The high-speed manipulation of light using the Pockels effect is a missing building block in photonic integrated chips based on amorphous materials. This work demonstrates a process for inducing a longlasting inscribed electric field responsible for an effective Pockels effect in amorphous silicon nitride electro-optic modulators at room temperature.

Recent advances in synthetic methods have allowed for the isolation and observation of superconductivity in nickelate compounds, but the exact nature of underlying mechanisms is still to be clarified. Here, the authors investigate the effect of nonmagnetic impurities on the superconductivity of La₃Ni₂O₇ under high-pressure and how this can impact differently on the s and d-wave order parameters.

Critical brain hypothesis states that neuronal networks in the brain operate close to criticality. This study demonstrates the importance of the hierarchical and modular structure of neuron connections in maintaining criticality: modules with sparsely connected neurons can sustain critical activity even when subjected to significant perturbations; however, as connection density increases, modularity alone proves inadequate to preserve criticality.

Quantum electrodynamics on a one-dimensional lattice can lead to the localization of charges, in contrast to the thermalizing dynamics ubiquitous in nature. In this work the authors identify the physical origin of such non-thermal signatures and attribute it to the fracturing of the underlying Hilbert space.

Solving optimization problems efficiently is a critical challenge, yet traditional physical minimizers that map the problem to a spin Hamiltonian suffer from limitations in their spin dynamics. We overcome this issue by introducing the Vector Ising Spin Annealer, which operates in a higher-dimensional space and demonstrates superior performance

The authors introduce DD2D, a physics-guided deep learning method that predicts 2D structures directly from diffraction patterns using a twin-tower framework. The method demonstrates high anti-interference, robust recognition, and up to 99.0% prediction accuracy, showing promise for future 2D materials discoveries.

The time resolution for X-ray imaging with synchrotron sources is limited by the speed of the detectors and a large amount of data to process. The authors implemented an inline dual-camera setup able to resolve the data using sparse events, achieving temporal super-resolution for synchrotron X-ray imaging.

The 21-cm forest is a powerful probe for exploring the small-scale structures in the early universe. In this work, the authors use deep learning-driven likelihood-free parameter inference to quickly and accurately estimate parameters from minimally simulated datasets, while addressing the challenge of non-Gaussian likelihood function

Topological flat bands in translationally invariant lattices are challenging to achieve due to the need for infinite hopping range. Here, the authors demonstrate perfectly flat bands in a topologically nontrivial Floquet-Lieb microring lattice using periodic driving, enabling light localization and potential exploration of novel phenomena in strongly correlated quantum systems.

Programmable photonic integrated circuits (PPICs) promise transformative advances in computing, yet precise voltage configuration remains a challenge due to manufacturing-induced static errors. Here, the authors introduce an off-chip method combining gradient descent and genetic algorithms to optimize circuit configurations, achieving high fidelity in matrix implementations and enhancing PPIC performance across various applications.

Microresonator frequency combs are versatile tools for sensing, data transmission and quantum applications. In this work the authors present the generation of low-noise frequency combs at repetition rates of 100 GHz by utilizing a cascaded forward-propagating Brillouin scattering process to seed soliton frequency comb generation.

The authors present a partially fault-tolerant implementation of the quantum approximate optimization algorithm. By encoding circuits with the iceberg error detection code, the authors improve QAOA’s performance on problems with up to 20 logical qubits on a trapped-ion quantum computer, outlining conditions required to surpass classical algorithms.

Power exhaust is one of the biggest challenges stopping fusion energy. This article shows experimental evidence for strategically shaping the power exhaust region as a solution to this challenge, utilising physics understanding to strike a balance between engineering complexity and power exhaust benefits, consistent with reduced models and simulations.