High-fidelity circuit execution on noisy intermediate-scale quantum devices is bottlenecked by compilation pipelines that disregard complex, correlated noise. To address this, this methodology article proposes a quantum machine learning control (QMLC) framework for generative quantum circuit synthesis from gate-set tomography (GST) data that bypasses the traditional two-step pipeline of characterizing native quantum gates via GST followed by unitary decomposition algorithms. Instead, a generative concept space is directly learnt from GST data, enabling conditional synthesis of quantum circuits on a desired output distribution. Our approach tokenizes GST germ circuits and embeds them into a structured latent space using a curriculum-learning-motivated strategy, starting with short circuits and progressively incorporating longer ones with diverse output statistics. The embedded sequences are processed by a set-vision transformer with permutation-invariant pooling, producing k-seed vectors that represent the learned concept space of the quantum device. Aggregating data across multiple circuits makes this latent representation inherently context-aware, capturing the shared physical noise environment (e.g., crosstalk, drift) that isolated gate metrics miss. We propose an unconditional diffusion model to sample from the concept space. During inference, a user provides a target measurement distribution, and the model generates a corresponding circuit. To ensure fidelity and robustness, the output is denoised using a diffusion model that operates on the target conditional covariance matrix. This end-to-end framework is a step towards context-aware, hardware-native circuit synthesis directly from raw GST data, which offers a new paradigm for integrating quantum control and compilation. The QMLC framework is particularly suited for near-term quantum devices with complex calibration procedures.

Memristor technology offers a promising route toward energy-efficient computing but faces challenges including resistance drift, variability, and the need for electroforming. Filamentary resistive random-access memory, one of the most studied memristive platforms, typically requires a high-voltage electroforming step to initiate conductive filaments, leading to increased power overhead and reduced endurance. Here we report HfO2-based forming-free memristive devices (PdNeuRAM) that operate at low voltages, support multi-bit functionality, and exhibit reduced variability. Through combined electrical and materials characterization, we identify a Pd-O-Hf interfacial configuration that lowers oxygen-vacancy formation and migration barriers, creating a dense network of shallow defect states. Together with a Ti top electrode acting as an oxygen reservoir and an ultrathin (5 nm) HfO2 layer, this interfacial engineering enables charge redistribution at room temperature and eliminates the need for electroforming. The fabricated devices provide tunable resistance states and reduce programming and read energy by 43% and 38%, respectively, in spiking neural network inference tasks. These results provide mechanistic insight into forming-free resistive switching and demonstrate the potential of Pd/HfO2 devices for energy-efficient neuromorphic computing.

Tin vacancy (SnV) centers in diamond are promising solid state qubits for integrated quantum photonics. Here, we fabricate and characterize a diamond on Al2O3 dual taper waveguide structure containing SnV centers, demonstrating optical coupling between the diamond nanobeam and the underlying Al2O3 waveguide. The devices are realized using a bilayer fabrication approach compatible with wafer scale lithography. Clear guided SnV- emission is observed in all optically active devices, indicating effective optical coupling in the integrated structure. These results demonstrate a scalable fabrication approach toward integrating diamond color centers with photonic waveguides.

While magnetic fields and superconductors are both central to classical and quantum technologies, their combined use is often challenging, as magnetic fields significantly affect superconducting device performance. In superconducting nanowire single-photon detectors (SNSPDs), magnetic fields drastically reduce detection efficiencies, hampering their application in magnetically-active classical and quantum photonics. Here, we systematically characterize the performance of NbTiN SNSPDs under magnetic fields and show the enhancement of their intrinsic detection efficiency (IDE) at lower bias currents and its suppression at higher currents. This leads to SNSPD performance degradation through reduced or disappearing saturation plateaus. We show that the magnitude of this degradation is highly dependent on nanowire width and demonstrate width-optimized SNSPDs with saturating IDE for a wide range of photon energies under application-relevant magnetic fields. Minimizing degradation in superconducting devices under magnetic fields enables applications like detector-integrated spin-optic and atomic quantum processors, high-sensitivity magnetometry, and quantum transduction.

We report progress toward a CMOS-integrated quantum diamond biosensing platform that combines nitrogen-vacancy (NV) centers in diamond with a custom 40 nm CMOS Single-Photon Avalanche Diode (SPAD) array. The system integrates on-chip active quenching and digital readout with external FPGA-based photon counting, compact microwave delivery, and practical optical excitation and collection schemes to support widefield optically detected magnetic resonance (ODMR). System-level design considerations spanning fluorescence collection efficiency, detector count-rate capability, and microwave homogeneity are analyzed with biological compatibility and scalability in mind. Using superparamagnetic iron oxide nanoparticle (SPION)-labeled HEK293T cells as a representative use case, simple dipole-field estimates indicate that sub-μT sensitivity is required to resolve ODMR shifts within typical ensemble linewidths. Based on the proposed architecture and efficiency analysis, a magnetic field sensitivity of approximately 90 nT/√(Hz) per pixel is estimated. These results outline a practical path from optics-heavy quantum diamond microscopes toward compact, CMOS-integrated NV-based biosensors for quantitative magnetic imaging in complex biological environments.

Accurate quantum state tomography (QST) is vital for calibrating quantum processors but faces exponential scaling challenges. We benchmark seven neural architectures—FCN, CNN, CGAN, Transformer, RNN, RBM, and SVAE—for QST reconstruction using expectation- and probability-based measurements. CNN and CGAN achieve high fidelity (F > 0.99), while SVAE enables efficient event-driven learning. To enhance scalability, memristor-based computation-in-memory (CiM) acceleration is proposed for CNN and SVAE, leveraging analog matrix–vector multiplication in HfO2 crossbars. The fabricated arrays show stable bipolar switching and STDP behavior, advancing energy-efficient, real-time quantum diagnostics through algorithm–hardware co-design.

Efficient elimination of unwanted light while transmitting lights with desired wavelengths is crucial in integrated photonics. We demonstrate a band-stop filter in alumina low-loss visible-light waveguides that eliminates green light while transmitting red light, applicable to diamond spin-based quantum computing. The filter employing a multimode waveguide Bragg grating (WBG) where forward TE0 and backward TE2 modes couple strongly, exhibits single-mode red light transmission and superior green light elimination compared to conventional TE0-TE0 coupling. The elimination wavelength changes linearly with the WBG period, offering tunability to target unwanted green wavelengths. A minimum transmittance of 10-3 is achieved for N = 10 000 periods. This band-stop filter is a key component for integrated photonics, especially for on-chip photonic quantum processors.

Diamond color-center qubits integrated with photonic circuits can be initialized, manipulated, entangled, and read individually with high fidelity, making them attractive for large-scale modular quantum computers, quantum networks, and distributed quantum sensing. However, the limited size of heteroepitaxially grown single-crystal diamond (SCD) and photonic-grade diamond-on-insulator (DOI) substrates remains a challenge for integration with existing manufacturing processes. Here, we develop a plasma etch recipe to thin direct-bonded (100) SCD membranes (<50 μm) into large-area, thin-film DOI substrates, and demonstrate free-standing photonic chiplets fabricated from the resulting DOI. The ICP-RIE recipe preserves diamond bonding, provides sufficient micromasking and surface-quality control, and enables thin-film DOI manufacture. We thin a 10 μm diamond plate bonded to SiO_2/Si and obtain a photonic-grade DOI substrate with diamond thickness ≤300 nm. The DOI film is around 300 nm thick over 0.5 × 0.5 mm^2, with surface roughness < 0.5 nm, while the bonding interface remains intact. Diamond photonic chiplets are fabricated on this DOI substrate using a standard two-step lithography process, without complex thin-film transfer, under-etching, or pedestal formation. We also present a colorimetric study of diamond visibility on SiO_2 and quantify color differences across thicknesses in common colorimetric spaces. This analysis enables automatic diamond-thickness extrapolation from standard optical microscope images with 5 nm resolution, in good agreement with white-light interferometry (WLI) measurements. The DOI substrate and colorimetric thickness-evaluation method provide an effective fabrication platform and reliable validation route for scalable manufacturing of diamond nanophotonic devices, opening a path toward large-scale integrated quantum systems.

Superconducting nanowire single-photon detectors (SNSPDs) have emerged as leading cryogenic photon detectors, thanks to their high detection efficiency and low jitter. However, their large-scale integration remains limited by the wiring bottleneck between the cryogenic detectors and their room-temperature readout electronics. In applications such as color-center-based quantum computers (QCs), thousands of detectors may need to operate in parallel within a limited cryogenic cooling budget, thus asking for a scalable, low-power cryogenic electronic readout. To address these needs, this work introduces a cryogenic readout circuit directly wire-bonded to the SNSPD and using a high-impedance input to maximize the quality of the detector signal, thus relaxing the requirement of the cascaded amplifier and reducing its power consumption. An active quenching circuit is then adopted to ensure a reliable reset after the latching of the detector induced by such high input impedance. Implemented in 40-nm CMOS with an active area of <0.14 mm(2), the system achieves competitive performance at 0.1 K, delivering low timing jitter (<40 ps), high speed (dead time of approximate to 5 ns), and dark count rates (DCRs) below 1 Hz, while achieving a 5 & times; reduction in power consumption (down to 20 mu W) with respect to the cryogenic-readout state-of-the-art. Its ultralow-power operation and compact footprint make the proposed solution well-suited for integration within large-scale quantum-computing architectures.

Color centers in diamond are a promising platform for quantum computing applications because of their optical and spin properties. However, diamond presents some technological challenges that limit its use in complex or large photonic circuits. To mitigate these limitations, it is technically effective to separate the smallest possible diamond photonic structures or chiplet containing the color center(s) from the rest of the circuit, which is fabricated on another material platform, and then heterogeneously integrate them. Considering efficient excitation and photon collection from waveguide-coupled color centers, we design a cross waveguide as the primary component of our chiplet to access the color centers, channeling excitation and emitted photons into different waveguides, and connecting the structure to the other components of the photonic circuit. The chiplet containing the cross waveguide and supporting structures requires careful optimization of each subcomponent. The receptor's design is also critical for optimal signal transmission. In this paper, we develop a simple but efficient methodology to optimize the main components constituting both the chiplet and the receptor for their synergistic operation. The designed structure has an excitation-to-emission conversion of more than 5.4

Quantum State Tomography (QST) is essential for characterizing and validating quantum systems, but its practical use is severely limited by the exponential growth of the Hilbert space and the number of measurements required for informational completeness. Many prior claims of performance have relied on architectural assumptions rather than systematic validation. We benchmark several neural network architectures to determine which scale effectively with qubit number and which fail to maintain high fidelity as system size increases.To address this, we perform a comprehensive benchmarking of diverse neural architectures across two quantum measurement strategies to evaluate their effectiveness in reconstructing both pure and mixed quantum states. Our results reveal that CNN and CGAN scale more robustly and achieve the highest fidelities, while Spiking Variational Autoencoder (SVAE) demonstrates moderate fidelity performance, making it a strong candidate for embedded, low-power hardware implementations.Recognizing that practical quantum diagnostics will require embedded, energy-efficient computation, we also discuss how memristor-based Computation-in-Memory (CiM) platforms can accelerate these models in hardware, mitigating memory bottlenecks and reducing energy consumption to enable scalable in-situ QST. This work identifies which architectures scale favorably for future quantum systems and lays the groundwork for quantum-classical co-design that is both computationally and physically scalable.

IntroductionIn 2012, potassium and sodium ion channels in Hodgkin-Huxley-based brain models were shown to exhibit memristive behavior. This positioned memristors as strong candidates for implementing biologically accurate artificial neurons. Memristor-based brain simulations offer advantages in energy efficiency, scalability, and compactness, benefiting fields such as soft robotics, embedded systems, and neuroprosthetics.MethodsPrevious approaches used current-controlled Mott memristors, which poorly matched the voltage-controlled nature of ion channels. This study employs volatile, oxide-based memristors that leverage electric-field-driven oxygen-vacancy migration to emulate voltage-dependent channel behavior. We selected candidate WOx and NbOx memristors and modeled their dynamics to verify performance as Hodgkin-Huxley potassium channels.ResultsThe device exhibits sigmoidal gating and voltage-dependent time constants consistent with the theoretical model. By scaling the passive circuitry around the memristors, we show that they capture the essential mechanisms of potassium ion-channels, although spike height is reduced due to strong non-linear voltage-dependence. Still, by cascading multiple compartments, typical spike propagation is retained.DiscussionThis is the first demonstration of a voltage-controlled memristor replicating the Hodgkin-Huxley potassium channel, validating its potential for more efficient brain simulation hardware.

Quantum computation represents a promising frontier in the domain of high-performance computing, blending quantum information theory with practical applications to overcome the limitations of classical computation. This study investigates the challenges of manufacturing high-fidelity and scalable quantum processors. Quantum gate set tomography (QGST) is a critical method for characterizing quantum processors and understanding their operational capabilities and limitations. This paper introduces Ml4Qgst as a novel approach to QGST by integrating machine learning techniques, specifically utilizing a transformer neural network model. Adapting the transformer model for QGST addresses the computational complexity of modeling quantum systems. Advanced training strategies, including data grouping and curriculum learning, are employed to enhance model performance, demonstrating significant congruence with ground-truth values. We benchmark this training pipeline on the constructed learning model, to successfully perform QGST for 2 and 3 gates on single-qubit and two-qubit systems, with over-rotation error and depolarizing noise estimation with comparable accuracy to pyGSTi. This research marks a pioneering step in applying deep neural networks to the complex problem of quantum gate set tomography, showcasing the potential of machine learning to tackle nonlinear tomography challenges in quantum computing.

Quantum computers are expected to unlock information processing capabilities and speed up simulation times to rates that cannot be achieved by classical computers [1]. But the technology is still in its nascent stages compared to the aspirations of achieving large-scale fault-tolerant computing.

Low-loss visible-light photonic circuits are crucial for high-performance photonic quantum processors. By using aluminum oxide (Al2O3) for its low visible-light absorption, we achieved waveguides exhibiting an exceptionally low propagation loss (1.39 dB/cm for the transverse electric mode) at red-light wavelengths. Directional coupler beam splitters fabricated using this platform exhibited good controllability of the optical splitting ratios. Furthermore, we fabricated a half beam splitter, which is an essential component of entangled photon generation in quantum optics. These results represent a significant advance toward developing low-loss photonic circuits, paving the way for improved performance in photonic quantum processors.

Cavity-enhanced diamond color center qubits can be initialized, manipulated, entangled, and read individually with high fidelity, which makes them ideal for large-scale, modular quantum computers, quantum networks, and distributed quantum sensing systems. However, diamond's unique material properties pose significant challenges in manufacturing nanophotonic devices, leading to fabrication-induced structural imperfections and inaccuracies in defect implantation, which hinder reproducibility, degrade optical properties and compromise the spatial coupling of color centers to small mode-volume cavities. A cavity design tolerant to fabrication imperfections-such as surface roughness, sidewall slant, and nonoptimal emitter positioning-can improve coupling efficiency while simplifying fabrication. To address this challenge, a deep learning-based optimization methodology is developed to enhance the fabrication error tolerance of nanophotonic devices. Convolutional neural networks (CNNs) are applied to promising designs, such as L2 and fishbone nanobeam cavities, predicting Q-factors at least one-million times faster than traditional finite-difference time-domain (FDTD) simulations, enabling efficient optimization of complex, high-dimensional parameter spaces. The CNNs achieve prediction errors below 3.99% and correlation coefficients up to 0.988. Optimized structures demonstrate a 52% reduction in Q-factor degradation, achieving quality factors of 5 x 104 under real-world conditions and a 2-fold expansion in field distribution, enabling efficient coupling of nonoptimally positioned emitters. Compared to previous deep-learning optimization methods, this approach achieves twice the Q-factor performance in the presence of fabrication errors, significantly enhancing device robustness. Hence, this methodology enables scalable, high-yield manufacturing of robust nanophotonic devices, including the cavity-enhanced diamond quantum systems developed in this study.

Diamond has emerged as a leading material for solid-state spin quantum systems and extreme environment electronics. However, a major limitation is that most diamond devices and structures are fabricated using bulk diamond plates. The absence of a suitable diamond-on-insulator (DOI) substrate hinders the advanced nanofabrication of diamond quantum and electronic devices, posing a significant roadblock to large-scale, on-chip diamond quantum photonics and electronics systems. In this work, we demonstrate the direct bonding of (100) single-crystal diamond plates to PECVD-grown SiO2/Si substrates at low temperatures and atmospheric conditions. The surfaces of the SiO2 and diamond plates are then activated using oxygen plasma and Piranha solution, respectively. Bonding occurs when the substrates are brought into contact with water in between and annealed at 200 °C under atmospheric conditions, resulting in a DOI substrate. We systematically studied the influence of Piranha solution treatment time and diamond surface roughness on the shear strength of the bonded substrate, devising an optimal bonding process that achieves a high yield rate of 90% and a maximum shear strength of 9.6 MPa. X-ray photoelectron spectroscopy was used for quantitative analysis of the surface chemicals at the bonding interface. It appears that the amount of –OH bindings increases with the initial roughness of the diamond, facilitating the strong bonding with SiO2. This direct bonding method will pave the way for scalable manufacturing of diamond nanophotonic devices and enable large-scale integration of diamond quantum and electronic systems.

We demonstrate interface-enhanced memristors (OxReRAM) tailored for cryogenic spin-qubit control. By engineering a sparse filament network, our devices achieve eight nonvolatile resistance levels with an ultra-low read noise rate of around 0.3 %. When embedded in a cryogenic gain stage with R-L = 30 k Omega and V-in = 0.3V, it will deliver a +/- 1 V output range and sub-100-mu V resolution using only six memristors per channel. This single-line biasing architecture will reduce wires, paving the way for large-scalce quantum processors.

The direct bonding process of a diamond‐on‐insulator (DOI) substrate enables monolithic integration of diamond photonic structures for quantum computing by improving photon collection efficiency and entanglement generation rate between emitters. It also addresses key fabrication challenges, such as robustness, bonding strength, and scalability. This study investigates strain effects in DOI substrates following direct bonding. Strain generation is expected near the diamond–SiO 2 /Si interface due to the thermal expansion coefficient mismatch between the bonded materials. Strain‐induced lattice distortions are characterized using nitrogen‐vacancy (NV) centers in diamond via optically detected magnetic resonance (ODMR) and photoluminescence (PL) mapping. PL mapping reveals interference fringes in unbonded regions, indicating bonding irregularities. Depth‐resolved ODMR measurements show a volumetric strain component increase of ≈0.45 MHz and a shear component increase of ≈0.71 MHz between the top surface and the DOI interface. However, ODMR signal contrast and peak linewidth remain largely unaffected, suggesting no visible deterioration in the optical properties of the emitters. By combining ODMR and PL mapping, this work establishes a robust methodology for assessing bonding quality and strain impact on NV centers, an essential step toward advancing scalable quantum technologies and integrated photonic circuits.