coupled superconducting qubits is a crucial step in the design of circuit quantum electrodynamics- (circuit-QED) based quantum processors. However, exact diagonalization of the device Hamiltonian is challenging for system sizes beyond tens of qubits. Here, we employ a tensor network method based on a variant of the density matrix renormalization group (DMRG) algorithm, DMRG-X, to efficiently obtain localized eigenstates of a two-dimensional transmon array without the need to first compute lower-energy states. We also introduce MTDMRG-X, a new algorithm that combines DMRG-X with multitarget DMRG to efficiently compute excited states even in regimes with strong eigenstate hybridization. We showcase the use of these methods for the analysis of long-range couplings in a multitransmon Hamiltonian including qubits and couplers, and we discuss eigenstate localization. These developments facilitate the design and parameter optimization of large-scale superconducting quantum processors.

Quantum computers based on cold-atom arrays offer long-lived qubits with programmable connectivity, yet the relatively low fidelity of site-selective, single-qubit control limits their programmability. We introduce an artificial-intelligence (AI) framework that overcomes this limitation. Trained on atom-laser dynamics in the presence of atomic motion in optical tweezers, a deep neural network autonomously designs robust composite pulses that increase local control fidelities tenfold while remaining compatible with existing control hardware. We further demonstrate the robustness of these pulses regarding optical aberrations and beam misalignment. This approach establishes AI-trained pulse compilation for high-fidelity qubit control, and can be readily extended to other atomlike platforms, such as trapped ions and solid-state color centers.

The automated design of parameterized quantum circuits for variational algorithms in the noisy intermediate-scale quantum (NISQ) era faces a fundamental limitation, as conventional differentiable architecture searches rely on classical models that fail to represent quantum gate interactions adequately under hardware noise. We introduce Quantum-Based Self-Attention for Differentiable Quantum Architecture Search (QBSA-DQAS), a metalearning framework featuring quantum-based self-attention and hardware-aware multiobjective searching for automated architecture discovery. The framework employs a two-stage quantum-based self-attention module. First, it computes contextual dependencies by mapping architectural parameters through parameterized quantum circuits, extracting high-dimensional feature representations that replace classical similarity metrics with quantum-derived attention scores. Second, it applies positionwise quantum transformations for feature enrichment. Architecture searching is guided by a task-agnostic multiobjective function, jointly optimizing noisy expressibility and the probability of successful trials. A postsearch optimization stage applies gate commutation, fusion, and elimination to reduce circuit complexity. Experimental validation demonstrates superior performance in variational quantum eigensolver (VQE) tasks and large-scale wireless sensor networks (WSNs). For VQE tasks on ${\mathrm{H}}_2$, QBSA-DQAS achieves 0.9 accuracy, compared with 0.89 for a standard differentiable quantum architecture search, outperforming classical applied attention baselines. Postsearch optimization via deterministic simplification rules reduces the complexity of the discovered circuits by up to 44% in gate count and 47% in depth without accuracy degradation. The framework maintains robust performance across three molecules and five IBM Quantum hardware noise models. For WSN routing, discovered circuits achieve 8.6% energy reduction versus the quantum approximate optimization algorithm and 40.7% versus the classical greedy method. These results establish the effectiveness of quantum-native architecture searching for NISQ applications.

Controlling the transport and nature of quantum excitations in low-dimensional systems is a key requirement for scalable quantum devices, including communication networks and quantum simulators. We propose a one-dimensional hybrid quantum lattice model, in which each lattice unit integrates a single-mode resonator that interacts with a two-level system, featuring direct coupling between adjacent two-level systems. This configuration enables the coherent propagation of excitations with tunable atomic, photonic, or polaritonic character. Beyond conventional single excitation transport, we demonstrate that appropriate impedance-matching and resonance conditions allow for the controlled swapping of excitation type as the excitation propagates along the lattice. We analyze the resulting dynamics using local observables and pairwise concurrence to track both transport and quantum correlations. Our results establish a minimal platform for controlled single excitation conversion, with direct relevance to hybrid quantum networks, on-chip quantum interconnects, and engineered quantum simulators.

In this work, we introduce a comprehensive statistical framework for assessing single-qubit quantum teleportation performance beyond the conventional average-fidelity benchmark. First, we derive a closed-form expression for the full probability density function of actual teleportation fidelities and apply it to both classical “measurement-and-prepare” schemes and standard quantum teleportation, considering two relevant noise models: Bell-diagonal resource states and local amplitude-damping channels. The results reveal that protocols with identical average fidelities can exhibit markedly different statistical behaviors, and that relying solely on average fidelity can mask inherent asymmetries introduced by local noise, potentially leading to spurious conclusions about symmetry. Second, we introduce a certification method based on prior importance functions (e.g., beta distributions), which unifies moment-based criteria and threshold-based success probabilities into a single figure of merit. Applying this framework, we see that certifying high-fidelity teleportation requires increasingly stronger entanglement or nonlocality, and we show that the “fighting noise with noise” effect is a feature specific to the first statistical moment (average fidelity), which masks the asymmetric degradation of local marginals, and vanishes for higher-order criteria. Our approach thus provides versatile tools for tailored, application-specific teleportation benchmarks.

Time-bin entanglement is a commonly used form of entanglement for quantum communication protocols over fiber networks, due to the natural resilience of this encoding scheme to thermal phase fluctuations and to birefringence in optical fibers. Projective measurements on some bases in the time-bin encoding need, however, postselection in the time domain of the measured events, introducing a loophole in Bell tests and requiring high temporal resolution. In this work, we demonstrate chip-integrated receivers for time-bin entanglement certification, including a high-speed optical switch to remove the temporal postselection loophole. The receivers are realized using thin-film lithium niobate and operate at a switching frequency of 5 GHz. We demonstrate a Bell inequality violation by more than 24 standard deviations without the need for time resolution at the time-bin separation level.

We introduce a protocol that modifies dynamical decoupling (DD) sequences to be robust to static $ZZ$ crosstalk when implemented with bounded control on 2-colorable qubit topologies. The protocol, which relies on modifications of the pulse timing, can be applied to any sequence with equidistant $\pi$ pulses. We motivate the method using suppression conditions identified through time-dependent perturbation theory. Theoretical findings are supported by demonstrations of widely studied sequences on several superconducting qubit devices offered by the IBM Quantum Platform. Using up to 20 qubits on fixed-coupler devices, we observe an at least 3 times improvement in the fidelity decay rate via our approach when compared with nonrobust DD variants. In addition, we leverage our approach to assess the impact of $ZZ$ errors on tunable-coupler devices. We find that $ZZ$-robust sequences perform nearly equivalently to nonrobust DD, affirming the reduced impact of such errors in a tunable-coupler architecture. Nevertheless, our demonstrations indicate that fixed-coupler devices, when subject to DD protection, can outperform tunable-coupler devices. Our method broadens the scope of practical DD protocols: with modest overhead and a reasonable constraint on the qubit topology, the method attains significant performance improvements on modern quantum computing devices.

The development of fault-tolerant quantum computers (FTQCs) is receiving increasing attention within the quantum computing community. Like conventional digital computers, FTQCs, which utilize error correction and millions of physical qubits, have the potential to address some of humanity’s grand challenges. However, accurate estimates of the tangible scale of future FTQCs, based on transparent assumptions, are uncommon. How many physical qubits are necessary to solve a practical problem intractable for classical hardware? What costs arise from distributing quantum computation across multiple machines? This paper presents an architectural model of a potential FTQC based on superconducting qubits, divided into discrete modules and interconnected via coherent links. We employ a resource-estimation framework and software tool to assess the physical resources required to execute specific quantum algorithms compiled into their graph-state form and arranged onto a modular superconducting hardware architecture. Our tool can predict the size, power consumption, and execution time of these algorithms based on explicit assumptions about the physical layout, thermal load, and modular connectivity of the system. We assess the resources needed for quantum computation examples that serve as building blocks of proposed applications, quantifying the architectural bottlenecks and trade-offs that remain to be addressed to deliver utility.

We propose a neural-network-based model capable of learning the broad landscape of working regimes in quantum dot simulators and using this knowledge to autotune these devices—based on transport measurements—toward obtaining Majorana modes in the structure. The model is trained in an unsupervised manner on synthetic data in the form of conductance maps, using a physics-informed loss that incorporates key properties of Majorana zero modes. We show that, with appropriate training, a deep vision-transformer network can efficiently memorize the relations between Hamiltonian parameters and structures on conductance maps and use them to propose parameter updates for a quantum dot chain that drives the system toward a topological phase. Starting from a broad range of initial detunings in parameter space, a single update step is sufficient to generate nontrivial zero modes. Moreover, by enabling an iterative tuning procedure—where the system acquires updated conductance maps at each step—we demonstrate that the method can address a much larger region of the parameter space.

Superinductances are superconducting circuit elements that combine a large inductance with a low parasitic capacitance to ground, resulting in a characteristic impedance exceeding the resistance quantum $R_Q=h/(2e{)}^2\simeq 6.45\mathrm{k}\mathrm{\Omega }$. In recent years, these components have become key enablers for emerging quantum circuit architectures. However, achieving high characteristic impedance while maintaining scalability and fabrication robustness remains a major challenge. In this work, we present two fabrication techniques for realizing superinductances based on vertically stacked Josephson junctions. Using a multiangle Manhattan (MAM) process and a zero-angle (ZA) evaporation technique—in which junction stacks are connected pairwise using airbridges—we fabricate one-dimensional chains of stacks that act as high-impedance superconducting transmission lines. Two-tone microwave spectroscopy reveals the expected $\sqrt{n}$ scaling of the impedance with the number of junctions per stack. The chain fabricated using the ZA process, with nine junctions per stack, achieves a characteristic impedance of approximately $16\mathrm{k}\mathrm{\Omega }$, a total inductance of $5.9\text{μ}\mathrm{H}$, and a maximum frequency-dependent impedance of $50\mathrm{k}\mathrm{\Omega }$ at 1.4 GHz. Our results establish junction stacking as a scalable, robust, and flexible platform for next-generation quantum circuits requiring ultrahigh impedance environments.

Hardware-algorithm co-design provides an effective route for implementing quantum optimization on near-term hardware. Counterdiabatic (CD) protocols can accelerate adiabatic optimization but are difficult to realize digitally due to increased circuit depth and noise sensitivity. Here, we report a photonic co-design strategy that directly implements CD dynamics on a programmable photonic integrated circuit based on a Mach-Zehnder interferometer mesh. By exploiting the ability of the photonic processor to realize arbitrary unitary transformations, our approach bypasses gate-level decomposition and significantly reduces circuit depth. We further enhance the protocol by mitigating higher-order many-body terms in conventional CD optimization. An experimental demonstration of integer factorization achieves success probabilities up to 86% with only three layers, together with accelerated energy convergence, highlighting programmable photonic platforms for quantum optimization.

Two-level systems (TLSs) are a dominant decoherence mechanism in superconducting qubits, microwave kinetic inductance detectors, and quantum dots. TLSs are tunneling states commonly found in amorphous materials and interfaces, which interact electromechanically, leading to energy loss. Fluctuations in the frequency of the TLSs are a significant problem for qubit operation and force recalibration every few hours. In this study, we probe the effect of alternating bias at cryogenic temperatures on TLSs in a purpose-built LC oscillator. When an in situ alternating bias is applied, the steady-state spectra of the TLSs disappear. Spectroscopy reveals transient behavior, in which the TLS frequency fluctuates on the order of minutes. Thermal cycling above 10 K reverses these effects, restoring the TLS spectrum to its original state. Importantly, the intrinsic loss tangent of the LC oscillator remains unchanged before and after the application of the alternating bias. We propose that the disappearance of the steady-state spectrum is caused by nonequilibrium energy buildup from strain in the oxide film introduced by the pulsed voltage-bias sequence. Understanding this nonequilibrium energy could shed light on TLS fluctuations and reduce the calibration requirements for qubits.

Measurement-device-independent quantum key distribution (MDI-QKD) can be immune to all detector-side loopholes. Compared with weak-coherent-state sources, heralded single-photon sources (HSPSs) can effectively suppress vacuum components, and thus substantially improve the transmission distance and secure key rate of MDI-QKD. Here, we experimentally demonstrate an efficient MDI-QKD protocol based on HSPSs without vacuum sources [Phys. Rev. A 98, 032303 (2018)]. It achieves a secure key rate that is more than 60 times higher than the previous reported work at 10-dB transmission loss and extends the maximum transmission loss to more than 30 dB. These improvements are attributed to the efficient protocol used and the high-quality HSPSs prepared in the experiment, which enable a Hong-Ou-Mandel interference visibility of 95.1% at 10 mW. Furthermore, with improvements to experimental technology, the maximum transmission distance is expected to exceed 400 km. Therefore, the current work is expected to play a crucial role in the future development of large-scale quantum communication networks.

In the noisy intermediate-scale quantum era, quantum devices can only execute shallow-depth circuits to maintain acceptable fidelity. This poses a significant challenge in implementing quantum algorithms that require interactions between nonadjacent qubits, especially on superconducting platforms with limited qubit connectivity. In conventional methods, swap gates are often employed to bridge distant qubits, at the cost of increased circuit depth and reduced overall fidelity. In this paper, we propose a scheme for implementing a controlled-$Z$ gate between next-nearest-neighbor qubits ($N^3\text{−}\mathrm{C}Z$), which enables direct high-fidelity entanglement between nonadjacent qubits without disturbing the intermediate qubit. Implemented on fixed-frequency and fixed-coupler superconducting qubits via simultaneous cross-resonance drives, the $N^3\text{−}\mathrm{C}Z$ gate achieves an average gate fidelity of 99.0% under realistic decoherence conditions, which is further confirmed by quantum process tomography with a fidelity of 98.9%. Moreover, we demonstrate that the $N^3\text{−}\mathrm{C}Z$ gate significantly reduces circuit depth and gate count in key quantum applications and circuit components, including quantum error correction, the quantum fan-out gate, the Toffoli gate, quantum ripple-carry adder, and graph state preparation. These results highlight the potential of the $N^3\text{−}\mathrm{C}Z$ gate as an efficient and hardware-friendly primitive for scalable quantum computation.

The integration of photonic components into surface ion traps provides a scalable approach for trapped-ion quantum computing, sensing, and metrology, enabling compact systems with enhanced stability and precision. However, the introduction of optical apertures in the trap electrodes can distort the trapping electric field. This can lead to excess micromotion and ion displacement, which degrade the performance of quantum logic operations and optical clocks. In this work, we systematically investigate the electric field distortion in a surface ion trap with integrated waveguides and grating couplers that require apertures in the surface electrode using finite element method simulations. We analyze methods to reduce these distortions by exploiting symmetries and transparent conductive oxide materials.