Reservoir computing (RC) provides an efficient framework for processing temporal information. However, spin-orbit-torque (SOT)-based physical RC systems are fundamentally limited by a trade-off between stochasticity in device responses and the richness of reservoir dynamics. Here, we address this challenge through a device-architecture co-design strategy. At the device level, we demonstrate that edge modulation of Hall-bar SOT devices effectively suppresses stochasticity in magnetization switching by regulating domain-wall depinning dynamics, thereby improving the stability of reservoir states. At the architecture level, we introduce a multiport configuration that integrates physically distinct nonlinear responses from different terminals, enhancing the dimensionality and dynamic richness of the reservoir. Within a time-division multiplexed RC framework, the reduced stochasticity achieved by edge modulation leads to improved accuracy in speech recognition and waveform classification tasks. Furthermore, in the Mackey–Glass chaotic time-series prediction task, the multiport reservoir achieves longer effective prediction horizons and lower normalized root-mean-square error compared with single-device reservoirs. These results demonstrate that jointly optimizing device-level stability and architecture-level reservoir diversity enables simultaneous enhancement of state reliability and representational capacity, providing an effective route toward high-performance spintronic physical RC systems.

So far, intense simulation efforts on hole-spin qubits in germanium (Ge) have focused primarily on singly occupied quantum dots. Here, we theoretically investigate three-hole qubits in germanium and demonstrate that their performance can rival that of single-hole qubits in both strained and unstrained systems. In particular, we find that—in the widely used quasi-circular geometry—a three-hole qubit encoding can yield enhancements of the Rabi frequencies by up to 2 orders of magnitude and a large advantage in terms of quality factors.

We characterize single-mode vacuum squeezing generated by a SNAIL parametric amplifier (SPA) operated under conditions representative of practical sensing and qubit-readout experiments. Motivated by prior expectations that Kerr-induced distortion limits squeezing in degenerate parametric amplifiers, we studied squeezing as a function of external flux and pump power to explore operating points where Kerr nonlinearity is theoretically minimized. We find that for squeezing operation at a fixed frequency, Kerr varied by about a factor of 2 and the achievable squeezing showed no significant dependence on Kerr. Rather than Kerr, the squeezing is dominated by internal resonator loss and insertion loss in the microwave chain as confirmed by theoretical modeling. These results indicate that, in practical SPAs, reducing loss, rather than suppressing Kerr, is the primary route to improved squeezing performance.

We propose and characterize a protocol for rapid near-ground state cooling of the center-of-mass (c.m.) mode of a large trapped-ion crystal. When the initial mean thermal occupation of the mode ${\overline{n}}_i$ is small compared to the number of ions $N$ (${\overline{n}}_i\ll N$), a red-sideband drive implements a beam-splitter-type state exchange (swap operation) between the mode and the collective spin of the $N$ ions, with the latter effectively serving as a quantum harmonic oscillator. Subsequently, a reset of the spins removes the entropy, leading to near-ground state cooling of the c.m. mode. We term this protocol as quantum beam-splitter cooling (QBSC). We analyze the impact of several practical imperfections on the final temperature achievable under QBSC, including finite ion number, off-resonant carrier and blue-sideband contributions, the sideband drives arising from spectator modes, and spatially inhomogeneous driving across the crystal. We outline practical strategies to eliminate the carrier drive. We extend our analysis to multiple cycles of QBSC and also generalize the analysis to demonstrate the cooling of non-c.m. modes using this protocol. Furthermore, we show that measuring the population statistics of the ions at the end of the swap operation can enable near-optimal quantum beam-splitter thermometry (QBST), with the classical Fisher information approaching the quantum Fisher information of a thermal state. We discuss the connection of QBSC with continuous sideband cooling and compare QBST with a recently proposed rapid adiabatic passage-based thermometry scheme. Our work constitutes an example of harnessing many-body effects for laser cooling and thermometry in large trapped-ion crystals.

This paper proposes a reflective anisotropic coding metasurface for multifunctional terahertz wavefront control. By integrating vanadium dioxide (${\mathrm{VO}}_2$), the design realizes multistate precise beam manipulation within a shared aperture. The $27\times 27$ unit cell array provides distinct reflection phase responses for two orthogonal linear polarizations. In the metallic state, it generates two and four reflected beams under $x$- and $y$-polarized incidence at 1.2 THz, respectively. Conversely, as ${\mathrm{VO}}_2$ switches to the insulating state, the metasurface generates zero-order Bessel beams at 1.6 THz. The $x$-polarized channel produces an on-axis beam, whereas the $y$-polarized channel produces a beam with a designed deflection angle of $25°$. Combining radial and linear phase gradients enables control of Bessel beam formation and deflection. Full-wave simulations show a well-defined central lobe and quasinondiffracting propagation over the designed range. The proposed design synergizes ${\mathrm{VO}}_2$ reconfigurability with polarization multiplexing, thereby enabling diverse beam manipulation functions on a single metasurface platform.

This work employs first-principles calculations to systematically investigate the anisotropic lattice thermal transport and thermoelectric properties of layered $X\text{CuSe}$ ($X=\mathrm{K},\text{Rb}$). The results reveal that, along the in-plane direction, low-frequency rattling vibrations of Rb atoms in RbCuSe strongly suppress the lattice thermal conductivity (${\kappa }_L$), resulting in substantially lower values compared with KCuSe. In contrast, the out-of-plane ${\kappa }_L$ exhibits an anomalous mass dependence at low temperatures: despite the larger atomic mass of Rb, RbCuSe shows a higher ${\kappa }_L$ than KCuSe. This anomalous behavior primarily originates from differences in phonon group velocities induced by strongly anharmonic Se vibrations under weak interlayer bonding. With increasing temperature, the marked expansion of the four-phonon scattering phase space in RbCuSe drives a transition in the dominant heat transport mechanism from phonon group velocity-controlled to higher-order anharmonic scattering dominated, thereby gradually suppressing the anomalous mass dependence. Moreover, optical phonon modes contribute more than 50% to the in-plane ${\kappa }_L$, highlighting the crucial role of nonacoustic phonons in heat transport within this layered system. Electronic structure analysis further indicates that the synergistic combination of strong band dispersion and high band degeneracy is favorable for achieving a large power factor. Within the present theoretical framework, p-type doping at 700 K is predicted to yield peak dimensionless thermoelectric figures of merit ($\mathrm{ZT}$) of 2.77 and 4.16 for KCuSe and RbCuSe, respectively, underscoring the promising thermoelectric potential of the XCuSe system in the mid-to-high temperature range.

Although conventional organic fluorescent 4-(Dicyanomethylene)-2-tert-butyl-6-(1,1,7,7-tetramethyljulolidin-4-yl-vinyl)-4H-pyran) (DCJTB) molecules have been widely employed for designing high-efficiency red or white organic light-emitting diodes (OLEDs), the reasons for their high performances are still vague because the existence of hot excitons and their high-level reverse intersystem crossing (i.e., $T_{2,\mathrm{DCJTB}}\to S_{1,\mathrm{DCJTB}}$) in DCJTB has not been previously recognized. Herein, a maximum external quantum efficiency (EQE) of 14.4% is achieved by co-doping the CzDBA sensitizer and the DCJTB emitter into the CBP host. Notably, this host-sensitizer-emitter-based OLED owns an excellent property of negligible efficiency roll-off; that is, the EQEs separately exhibit 14.0% and 13.6% at $500$ and $1000\text{cd}/{\mathrm{m}}^2$ as compared to the maximum EQE of 14.4%. These significant achievements are primarily attributed to the synergistic utilization of enhanced singlet- and hot-exciton energy through multiple exciton energy transfer channels enabled by inserting the CzDBA sensitizer with the thermally activated delayed fluorescence (TADF) property and appropriate charge-transfer state energy. These efficient energy transfer channels were demonstrated through spectroscopic analyses of the film materials, fingerprint magneto-electroluminescence (MEL) studies of the control devices, and analysis of the transient electroluminescence (TEL) response. Moreover, these host-sensitizer-emitter-based devices exhibit abundant variations in optical, electrical, and magnetic properties in the presence and absence of CzDBA TADF sensitizer, which also directly verify the crucial role of an appropriate sensitizer in fully harvesting hot excitons of DCJTB emitters. Thus, our work not only deepens the physical understanding of recently discovered hot-exciton channels from typical organic fluorescent semiconductor materials but also paves a promising pathway to fabricate further high-performance DCJTB-based red or white fully fluorescent OLEDs for lighting and flat-panel display applications.

Generating and preserving metrologically useful quantum states is a central challenge in quantum-enhanced metrology. In low-field atomic magnetometry with multilevel atoms, the nonlinear Zeeman (NLZ) effect is both a resource and a limitation. It can generate internal spin squeezing within a single atomic qudit, but under fixed readout, it also rotates and distorts the measurement-relevant quadrature, limiting the usable metrological gain. The problem is further complicated by the time dependence of both the squeezing axis and the nonlinear evolution itself. Here we show that reinforcement learning can transform NLZ dynamics from a source of readout degradation into a sustained metrological resource. Using only experimentally accessible low-order spin moments, a trained agent identifies a unified control policy for this class of intrinsically nonlinear sensing dynamics. We illustrate the approach in the $f=21/2$ manifold of ${}^{161}\mathrm{Dy}$, where the learned policy rapidly prepares strongly squeezed internal states and stabilizes more than $4\mathrm{dB}$ of fixed-axis spin squeezing under continuous NLZ evolution. Including state-preparation overhead, the learned protocol yields a single-atom magnetic-field sensitivity of $13.9\mathrm{pT}/\sqrt{\mathrm{Hz}}$, approximately $3\mathrm{dB}$ beyond the standard quantum limit. Our results establish learning-based control as an experimentally feasible route for converting unavoidable intrinsic nonlinear dynamics in multilevel atomic sensors into operational metrological advantage.

Antiartificial spin ice (AASI), realized through spin-ice-derived antidot lattices of stadium-shaped holes in continuous Permalloy (${\text{Ni}}_{81}{\text{Fe}}_{19}$) films, provides a powerful route for embedding geometry-induced magnetic anisotropy within a connected magnonic medium. Here, homogeneous and bicomponent AASI lattices are investigated through a combined experimental and micromagnetic approach, establishing a direct link between lattice geometry, magnetization reversal, and spin-wave (SW) dynamics. Magneto-optical Kerr effect measurements, Brillouin light scattering spectroscopy, and simulations reveal that antidot size, lattice heterogeneity, and magnetic field orientation collectively govern both equilibrium states and dynamic excitations. The static response exhibits pronounced angular anisotropy in coercivity and remanence, with bicomponent lattices showing distinct reversal pathways and shifted anisotropy extrema relative to homogeneous systems. These variations are directly reflected in the SW spectra, where edge-, node-, and domain-wall-localized modes, as well as mode splitting and frequency flattening, emerge as signatures of geometry-induced magnetization textures. The excellent agreement between experiment and simulation establishes a microscopic connection between equilibrium magnetic configurations and SW excitations. These findings demonstrate a direct transfer of configurational anisotropy into tunable magnonic functionality, highlighting AASI as a versatile platform for geometry-encoded SW architectures.

We demonstrate efficient diffraction of intense ultrashort laser pulses using optical-field-ionization-induced plasma-neutral gratings formed by spatially structured ionization of a neutral molecular gas in the interference field of two femtosecond pump pulses. The transient refractive index modulation of the plasma structure persists for at least 10 ps and is used to diffract intense femtosecond signal pulses into the first order of diffraction with an average efficiency of up to 35%. Plasma gratings are shown to provide stable diffraction at signal laser intensities greater than ${10}^{14}\mathrm{W}/{\mathrm{cm}}^2$, exceeding the damage thresholds of conventional solid-state optics by more than two orders of magnitude, continuously for at least one hour at a 10-Hz repetition rate. The experimental diffraction efficiency scales with the grating aperture allowing for a larger millimeter-scale plasma optic, increases with the pump energy and electron density, and reaches a maximum at a specific grating length in agreement with the coupled-mode theory for periodic media. These results demonstrate the scalability, tunability, and high damage threshold of transmissive plasma-based photonic structures, opening new prospects for controlling multi-petawatt laser beams.

Increasing connectivity and decreasing qubit-state delocalization without compromising the speed and accuracy of elementary gate operations are topical challenges in the development of large-scale superconducting quantum computers. In this theoretical work, we study a special honeycomb qubit lattice where each qubit inside a unit cell is coupled to every other one via two dedicated tunable couplers and a common central element. This results in an effective multimode interaction enabling tunable, on-demand, all-to-all connectivity between each qubit pair within the unit cell. We provide a thorough analysis of the unit cell, including a proposal for a novel and efficient conditional-$Z$ gate scheme, which takes advantage of the effective multimode coupling. We develop an experimentally viable pulse protocol for a single-step gate implementation, which considerably improves the gate speed compared to the previous two-qubit-gate realizations suggested for architectures utilizing a center mode. We also show numerical results on how the presence of spectator qubits affects the average two-qubit-gate fidelity and analyze how the multimode coupling structure mitigates the delocalization-induced crosstalk during simultaneous single-qubit gates within the unit cell. We also provide analytical estimates for the errors caused by relaxation and dephasing during a two-qubit-gate operation, including noise terms for the multimode coupling structure. Our multimode coupling architecture results in a good balance between increased connectivity and available parallelism, especially when several interacting unit cells form a quantum-processing unit. We anticipate that the obtained results pave the way toward high-connectivity quantum processors with efficient and low-overhead quantum algorithms.

In a circuit quantum electrodynamic architecture, we experimentally demonstrate a hardware-efficient and qubit-state-dependent single-step phase-engineered (SSPE) pulse scheme for actively depopulating a readout cavity. The protocol appends a reset segment with tailored amplitude and phase to a square readout pulse. Within the linear-response regime, the optimal reset amplitude scales proportionally with the readout amplitude, while the optimal reset phase remains invariant, significantly simplifying the experimental calibration procedure. For a known qubit state, time-resolved measurements demonstrate that the SSPE pulse achieves faster cavity depletion than the CLEAR protocol. Crucially, this approach enables arbitrarily fast, overshoot-free depletion of the cavity photon population, with the ultimate reset rate constrained by the available drive amplitude and the finite analog bandwidth of the measurement chain. A comprehensive evaluation of the quantum nondemolition character reveals no measurable increase in non-QND measurement errors (non-QNDness) relative to the free-decay and CLEAR protocols. The remaining non-QNDness is predominantly attributable to state switching induced by qubit relaxation during the readout process. These results establish the SSPE scheme as a practical and scalable approach for achieving rapid and smooth cavity reset in superconducting quantum circuits in the case of a known qubit state.

Single-layer diffractive neural networks often face classification accuracy bottlenecks due to limited wavefront modulation capabilities. Edge detection, as an optical image processing technique, extracts image contours and offers a promising way to simplify classification tasks. However, integrating edge detection and diffractive neural network (DNN)-based classification on a single chip remains a challenge. Here, we propose an integrated nonlocal meta-platform that sequentially achieves all-optical edge detection and classification via spin-multiplexing. By exploiting the dispersion properties of the nonlocal Huygens’ metasurface, the co-polarized component in the output light performs momentum-space filtering for real-time edge detection. Subsequently, the cross-polarized component undergoes geometric phase modulation to execute image classification within the DNN. We couple quasibound states in the continuum and magnetic dipole resonances in crescent-shaped nanopillars, achieving a high polarization conversion efficiency of approximately 55%. This edge-enhanced DNN architecture significantly reduces data redundancy, elevating the classification accuracy of the single-layer network on the MNIST dataset from 64.2% to 80.7%. Our work provides a compact, high-efficiency solution for integrated all-optical machine vision and intelligent photonic computing.

We report recent progress in the growth and fabrication of advanced PbTe-Pb Majorana devices for three-terminal correlation experiments. A critical prerequisite for these devices is a continuous Pb film spanning from the PbTe nanowire to the substrate. In previous studies, PbTe nanowires were spatially separated from the substrate mask, making them unable to fulfill this requirement. To address this issue, we have developed a growth and fabrication approach. Key fabrication steps are introduced between the PbTe growth and Pb deposition. The top facet of the PbTe nanowire is covered with a Te capping layer before breaking the ultrahigh vacuum, enabling ex situ processing for subsequent steps. The mask is then removed, followed by the fabrication of shadow walls. After reloading the chip into the growth chamber, the Te protection layer is removed through thermal decomposition, and the Pb film is subsequently deposited. Notably, a pristine PbTe-Pb interface with an in situ-equivalent quality can be achieved despite the ex situ processing. Our technique facilitates the development of advanced PbTe-Pb quantum devices with complex designs, paving the way toward the realization of topological qubits.

We investigate how nonstabilizer resources enable the emergence of quantum state designs within the projected ensemble. Starting from initial states with finite magic and applying resource-free Clifford circuits to scramble them, we analyze the ensemble generated by performing projective Pauli measurements on a subsystem of the final state. Using both analytical arguments and large-scale numerics, we show that the projected ensemble converges toward a state $k$-design with an error that decays exponentially with the $k$th stabilizer Rényi entropy of the premeasurement state, via a magic-induced design ansatz (MIDA) that we introduce. We identify a scaling form that depends on the initial state only through its magic $M_k$, which we corroborate for distinct classes of initial states through numerical simulations and analytical calculations of the frame potential. For finite-depth Clifford unitaries, we show that the timescales at which state designs emerge are controlled by the transport of magic. We identify a “magic teleportation” mechanism whereby non-Clifford resources injected locally spread through Clifford scrambling and measurements across distances beyond the light-cone. Our results demonstrate that an amount of magic scaling with the probed subsystem size, yet independent of the total system size, suffices to generate highly random states, providing a systematic route toward generating quantum state designs in early fault-tolerant devices.

With its pronounced photon-fluctuation characteristics, superthermal light provides a fundamental physical basis for overcoming the long-standing bottlenecks of sampling efficiency and image quality in conventional optical imaging. Building on previous demonstrations of superthermal ghost imaging, this work integrates a superthermal light source generated by secondary speckle modulation with compressive-sensing ghost imaging reconstruction into a unified imaging framework. Through two independent speckle modulations, a superthermal light field with a $g^{(2)}(0)$ value enhanced from the thermal level of approximately 2 to nearly 4 is successfully generated, and the derived expression of $g^{(2)}(0)$ depends on the effective mode-number parameter and coherence length of the optical field. Under the sampling conditions studied here, this integrated approach improves the target-background separation and the quality of compressive-sensing reconstruction, demonstrating the application potential of engineered superthermal statistics in advanced optical imaging.

Signal readout in optically pumped magnetometers (OPMs) is usually described in terms of ground-state spin polarization and its Larmor response, while excited-state fine-structure transfer (FST) is absorbed into effective relaxation terms. Here, we treat collision-mediated FST as a fluorescence readout channel under OPM operating conditions. Using steady-state fluorescence spectroscopy, low-frequency magnetic-field modulation, and temperature-dependent Rb-He measurements, we observe collision-generated D2 fluorescence under single-wavelength D1 pumping without direct D2 excitation. The D1 signal is dominated by the modulation frequency ${\mathrm{\Omega }}_m$, whereas the D2 signal contains a pronounced $2{\mathrm{\Omega }}_m$ component whose amplitude scales quadratically with the modulation-field amplitude. Unpolarized D2 fluorescence from the $P_{3/2}$ level carries a rank-2 alignment term that D1 fluorescence from the $P_{1/2}$ level lacks. This selection rule, the $2{\mathrm{\Omega }}_m$ dominance, and the quadratic scaling together identify the D2 channel as a rank-2-weighted readout opened by FST. From the temperature dependence of the $\mathrm{D}2/\mathrm{D}1$ fluorescence ratio, we obtain an effective Rb-He FST cross section ${\sigma }_{\mathrm{eff}}=(2.7\pm 0.1)\times {10}^{-17}{\mathrm{cm}}^2$ under an effective symmetric-mixing approximation, consistent with reported Rb-He FST values. The same collision conditions occur in spin-exchange-relaxation-free (SERF) magnetometers, NMR gyroscopes, and spin-exchange optical-pumping cells, where FST affects fluorescence readout and optical-pumping models.

We present an explicit quantum circuit construction for Hamiltonian simulation of a first-order velocity-stress formulation of the three-dimensional elastic wave equation in homogeneous isotropic media. Previous studies have shown how elastic wave equations can be cast into forms amenable to Hamiltonian simulation, but they typically rely on black box Hamiltonian access assumptions, making gate complexity estimation difficult. Starting from the first-order velocity-stress formulation, we discretize the system by finite differences, transform it into Schrödinger form, and exploit the separation between the component register and the spatial register to decompose the Hamiltonian into structured tensor product terms. This yields explicit implementations of first-order and second-order Trotter formulas for the resulting time evolution operator. We derive the corresponding error bounds and qubit and cnot complexity estimates with explicit constants in terms of the discretization parameter, simulation time, target accuracy, and material parameters, exhibiting polynomially better upper bounds in time complexity compared with those of classical finite difference methods. Numerical experiments compare the analytical cnot gate-count bounds with the compiled cnot gate counts and validate the proposed framework against the exact evolution.