Guest Editors Kerem Camsari and Supriyo Datta reflect on the Collection at its closing.

Pulsed squeezed light is a key resource for continuous-variable quantum information processing and photonic quantum technologies, and can be generated efficiently using Si${}_3$N${}_4$ microresonators. Under strong pulsed pumping, however, nonlinear effects complicate control of such light’s temporal and spectral properties, limiting performance and practical utility. This study in the high-gain regime investigates the impact of pump detuning and pulse duration on key metrics, including output photon flux and various correlations. Its results deepen our understanding of pulsed squeezed light in chip-scale resonators, and provide guidance for optimizing integrated quantum light sources.

Defect-free mixed-species atom arrays are promising for quantum computing, simulation, and metrology, but their scalability has been hindered. This study overcomes the barriers by expanding the tweezer-array size, improving atom-transfer efficiency, and introducing a powerful rearrangement algorithm. The authors successfully assemble defect-free arrays containing 120 mixed-species atoms with a filling fraction of 98.3% and a 14% defect-free probability—a real leap beyond prior demonstrations. This enhanced approach can be extended to other atomic species and is expected to accelerate progress in quantum error correction, many-body quantum simulations, and precision metrology.

Spin shuttling is a promising strategy for scaling silicon-based quantum processors by overcoming the connectivity constraints inherent in quantum dots. This study employs Pauli spin blockade to characterize spin-shuttling coherence at different external magnetic fields, facilitating a systematic investigation of the impact of various operational parameters, which can drive up to a twentyfold variation in error rates. Through targeted optimization, the authors achieve an average shuttling fidelity of 99.8%. Their findings provide critical insights for optimizing high-performance spin shuttling in future large-scale quantum processors.

Optical coherence tomography (OCT) using undetected photons is a promising technique for studying layered materials at wavelengths including the midinfrared, where traditional methods face challenges. However, OCT relies on large optical setups that require high laser power and are difficult to miniaturize. This work explores performance benchmarks for integrated sensors, which offer a path toward smaller, more practical devices, and finds that a less-common system configuration exploiting induced coherence works particularly well in integrated setups. This result not only improves performance but also provides useful guidance for designing future compact quantum sensing systems.

Noise-aware compilation on near-term quantum processors requires accurate noise models, but standard ones often miss algorithm- and hardware-specific error mechanisms, and full characterization can be costly. In this study a data-efficient framework combines a physically motivated parametrized noise model with Bayesian optimization, to infer algorithm- and hardware-specific error parameters from routine circuit-execution data. Remarkably, models trained only on small-circuit data generalize well to larger validation circuits, yielding 65% better model fidelity. This provides a scalable, low-overhead route to more predictive, application-aware noise models for quantum compilation workflows.

Birefringence in optical fiber causes polarization mode dispersion (PMD), which can broaden telecommunication signals, degrade fiber-sensor measurements, and scramble polarization-encoded quantum states. Though widely studied, PMD is typically modeled using statistical descriptions that obscure its underlying physical origins. The authors present BIFROST, a first-principles model that links PMD to specific physical parameters such as core geometry, temperature, and bend radius. Using this model, they simulate the impact of environmental variations on PMD compensation and demonstrate how knowledge of fiber properties (e.g. fiber spinning) can be applied to emerging quantum networks.

Hybrid sound-magnet interactions attract growing interest for advanced signal processing and computing, but generating complex, controllable nonlinear magnetic responses in such systems remains challenging for chip-scale platforms. The authors use a device that concentrates high-frequency sound waves to strongly couple a magnetic film to acoustic motion, revealing nonlinear magnetoelastic waves that generate phase-locked harmonic and subharmonic magnetic signals. Notably, the subharmonic process closely resembles optical parametric down-conversion. These nonlinear magnon-phonon hybrid excitations may represent an important step toward quantum magnonics with propagating excitations.

Probabilistic processors promise significant efficiency gains over GPUs, but scaling is bottlenecked by a reliance on hard-to-manufacture hardware for random number generation. In this study, researchers overcome this hurdle by demonstrating circuits built entirely from standard transistors that harness intrinsic thermal noise to efficiently sample from programmable probability distributions. Because these probabilistic circuits can be seamlessly integrated alongside standard CMOS cells, this approach paves the way for scalable, near-term probabilistic computing.

High-coherence superconducting quantum technologies demand as little microwave loss as possible, and niobium-based devices are often limited by dissipation associated with surface oxides. Inspired by processing developed for accelerator cavities, the authors employ a strategy to reduce oxide-related loss in a three-dimensional niobium cavity, achieving ultralow dissipation in the single-photon regime at millikelvin temperatures. This improved performance is largely preserved across multiple cooldown cycles, and after hours of air exposure. These results highlight oxide engineering as a practical route to longer-lived niobium-based qubits and resonators.

Thermoelectric technology is promising for energy conversion and solid-state refrigeration. As is well known, though, the strong coupling among the Seebeck coefficient, electrical conductivity, and lattice thermal conductivity substantially limits thermoelectric efficiency. Guided by orbital-mixing theory and first-principles calculations, the authors propose a biaxial-strain strategy to increase the Seebeck coefficient without sacrificing electrical conductivity (enhancing the power factor) and to weaken chemical bonding (suppressing lattice thermal conductivity), in two candidate materials. Consequently, tensile strains of 1–3% yield can double or triple the figure of merit at 300 K.

Cryogenic electronics requires amplifiers that can operate at millikelvin temperatures with low noise, while dissipating almost no power—two serious challenges. The authors propose and numerically analyze a fully voltage-controlled three-terminal superconducting transconductance amplifier based on thermally modulating a SINIS structure via quasiparticle injection through an additional NIS tunnel junction. Simulations predict millisiemens-level transconductance and high current gain with nanowatt power dissipation, suggesting a possible route to scalable low-power cryogenic amplification for quantum technologies and low-temperature detectors.

Although nonlocal connectivity is essential for universal logical gates and low-overhead quantum error correction, it is largely absent from today’s superconducting platforms, which are restricted to nearest-neighbor coupling. This work demonstrates an on-chip coupler with centimeter-scale interaction length, to provide high-fidelity, low-crosstalk nonlocal qubit coupling and serve as a building block for binary-tree connectivity graphs, reducing the average entangling distance from $O$($N$) to $O$(ln $N$). This capability supports the implementation of innovative quantum algorithms on superconducting processors, and strengthens their competitiveness with other hardware platforms.

Low-frequency cryogenic coolers typically set the infrastructure requirements for reaching extremely low temperatures, and unfortunately operate at just 1% of the Carnot limit. Using measurements and thermoacoustic analysis, the authors find that the status quo compression methodology is the greatest outstanding bottleneck for the overall efficiency of these cryocoolers. Comparison to a fresh compressor architecture and compressors for high-frequency cryocoolers suggests that we could promptly realize low-temperature technologies requiring significantly less electricity.

Imaging out-of-plane magnetization is essential for understanding and controlling nanoscale spintronic devices. Here researchers demonstrate a simple magnetic imaging technique based on the anomalous Nernst effect, using a conventional atomic force microscope. By touching the probe tip with a heated nanowire to induce a lateral temperature gradient, multidomain structures of out-of-plane magnetization are visualized with sub-200-nm resolution. This approach provides a practical, robust means of characterizing spin–orbit-torque-driven magnetic structures in nanoscale devices.

Precision frequency references are essential for quantum technologies and navigation systems, but stabilization methods are often constrained by narrow locking ranges and noise sensitivity, which thwart long-term stability and operation outside a laboratory. The authors use continuous spectroscopy with the vertical position of a magneto-optical trap as a frequency reference, to achieve resolution 30 times below the natural transition linewidth and expand locking range by orders of magnitude. Their direct extraction of optical and rf references with superior long-term stability offers a practical route toward robust references for quantum technology and potential GPS redundancy.

Frequency noise in nanomechanical resonators limits their performance in precision sensing and frequency-tracking applications, yet its origin—particularly that of intrinsic flicker ($1/f$) noise—remains poorly understood. This study uses ultracoherent nanomechanical resonators cointegrated with a photonic cavity to reveal strong correlations in the $1/f$ noise of distinct mechanical modes. By exploiting nonlinear optomechanical transduction, the authors generate an on-chip difference signal with strongly suppressed thermal and flicker frequency fluctuations, enabling direct frequency-noise cancellation.

Laser-induced discharge in a xenon jet is considered a promising source of extreme ultraviolet (EUV) light for lithography. In this Letter, the authors experimentally demonstrate that the EUV-emitting region extends outside the laser beam, which is important for understanding innovative EUV emission schemes for industrial applications.

Pulsed squeezed light is a key resource for continuous-variable quantum information processing and photonic quantum technologies, and can be generated efficiently using Si${}_3$N${}_4$ microresonators. Under strong pulsed pumping, however, nonlinear effects complicate control of such light’s temporal and spectral properties, limiting performance and practical utility. This study in the high-gain regime investigates the impact of pump detuning and pulse duration on key metrics, including output photon flux and various correlations. Its results deepen our understanding of pulsed squeezed light in chip-scale resonators, and provide guidance for optimizing integrated quantum light sources.

Defect-free mixed-species atom arrays are promising for quantum computing, simulation, and metrology, but their scalability has been hindered. This study overcomes the barriers by expanding the tweezer-array size, improving atom-transfer efficiency, and introducing a powerful rearrangement algorithm. The authors successfully assemble defect-free arrays containing 120 mixed-species atoms with a filling fraction of 98.3% and a 14% defect-free probability—a real leap beyond prior demonstrations. This enhanced approach can be extended to other atomic species and is expected to accelerate progress in quantum error correction, many-body quantum simulations, and precision metrology.

High-frequency A1-mode Lamb-wave resonators are promising for next-generation wireless communication, due to their strong electromechanical coupling, but spurious modes limit their performance. This study presents a circular resonator design that exploits anisotropic acoustic velocities in Z-cut lithium niobate and radially varying interdigital transducers to suppress unwanted modes, while maintaining efficient excitation of the A1 mode. Unexpectedly, the spatial interference of independent spurious modes enhances suppression beyond conventional designs. This approach offers a pathway to high-performance acoustic resonators for wideband rf applications.

Spin shuttling is a promising strategy for scaling silicon-based quantum processors by overcoming the connectivity constraints inherent in quantum dots. This study employs Pauli spin blockade to characterize spin-shuttling coherence at different external magnetic fields, facilitating a systematic investigation of the impact of various operational parameters, which can drive up to a twentyfold variation in error rates. Through targeted optimization, the authors achieve an average shuttling fidelity of 99.8%. Their findings provide critical insights for optimizing high-performance spin shuttling in future large-scale quantum processors.

Optical coherence tomography (OCT) using undetected photons is a promising technique for studying layered materials at wavelengths including the midinfrared, where traditional methods face challenges. However, OCT relies on large optical setups that require high laser power and are difficult to miniaturize. This work explores performance benchmarks for integrated sensors, which offer a path toward smaller, more practical devices, and finds that a less-common system configuration exploiting induced coherence works particularly well in integrated setups. This result not only improves performance but also provides useful guidance for designing future compact quantum sensing systems.

Magnetoelastic coupling between surface acoustic waves and spin waves in ferromagnetic thin films enables nonreciprocal signal processing and programmable wave filters, but self-consistent simulations of layered heterostructures are tricky, in terms of interfacial stress and strain discontinuities. This work presents a finite-difference time-integration scheme within an open-source micromagnetic library, to self-consistently solve the coupled magnetic and elastodynamic equations with rigorously enforced jump conditions at interfaces. Validated against analytical solutions and experimental data, this solver establishes a robust, open-source tool for designing magnon-phonon hybrid devices.

Noise-aware compilation on near-term quantum processors requires accurate noise models, but standard ones often miss algorithm- and hardware-specific error mechanisms, and full characterization can be costly. In this study a data-efficient framework combines a physically motivated parametrized noise model with Bayesian optimization, to infer algorithm- and hardware-specific error parameters from routine circuit-execution data. Remarkably, models trained only on small-circuit data generalize well to larger validation circuits, yielding 65% better model fidelity. This provides a scalable, low-overhead route to more predictive, application-aware noise models for quantum compilation workflows.

Birefringence in optical fiber causes polarization mode dispersion (PMD), which can broaden telecommunication signals, degrade fiber-sensor measurements, and scramble polarization-encoded quantum states. Though widely studied, PMD is typically modeled using statistical descriptions that obscure its underlying physical origins. The authors present BIFROST, a first-principles model that links PMD to specific physical parameters such as core geometry, temperature, and bend radius. Using this model, they simulate the impact of environmental variations on PMD compensation and demonstrate how knowledge of fiber properties (e.g. fiber spinning) can be applied to emerging quantum networks.

Hybrid sound-magnet interactions attract growing interest for advanced signal processing and computing, but generating complex, controllable nonlinear magnetic responses in such systems remains challenging for chip-scale platforms. The authors use a device that concentrates high-frequency sound waves to strongly couple a magnetic film to acoustic motion, revealing nonlinear magnetoelastic waves that generate phase-locked harmonic and subharmonic magnetic signals. Notably, the subharmonic process closely resembles optical parametric down-conversion. These nonlinear magnon-phonon hybrid excitations may represent an important step toward quantum magnonics with propagating excitations.

Probabilistic processors promise significant efficiency gains over GPUs, but scaling is bottlenecked by a reliance on hard-to-manufacture hardware for random number generation. In this study, researchers overcome this hurdle by demonstrating circuits built entirely from standard transistors that harness intrinsic thermal noise to efficiently sample from programmable probability distributions. Because these probabilistic circuits can be seamlessly integrated alongside standard CMOS cells, this approach paves the way for scalable, near-term probabilistic computing.

High-coherence superconducting quantum technologies demand as little microwave loss as possible, and niobium-based devices are often limited by dissipation associated with surface oxides. Inspired by processing developed for accelerator cavities, the authors employ a strategy to reduce oxide-related loss in a three-dimensional niobium cavity, achieving ultralow dissipation in the single-photon regime at millikelvin temperatures. This improved performance is largely preserved across multiple cooldown cycles, and after hours of air exposure. These results highlight oxide engineering as a practical route to longer-lived niobium-based qubits and resonators.

$XY$ (planar-spin) Hamiltonians arise in phase synchronization and retrieval and analog formulations of hard optimization, which motivates fast, low-power physical solvers. However, gain-based $XY$ systems that encode each spin with a single complex field can become trapped in metastable states. The authors introduce an annealer that represents each spin with two coupled complex components, and uses a graph-independent locking term that exploits the extra degree of freedom to bypass barriers. For challenging graph families, this higher-dimensional annealing improves ground-state recovery compared to one-component approaches, supporting more reliable photonic and analog $XY$ optimization.

Thermoelectric technology is promising for energy conversion and solid-state refrigeration. As is well known, though, the strong coupling among the Seebeck coefficient, electrical conductivity, and lattice thermal conductivity substantially limits thermoelectric efficiency. Guided by orbital-mixing theory and first-principles calculations, the authors propose a biaxial-strain strategy to increase the Seebeck coefficient without sacrificing electrical conductivity (enhancing the power factor) and to weaken chemical bonding (suppressing lattice thermal conductivity), in two candidate materials. Consequently, tensile strains of 1–3% yield can double or triple the figure of merit at 300 K.

Cryogenic electronics requires amplifiers that can operate at millikelvin temperatures with low noise, while dissipating almost no power—two serious challenges. The authors propose and numerically analyze a fully voltage-controlled three-terminal superconducting transconductance amplifier based on thermally modulating a SINIS structure via quasiparticle injection through an additional NIS tunnel junction. Simulations predict millisiemens-level transconductance and high current gain with nanowatt power dissipation, suggesting a possible route to scalable low-power cryogenic amplification for quantum technologies and low-temperature detectors.

Although nonlocal connectivity is essential for universal logical gates and low-overhead quantum error correction, it is largely absent from today’s superconducting platforms, which are restricted to nearest-neighbor coupling. This work demonstrates an on-chip coupler with centimeter-scale interaction length, to provide high-fidelity, low-crosstalk nonlocal qubit coupling and serve as a building block for binary-tree connectivity graphs, reducing the average entangling distance from $O$($N$) to $O$(ln $N$). This capability supports the implementation of innovative quantum algorithms on superconducting processors, and strengthens their competitiveness with other hardware platforms.