Papers

We present two methods to estimate bottom topography in a shallow water flow using only surface deformation measurements. One is based on Physics-Informed Neural Networks (PINNs) and the other on the Adjoint State Method. We test both methods using synthetic data in 1D and 2D cases. Both are able to successfully reconstruct not only the bottom topography but also the surface velocity. Both also show robustness against noise and data sparsity up to reasonable levels.

Based on experimental results, this work proposes the influence of the Oxygen order present in the Ortho-II superstructure of YBa$_2$Cu$_3$O$_{7-\delta}$ (YBCO), on the final ordering of Oxygens in its Orthorhombic phase for $\delta$ $\approx$ 0. Isothermal oxygenation (oxyg) of YBCO powder material is performed, starting from non-oxygenated material ($\delta$ $=$ 1) and evolving until saturation in an oxygen atmosphere. The oxyg process is carried out within a temperature range from 300 $^o$C to 800 $^o$C (300 $^o$C $<$ T$_O$ $<$ 800 $^o$C). During the oxyg process, and using a thermogravimetric balance, the evolution of mass (m) and the differential thermal analysis (DTA) of the material are monitored with respect to an inert reference material subjected to the same conditions as the YBCO powder. These results allow observation of the Tetragonal-Orthorhombic (T-O) transition occurring in the YBCO material. From these results, oxygenated YBCO material is obtained by working at different temperatures and under two different conditions: through a direct T-O transition into the Ortho-I superstructure, and by passing through the Ortho-II superstructure along the transition. The material obtained under these two conditions is studied by X-Ray diffraction, revealing differences in the resulting diffractograms. Furthermore, we propose that, for low values of T$_O$ (T$_O$ $<$ 400 $^o$C), the T-O transition proceeds through the region of the phase diagram where the Ortho-II superstructure is present, leading to progressive ordering of the Oxygen atoms within the material. This ordering leaves a fingerprint in the final configuration reached by the YBCO material, even beyond the region where the Ortho-II superstructure is stable. Finally, we suggest that this mechanism is responsible for the differences observed between the diffractograms obtained under both conditions.

The space-time correlation of a passive scalar advected by a Gaussian colored-noise velocity with wavenumber-dependent correlation times and power-law spatial spectra is investigated in the present paper. Within the inertial-convective subrange, we derive an analytical solution for the space-time correlation. This solution validates the elliptic approximation (EA) model [He and Zhang, Phys. Rev. E 73, 055303(R) (2006)], demonstrating that the iso-correlation contours are self-similar in the co-moving space-time frame $(r-U\tau, V\tau)$, with a universal spatial-to-temporal intercept ratio of 1.55. Unlike the classic Kraichnan white-noise model, our formulation simultaneously recovers the Obukhov–Corrsin scaling for spatial correlations (when the velocity obeys Kolmogorov scaling) and reproduces the random-sweeping mechanism, yielding Gaussian (rather than exponential) temporal decorrelation of scalar Fourier modes. Our results clarify the underlying decorrelation mechanism of passive scalars: mean-flow advection and large-scale sweeping dominate temporal decorrelation, and small-scale distortion dominates spatial decorrelation.

The $\vartheta$-MGGA class of density functionals is formally reformulated as Hessian-level meta-generalized gradient approximations (HL-MGGAs). In contrast to standard meta-GGAs that rely on the orbital-dependent kinetic-energy density or the density Laplacian, HL-MGGAs utilize the full density Hessian. We introduce a simplified, non-empirical functional, $\vartheta$-PBE, and present a roadmap for its self-consistent implementation within the projector augmented-wave (PAW) method. By utilizing the complete set of spatial second-order density derivatives, the functional's underlying descriptor successfully distinguishes between distinct one-electron density limits, such as single-center atomic densities and two-center bonds, that standard iso-orbital indicators often conflate. Benchmarks across molecular and solid-state datasets reveal that while $\vartheta$-PBE delivers accurate chemisorption energies and molecular properties, challenges remain in predicting bulk lattice constants. Ultimately, this work demonstrates the physical utility and feasibility of designing orbital-independent, Hessian-based exchange-correlation functionals.

Bicycle node networks are regional bicycle networks equipped with a wayfinding system of numbered nodes to ease recreational cycling. They spur sustainable bicycle tourism, economic spending, and local culture. Due to their country-wide scale, implementing bicycle node networks is a considerable effort and investment. Despite this investment, planning is a manual ad-hoc process that follows general design principles, but without clear performance metrics that account for the human cycling experience. Here we analyze a 28,215 km long bicycle node network spanning Denmark, developing and studying such metrics. First, a spatial analysis of geometric and topological properties reveals high heterogeneity and local clusters of node density, face loop lengths, gradients, and feature-rich areas. Next, taking the perspective of a recreational cyclist starting at any node on the network, we create a loop census that lists all loops in the network up to day-trip length. The loop census identifies the feasible points on the network from which to take a day trip and quantifies the number of round trip choices, unveiling different levels of choice depending on the considered demographic group. While long-range cyclists can access most of the country with often overabundant choices, cyclists with stronger length and gradient limitations like families with small children can not - which could be overcome by e-bikes. Our open-source analysis methods provide data-driven decision support for bicycle node network planning with the potential to boost the development of rural cycling and cycling tourism.

We take causality and uniqueness of events observation as our driving forces. They are built in in the way we define distinct observers, which then require a finite time to communicate between each other. This unavoidably leads to the existence of maximal transfer-information velocity between arbitrary (not necessarily inertial) reference frames. Inertial reference frames are defined by fixing the geometrical properties of (spatial) distance without any reference to relativity, electromagnetism, or laws of physics in general. For these inertial reference frames, the causality condition fixes the causal group to be the orthochronous inhomogeneous Lorentz group times dilatations. The mathematics we will use are quite basic.

We design and experimentally characterize a balanced homodyne detector optimized for high-repetition-rate (100 MHz) pulsed optical sources. Unlike conventional transimpedance-amplifier architectures, which suffer from nonlinearities and dynamic instabilities with ultrashort pulses, our approach allows to directly amplify the photocurrent extracted at the common photodiode node without feedback loops. A theoretical model describing the detector response, noise, and pulse-to-pulse correlations is developed, providing quantitative predictions for the signal variance, signal-to-noise ratio (SNR), and inter-pulse correlations. Implemented with two matched InGaAs photodiodes illuminated by a 1030 nm mode-locked laser at 100 MHz, the detector exhibits excellent linearity and shot-noise-limited scaling of the signal variance with optical power. Optimizing the temporal integration window yields a maximum SNR of about 14 dB, while correlation measurements confirm negligible inter-pulse correlations. These results demonstrate that the proposed architecture offers a robust and simple solution for high-speed pulsed homodyne detection, suitable for quantum optics and continuous-variable quantum information applications.

Cost-effective wireless electrocardiograms (ECGs) enable long-term and scalable monitoring of cardiac patients in their home and work environments. Because they offer greater freedom of movement, they are also suitable for investigating the relationship between cardiac workload and underlying physical exertion. However, this requires that the quality of the generated data meets the standards of clinical devices. The aim of this study is to examine this closely. We therefore analyze data from 54 healthy subjects who performed five physical activities using wireless ECGs outside of clinical settings and without medical supervision. The results are compared with clinically collected data from standard 12-lead ECGs (2493 subjects) and Holter ECGs (29 subjects), with particular attention to the RR interval time series (tachogram) and heart rate variability (HRV). Our study shows significant statistical agreement between the different datasets. We calculated the 95% confidence intervals for the mean RR interval and HRV assuming that (1) the statistics of the 12-lead ECGs could serve as reliable reference, and (2) the statistics of the 12-lead ECGs cannot be taken as reliable reference. The p-values for both conditions (for the RR interval: 0.23 and 0.26 respectively; for HRV: 0.10 and 0.11 respectively) suggest that there is insufficient evidence to reject the hypothesis that significant statistical agreement exists between the different datasets.

The soliton-like coherent structure (SCS), which has been verified to exist in both transitional and turbulent boundary layers1-4, still poses a challenge in the understanding of its formation and behavior. In our previous study (Niu et al.5), the SCS was also found to exist in the transitional wake flow behind a sphere. In present study, the formation and evolution of the SCS is further investigated at four Reynolds numbers by numerical simulation. The results show that at the early stage of the turbulence transition, the SCS appears as a form of wave packet during the Tollmien-Schlichting (T-S) wave stage. With the increase of the Reynolds number, the SCS reaches its maximum amplitude downstream where the velocity discontinuity occurs. This position is located after the breakdown of the T-S wave and the three-dimensional structure is formed. Then, the SCS conserves its shape and amplitude over a long distance downstream. The relationships among the SCS, the spikes, the vortex structures, and the high-shear layers are further analyzed. It is found that the SCS in the wake flow has similarities to the phenomena observed in boundary layer flows during the turbulent transition. The vortex structures and high-shear layers mostly wrap around the border of the SCS. The vortex structure is considered to be as a consequence of the development of the SCS rather than its cause.

We investigate the sliding strength of thin filaments in frictional contact with a translating cylinder, perpendicular to the filaments' axes, in knotted (clove hitch) and unknotted (capstan) configurations. Recent work reported superlinear scaling for surgical knots with elasto-plastic filaments [1]. Testing the clove hitch with various materials (elastomeric rods, metallic wires, braided ropes) reveals similar nonlinear behavior, ruling out plasticity. To explore the source of the previously reported nonlinear behavior, we perform three-dimensional FEM simulations (resolving full 3D mechanics) and reduced-order DER simulations (isolating geometric effects by neglecting cross-sectional deformation). Both FEM and DER simulations reproduce the experimental scaling. Simplifying the knot topology by studying capstan angles from $\pi/4$ to $4\pi$ yields comparable superlinear behavior, transitioning to linearity at smaller angles. We rationalize the results by developing an analytical model based on planar elastica theory for the capstan configuration (which exhibits behavior similar to the clove hitch but with a simpler topology). The model reproduces the observed superlinear behavior and rationalizes it by coupling the evolution of normal forces and contact arclength during tightening. The analysis further predicts transition to linearity when full contact between the filament and the cylinder is established, providing a mechanical framework applicable across materials, geometries, and topologies.

A widely tunable free spectral range (FSR) is essential for many microresonator applications, but achieving it in compact high-Q systems remains a significant challenge. Surface nanoscale axial photonics (SNAP) provides a route to such devices by localizing whispering-gallery modes through nanoscale effective-radius variations along an optical fiber. Recently, it has been experimentally demonstrated that side-coupling between two optical fibers can induce high-Q SNAP microresonators. In contrast to broadly explored monolithic microresonators, this configuration enables extensive mechanical tuning of the microresonator configuration and FSR through fiber bending, tilting, and twisting. Here, we develop the theory of SNAP microresonators induced at optical fiber intersections, which has remained largely unexplored. We consider weakly twisted fibers, whose geometry can be decomposed into tilting and bending. We show that extremely small fiber curvature critically affects the coupling-induced cutoff frequency variation, the shape of the effective axial potential, and the spectrum of the induced resonator. We discuss the physical origin of this curvature and show that accounting for it resolves the discrepancy between theory and experiment. These findings establish side-coupled optical fibers as a simple, reconfigurable platform for realizing microresonators with a widely tunable FSR.

Cavities with high quality (Q) factor and small mode-volume are crucial to realize high-performance nanolasers suitable for optical interconnects. In this work, we propose a novel one-dimensional photonic crystal nanobeam cavity design with fins for controlled electrical injection into the active region. An effective optimization algorithm based on first-order perturbation theory of quasinormal modes is implemented and shown to significantly enhance the cavity quality factor. The one-dimensional geometry of the cavity lends itself to unidirectional coupling of the resonant mode into the waveguide by introducing asymmetry of the mirror. The resulting design is shown to achieve high extraction efficiencies ($>90\%$) while maintaining a high Q-factor ($>10 \cdot 10^3$). Through an analysis of the cavity's decay channels, we find that the introduced asymmetry induces unexpected interactions between the cavity's decay channels. Passive InP cavities are fabricated and experimentally characterized, demonstrating record-high quality factors exceeding $170 \cdot 10^3$ for designs without fins and up to $70 \cdot 10^3$ for designs with fins, confirming the efficacy of the optimization method and quality of the fabrication process.

This paper explores the electrification of mezcal distilling in Oaxaca, Mexico, as a sustainable alternative to traditional firewood methods. We investigate the mezcal process, including cooking, grinding, fermentation, and distillation, and propose a photovoltaic system for distillation. The research also includes scientific outreach activities in the producing communities. We, in collaboration with the communities, proposed novel uses of renewable energies. The results of chemical analysis (chromatography and FTIR) and sensory data for distillation using firewood and electricity are presented to compare the mezcal produced with solar energy and traditional mezcal. Our studies conclude that electrical distillation can reduce environmental impact and improve energy efficiency without compromising product quality.

Wavelet phase is a critical parameter in seismic processing, where zero-phase wavelets are essential for maximizing temporal resolution and ensuring accurate interpretation of subsurface structures. In practice, however, the seismic wavelet is often nonstationary, exhibiting a phase that varies in space and time due to physical factors such as attenuation, dispersion, and thin-bed tuning effects. Higher-order statistical measures-specifically kurtosis and skewness-are traditionally maximized to drive the signal toward a maximally non-Gaussian or maximally asymmetric zero-phase state. This paper addresses the computational and stability challenges inherent in nonstationary estimation by casting the problem as a regularized non-convex optimization task. We propose a robust framework based on the Alternating Direction Method of Multipliers (ADMM) that eliminates the instability and artifacts associated with traditional piecewise-stationary windowed approaches. The core of our contribution is the derivation of the first closed-form proximity operators for the scale-invariant inverse kurtosis and inverse skewness functionals. By exploiting the signed permutation invariance of these statistical measures, we reduce the high-dimensional proximal subproblems to efficient one-dimensional root-finding tasks. We provide a detailed geometric interpretation of the optimality conditions, demonstrating that the global minimizer is governed by a branch-separation property. Furthermore, we derive an explicit critical threshold parameter which provides a theoretical rule for identifying the global minimum among multiple stationary points. Numerical validations on synthetic and real seismic data demonstrate that the proposed proximal algorithms achieve linear computational complexity and superior stability compared to traditional methods, effectively enabling nonstationary phase correction.

Retrieving suspension parameters from light scattered by a suspension microdroplet is a difficult inverse problem because the recorded optical pattern is shaped by droplet morphology and by perturbations of the internal optical field introduced by the dispersed phase. Here we demonstrate experimental, data-driven recognition of microdroplet and nanoparticle-suspension parameters from laser speckle images recorded for individual levitated suspension microdroplets. We confined slowly evaporating TiO$_2$ nanoparticle suspensions in diethylene glycol in a linear electrodynamic quadrupole trap and classified their speckle patterns using a convolutional neural network (CNN). The CNN was trained and tested to recognize droplet diameter, nanoparticle mass concentration, and nanoparticle diameter, first separately and then in simultaneous two- and three-parameter tasks. The CNN recognized droplet diameter most robustly, with an estimated accuracy of approximately 6% on independent test data. Nanoparticle concentration required sufficiently separated classes. Nanoparticle-diameter classes were also recognized, although particle-size effects were coupled to changes in particle number at fixed mass concentration. We demonstrated simultaneous three-parameter classification across 27 classes. As an exploratory out-of-distribution test, we applied a CNN trained on single-component suspensions to dual-component suspensions containing two nanoparticle diameters. The CNN retained meaningful predictive capability, especially at higher concentrations, although robust decomposition of mixed nanoparticle populations was not achieved. These results show that speckle images of single levitated suspension microdroplets contain sufficient information for CNN-based multi-parameter recognition and support further development of speckle-based optical diagnostics for complex microdroplet and aerosol-like systems.

This work attempts to present a complete theory of the $\mu$H hyperfine splitting, including all contributions above 1 ppm. Quantum electrodynamic and recoil corrections are calculated directly, while the proton structure correction is obtained with the help of the H hyperfine splitting. The resulting theoretical prediction for the ground state of $\mu$H is $E_\mathrm{hfs} = 182\,626(5)$ $\mu$eV.

We implemented ab initio Hubbard parameter calculation schemes in the k-point sampling real-time TDDFT (RT-TDDFT) program in CP2K. We propose a new linear-response-based calculation scheme for energy-dependent Hubbard parameters. Our scheme extends the minimum-tracking linear-response method proposed in [Moynihan et al., arXiv preprint arXiv:1704.08076(2017); E. B. Linscott et al., Phys. Rev. B 98, 235157 (2018)] to realize the calculation of energy-dependent Hubbard parameters that reflect the exchange-correlation (xc) effects included in the xc-functional. We discuss the properties of the minimum-tracking linear-response method in comparison to another promising scheme, ACBN0 [Agapito et al., Phys. Rev. X, 5, 011006 (2015)]. We show that, while neither clearly outperforms the other in the accuracy of static property calculations, each has a distinct dynamical application depending on its theoretical formulation.

Optically levitated micro- and nanoparticles are an ideal optomechanical platform for precision measurements, particularly enabling the detection of ultraweak forces. Nevertheless, quantum backaction and inherent instabilities induced by the trapping laser fundamentally restrict further improvements in force sensitivity and resolution. To circumvent these bottlenecks, we actively drive the levitated nanoparticle's mechanical motion in a phase-locked phonon laser mode and integrate a carrier-modulation measurement architecture to enhance force sensing capabilities. The stable and high-amplitude oscillation of the phonon laser allows for the robust trapping under 1 mW-level laser power, which in turn reduces the force noise to 4.0(3)*10^-22 N/Hz^1/2. Furthermore, by using phase-locked phonon laser, the measurement system achieves active stabilization and extended coherence time with the measured signal to 12,500 seconds, realizing a measurement resolution of 8(4)*10^-24 N with a sensitivity of 9.3(7)*10^-22 N/Hz^1/2 under a loaded force. These results establish the phonon laser as a low-noise, long-coherence-time, self-stabilizing platform for precision measurements, as well as in quantum and fundamental physics tests.

Steady-state RANS simulations are reported for 2D flow around an ellipse and a von-Kármán-Trefftz airfoil at seven different angles of attack and two different Reynolds numbers, computed using the $k \omega SST$ turbulence model in OpenFOAM. The dataset contains surface pressure distribution, skin friction distribution, lift and drag coefficients, stagnation point location and separation point locations. The results serve as a benchmark for calibration and evaluation of extended potential flow models.

Epithelial tissues dynamically reshape through local mechanical interactions among cells, a process well captured by vertex models. Yet their many tunable parameters make inference and optimization challenging, motivating computational frameworks that flexibly model and learn tissue mechanics. We introduce VertAX, a differentiable JAX-based framework for vertex-modeling of confluent epithelia. VertAX provides automatic differentiation, GPU acceleration, and end-to-end bilevel optimization for forward simulation, parameter inference, and inverse mechanical design. Users can define arbitrary energy and cost functions in pure Python, enabling seamless integration with machine-learning pipelines. We demonstrate VertAX on three representative tasks: (i) forward modeling of tissue morphogenesis, (ii) mechanical parameter inference, and (iii) inverse design of tissue-scale behaviors. We benchmark three differentiation strategies-automatic differentiation, implicit differentiation, and equilibrium propagation-showing that the latter can approximate gradients using repeated forward, adjoint-free simulations alone, offering a simple route for extending inverse biophysical problems to non-differentiable simulators with limited additional engineering effort.

Neural operators have emerged as powerful surrogates for dynamical systems due to their grid-invariant properties and computational efficiency. However, Fourier-based variants inherently truncate high-frequency components in spectral space, resulting in the loss of small-scale structures and degraded prediction quality at high resolutions when trained on low-resolution data. While diffusion-based enhancement methods can recover multi-scale features, they introduce substantial inference overhead that undermines the efficiency advantage of neural operators. In this work, we introduce MeanFlow-Enhanced Neural Operators (MENO), a novel framework that achieves accurate all-scale predictions with minimal inference cost. By leveraging the improved MeanFlow method, MENO restores both small-scale details and large-scale dynamics with superior physical fidelity and statistical accuracy. We evaluate MENO on three challenging dynamical systems, including phase-field dynamics, 2D Kolmogorov flow, and active matter dynamics, at resolutions up to 256$\times$256. Across all benchmarks, MENO improves the power spectrum density accuracy by up to a factor of 2 compared to baseline neural operators while achieving up to $14\times$ faster inference than the state-of-the-art Denoising Diffusion Implicit Model (DDIM)-enhanced counterparts, effectively bridging the gap between accuracy and efficiency. The flexibility and efficiency of MENO position it as an efficient surrogate model for scientific machine learning applications where both statistical integrity and computational efficiency are paramount.

Secure communications with quantum key distribution over fiber-optic links is one of the few recognized applications of quantum physics at the level of individual quanta – single C-band photons. Currently, the widely used sources of such photons are highly attenuated laser pulses, featured by a low probability of single photon occurrence. Here, we present an efficient source with an InAs/GaAs quantum dot on a metamorphic buffer layer inside a micropillar-shaped microcavity. The key innovation is the use of different semiconductor and dielectric materials to form the lower (GaAs/AlGaAs) and upper (Si/SiO$_2$) Bragg reflectors. Compatibility of these materials in a monolithic source is achieved by depositing a small amount of Si/SiO$_2$ pairs on an incomplete micropillar made from a coherent heterostructure grown by molecular beam epitaxy. This design enables resonant excitation with $\pi$-pulses and generation of polarized photons with a record-breaking end-to-end efficiency of 11%.

Active matter classifies systems consisting of self-propelled units which convert the energy stored locally or extracted from their environment into directed motion. It has recently attracted considerable attention due to rich new physics it displays and potential applications in various fields including materials science. Active matter found in nature is inherently complex, so model systems are of interest where the main relevant features can be isolated and studied in laboratory experiments. An interesting instance of active matter is a suspension of active particles (e.g., the so-called Janus particles, where the two halves have different properties) in a gas discharge plasma. Such complex plasmas with active particles are excellent model systems which can enhance our understanding of natural active matter systems not easily amenable to experiment. In the present experimental study, micrometer-size plastic microspheres with thin gold coating on one side were suspended as a single layer in the plasma sheath of a radio-frequency discharge in argon and driven by a combination of laser-induced photophoretic force and asymmetric ion drag force. Enhanced particle activity in this highly driven, inertial active-matter system leads to collective particle dynamics characterized by extended self-similarity of the velocity field, intermittency, and the emergence of a direct energy cascade with non-universal scaling exponent.

Local polynomial smoothing is a widespread technique in data analysis, and Savitzky-Golay (SG) filters are one of its most well-known realizations. In real settings, the effectiveness of SG filtering depends critically on proper tuning of its parameters, constrained in turn by repeated polynomial fitting over large data windows and for varying polynomial degrees. Standard implementations based on monomial bases and Vandermonde matrix formulations are known to suffer from ill-conditioning and unfavorable scaling as the problem size increases. In this work, we present a fast and numerically stable method for computing polynomial fitting and differentiation matrices by reformulating the problem in terms of discrete orthogonal (Chebyshev) polynomials. Exploiting their recursive structure and the intrinsic symmetry properties of the resulting matrices, we derive two algorithms designed to reduce computational overhead. Both methods significantly reduce memory usage and improve scalability with respect to the polynomial degree and window length. A discussion of the performance demonstrates that the proposed algorithms achieve orders-of-magnitude improvements in numerical accuracy compared to standard matrix multiplication, while also providing potential gains in execution time for large-scale problems. These features make the approach particularly well suited for applications requiring repeated local polynomial fits, such as the optimization of SG filters in high-resolution spectral analyses, including axion dark matter searches and the ALPHA haloscope.

Strongly correlated molecules often contain dense manifolds of low-lying spin states, making total-spin symmetry essential for predictive electronic-structure theory. Neural-network quantum states provide flexible variational wavefunctions, but commonly used fermionic architectures do not enforce this symmetry and can therefore converge to spin-contaminated states with misleading energies and properties. Here we introduce a spin-adapted neural-network backflow (SA-NNBF) ansatz in second quantization, which combines configuration-dependent spatial orbitals with a compressed spin eigenfunction. A projected tensor compression scheme for spin eigenfunctions and a particle-hole representation make variational Monte Carlo calculations with SA-NNBF practical for active spaces containing more than one hundred electrons. Across hydrogen chains and iron-sulfur clusters, SA-NNBF eliminates spin contamination and consistently achieves lower variational energies than standard NNBF with a comparable number of parameters. For the CAS(113e,76o) active-space model of FeMoco, SA-NNBF yields a highly compact spin-adapted variational state, achieving an energy competitive with recent spin-adapted DMRG calculations at bond dimension $D=10000$ while using orders of magnitude fewer parameters. Our work establishes a general framework for developing spin-symmetry-preserving neural-network quantum states for chemically realistic strongly correlated electrons.

Conventional multi-image super-resolution (MISR) methods, such as burst and video SR, rely on sequential frames from a single camera. Consequently, they suffer from complex image degradation and severe occlusion, increasing the difficulty of accurate image restoration. In contrast, multi-aperture camera-array imaging captures spatially distributed views with sampling offsets forming a stable disk-like distribution, which enhances the non-redundancy of observed data. Existing MISR algorithms fail to fully exploit these unique properties. Supervised MISR methods tend to overfit the degradation patterns in training data, and current self-supervised learning (SSL) techniques struggle to recover fine-grained details. To address these issues, this paper thoroughly investigates the strengths, limitations and applicability boundaries of multi-image-to-single-image (Multi-to-Single) and multi-image-to-multi-image (Multi-to-Multi) SSL methods. We propose the Multi-to-Single-Guided Multi-to-Multi SSL framework that combines the advantages of Multi-to-Single and Multi-to-Multi to generate visually appealing and high-fidelity images rich in texture details. The Multi-to-Single-Guided Multi-to-Multi SSL framework provides a new paradigm for integrating deep neural network with classical physics-based variational methods. To enhance the ability of MISR network to recover high-frequency details from aliased artifacts, this paper proposes a novel camera-array SR network called dual Transformer suitable for SSL. Experiments on synthetic and real-world datasets demonstrate the superiority of the proposed method.

We present the first observational test of the hybrid ring strategy, a general coordinated signaling scheme proposed by Seto (2025), which provides a practical Schelling-point realization for interstellar signaling. We use the exceptionally bright GRB 221009A as the anchoring flash for the scheme, together with the accurately measured distance to the Galactic center. This combination provides a high-precision relation linking sky position to a tightly constrained arrival-time window. TESS observed the region around the GRB nearly continuously for ~50 days in 2024, providing survey light curves that enable a direct test of this scheme with sharply predicted arrival-time windows of $\sim$3.4 days. Among 58 carefully selected stars, we identify two that show noticeable single-time-bin brightenings inside their predicted windows (where each time bin corresponds to a 200 s integrated TESS exposure). In both cases the brightenings coincide with excursions in at least one nearby star and are therefore most consistent with instrumental origins. This test demonstrates that the hybrid ring strategy is practical with existing survey data and could serve as a promising basis for future technosignature searches.

In this study, we experimentally investigated the wall heat transfer and flow field configuration of incident-reflected shock wave-turbulent boundary layer interactions on a cooled wall in supersonic flow. Wind tunnel experiments were conducted at a Mach number of 2.0 and a total temperature of 289 K. To create a cooled-wall state, the wind tunnel wall was cooled to a cryogenic temperature using liquid nitrogen at 77.4 K. In addition to conventional measurements, such as the schlieren visualization method and pressure measurements, cryogenic temperature-sensitive paint was employed to clarify the relationship between the flow field configuration and wall heat flux on a cryogenically cooled wall. The wall surface temperature of the cryogenically cooled wall was 95 K, corresponding to a wall-to-recovery temperature ratio of 0.34. The oil flow image and wall surface temperature distribution indicated a quasi-two-dimensional flow at the center of the wind tunnel. The schlieren images and wall pressure distributions showed that the separation point under the cooled-wall condition shifted downstream compared with that under the uncooled-wall condition. Based on the temperature distribution obtained from the cooled-wall experiments, the wall heat flux at the separation point reduced due to the outward flow from the wall. The peak wall pressure ratio and wall heat flux ratio normalized by their upstream values exhibited trends consistent with previously reported data under the cooled-wall condition. These results suggest that the cryogenic temperature-sensitive paint is a powerful tool for investigating the effects of wall temperature on the shock wave-turbulent boundary layer interactions on cryogenically cooled walls.

Suprathermal fusion reactions, initiated by energetic particles slowing down and scattering in dense plasmas, can modify the burn dynamics at inertial confinement fusion (ICF) regimes. A 0D time-dependent Monte-Carlo code has been developed to assess the suprathermal energy gain from fast fusions in DT, deuterium, $^{11}$BH$_3$ and $^{11}$BHDT fuels. It incorporates modified Li-Petrasso stopping powers, thermal broadening of cross-sections, anisotropic nuclear elastic and neutron elastic scattering, and a physical model for the p$^{11}$B alpha-particle spectra. Results show that earlier predictions of suprathermal criticality in pure deuterium are overestimated by more than an order of magnitude; no realistic density-temperature regime supports a self-sustaining chain reaction. Only DT demonstrates a critical regime provided there is no neutron leakage. Fast protons in $^{11}$BH$_3$ have an optimum energy of 4 MeV for maximising suprathermal enhancement. In this case the additional energy from fast fusions is unlikely to exceed 40% of the initial proton beam energy. The possibility of an alpha-particle-driven "avalanche" mechanism is ruled out since the ionic stopping is dominated by collisions involving small energy transfer. Suprathermal multiplication processes are dominated by neutron-driven ion up-scattering and likely play a limited role in purely aneutronic fuels.

Artificial atoms based on color centers in silicon (SiCCs) have recently emerged as promising candidates for highly integrable and scalable key components in photonic quantum technology, including telecom single-photon sources and spin memory devices. A novel all-epitaxial fabrication technique for SiCCs, based on ultra-low-temperature (ULT) molecular beam epitaxy (MBE), addresses limitations of conventional fabrication via ion implantation, such as vertical ion straggle and collateral crystal lattice damage. This method solely relies on self-assembly of SiCCs during kinetically-limited growth of (carbon-doped) Si(:C) at ULTs <~350{\deg}C. The latter requires an extraordinary pristine growth environment to prevent unintended defect formation caused by the incorporation of impurities from the background vapor; however, so far, no study has specifically addressed how exactly the vacuum conditions during epitaxy influence SiCC formation, their optical properties, and the quality of the surrounding crystal matrix. Here, we investigate the impact of the growth pressure and the substrate temperature on the self-assembly and photoluminescence (PL) properties of important SiCCs, such as W, G, G', and T centers. Further, we use PL and Doppler broadening variable energy positron annihilation spectroscopy to emphasize the role of the growth pressure in suppressing the luminescence background, which is crucial for advancing quantum photonics applications.