Articles in 2025

Intensity-rotating beams are a class of structured beams that can potentially enable advanced optical manipulation techniques, as well as applications in free-space optical communication and high-resolution imaging, but their exploration has remained mostly theoretical. The authors demonstrate galactic-form spinning beams, i.e a stable, intensity-rotating beam whose dynamics and divergence can be precisely controlled.

Rhythmic phenomena, ubiquitous in biological systems, are often modeled as networks of coupled limit-cycle oscillators. This study introduces an adaptive LASSO based approach to infer the interaction type-"including pairwise and higher-order interactions-" from time-series data, demonstrating its effectiveness on synthetic datasets and human brain network data.

Systems across logistics, economics, ecology, and computing, where agents aim to maximize gain or minimize cost in resource allocation, exhibit behavior consistent with constrained optimization. This study combines optimal transport theory and maximum-entropy models to introduce a class of bipartite random graphs that enables systematic characterization of how dense, entropic graphs evolve into sparse, efficient structures, revealing the most probable network configurations under partial optimization.

Liquid crystal skyrmions are quasiparticle structures whose interactions at large separations can be understood in terms of elastic multipoles. Here, the authors derive these multipole interactions analytically and show how they drive the formation of chains and clusters under oscillating electric fields.

Stabilizing the ferroelectric phase in HfO₂ is a critical challenge for its application in next-generation nanoelectronics. Here, through a combination of first-principles calculations and experimental results the authors identify a critical extra-hole concentration for the stabilisation of the ferroelectric phase in HfO₂ thin films.

The authors develop a microwave interferometer method to measure how electric and acoustic waves mix inside piezoelectric materials. Using this approach, they detect and quantified electric-acoustic heterodyning in lead zirconium titanate, showing the mixed signal scales linearly with both input powers.

Ranking items based on pairwise comparisons, such as using match outcomes to rank sports teams, is a common task that becomes challenging when data is limited or noisy. Here, the authors introduce an efficient nonparametric Bayesian method for learning partial rankings that breaks ties among item ranks only when supported by sufficient statistical evidence in the data.

The structure of ties in social networks determines who receives information first, and those early opportunities often confer a competitive advantage. The authors develop the H3 hypergraph model, which reveals how hyperedge homophily drives inequality in information access and suggests that targeted interventions informed by higher-order dynamics can help close those gaps.

Magnetic skyrmions are chiral spin structures that can form two-dimensional lattices. Here, the authors show how lattice domains grow under the influence of pinning effects in a non-flat energy landscape and that magnetic field oscillations help to stabilize order.

High-energy electron beams interacting with targets produce electromagnetic showers, a pathway for producing electronpositron pair plasma in the laboratory. Here, the authors solve kinetic equations to derive spectra and pair numbers for varying target thicknesses, with results validated by Geant4 simulations, an explicit formula for the density of outgoing pairs is then obtain.

Synchronization of nanoscale photonic devices generating pulses is of prime importance for on-chip communications and analog all-optical machines. Beyond synchronization, by using non-identical thermo-optical self-pulsing photonic crystal integrated on silicon circuit, the authors demonstrate multiple complex behaviors of the temporal dynamics in the parameter space.

Amorphous silicon carbide microring sensors provide a compact platform for optical ultrasound detection. The authors demonstrate 20- element array coupled to single bus waveguide attains high finesse and sensitivity, enabling high-resolution imaging in a miniaturized, easily integrated system.

Rare-earth ions are often crucial for the emergence of multiferroicity, yet their 4f-driven magnetic domain patterns remain elusive. Using optical second harmonic generation, the authors uncover columnar Dy/Tb domains in multiferroic Dy_0.7Tb_0.3FeO₃. The domain orientation enforces charged ferroelectric walls, underscoring the pivotal role of 4f order in shaping multiferroic domains.

Causality plays a central role in understanding interactions between variables in complex systems. Here, the authors introduce a state-aware causal inference method that quantifies causality in terms of information gain about future states, demonstrating its applicability in the context of interactions between motions across scales in a turbulent boundary layer and the Walker circulation phenomenon in tropical Pacific climate dynamics.

Quantum networks with bipartite resources and shared randomness are important for the future quantum internet, yet the types of entanglement they can generate remain unclear. The authors establish improved upper bounds on fidelity for graph states and propose efficient protocols for generating genuine multipartite entangled states, enhancing quantum network capabilities.

Spin photocurrents can be efficiently generated by the resonant excitation associated with defect states in unintentionally doped GaN. Here, the authors show that shallow Si-donor states can be utilised to enhance the magnitude of the spin photocurrent, and that defect induced deep level states provide a feasible method to modulate the direction of the spin photocurrent.

Laser wakefield acceleration offers compact high-energy electron sources, but controlling electron injection remains challenging. The authors experimentally demonstrate a controlled “scissor-cross ionization injection” mechanism that achieves localized, stable, and tunable electron injection using an oblique trigger laser.

Non-Hermitian physics extends criticality beyond conventional Hermitian systems. Here, the authors report a universal many-body critical skin effect arising from the interplay of multiple pumping channels and interactions, revealing new collective edge phenomena and real-to-complex transitions.

Traditional network diffusion models primarily focus on direct connections, often overlooking the role of indirect pathways. Using the d-path Laplacian framework, this study quantifies indirect influence and reveals a structural phase transition associated with altered diffusion dynamics, providing new insights into network behavior and applications in information spreading, disease propagation, or transport dynamics across physical and biological systems.

Using high-resolution measurements in a magnetically confined plasma, two distinct turbulence types were directly observed: local turbulence, which carries heat, and nonlocal turbulence, which rapidly links distant regions. The authors show that nonlocal turbulence appears first after heating and becomes increasingly dominant for shorter pulses, revealing how mediator structures enhance nonlocal transport.