PRX Intelligence

Chief Editor Anatole von Lilienfeld and the PRX Intelligence editorial team outline the newest APS journal’s vision, scope and philosophy in the context of the rapidly evolving landscape of AI and scientific discovery.

A first-principles method that builds neural network representations of many-body excited states provides an accurate and parallelizable tool for estimating critical exponents.

Augmenting autonomous experimentation with human-in-the-loop guidance significantly improves efficiency of AI-based materials exploration.

A neural-network variational Monte Carlo method is used to study the fractional quantum Hall effect on the torus, establishing neural-network wavefunctions as a powerful tool for investigating anyonic properties.

A resolution-robust data-driven closure for the electron heat flux bridges kinetic and fluid descriptions of inertial confinement fusion plasmas.

A framework that combines diffusion-based materials generation with hierarchical thermodynamic and electronic structure screening accelerates discovery of inorganic electrides.

A Symmetry-adapted Gaussian process regression model is employed for electronic structure prediction of twisted bilayer moiré materials, unraveling the impact of long-range interactions.

Graph neural networks enable efficient prediction of the electronic density of states, making it possible to identify application-specific, structurally diverse compounds with similar electronic properties across vast chemical spaces.

Resolving the angular content of internal representations, layer by layer, reveals how unconstrained machine-learning models learn near-exact equivariance, and which minimal inductive biases push symmetry breaking down to harmless levels.

Joint diffusion is leveraged as a multimodal generative surrogate model for inertial confinement fusion (ICF), enabling post-shot inference, diagnostic optimization, and accelerated ICF design.

Broken scaling laws are observed across three multiobjective graph neural network architectures trained to predict the optical properties of solids.

A self-attention neural-network variational wavefunction is capable of describing both fractional quantum Hall liquids and electron crystals within the same architecture.

A quantum circuit realization of flow matching enables efficient interpolation between density matrices, and can be implemented in existing quantum computing architectures without costly redesigns.

Neural decoders can serve as a robust and accurate foundation for fault-tolerant quantum algorithms under realistic conditions.

An efficient reciprocal space based predictor of crystal properties achieves improved accuracy in predicting various crystal properties compared to existing crystal graph convolutional neural network methods, at a much lower computational cost.

A message-passing neural network method is presented that enables accurate many-body dispersion calculations spanning the periodic table for organic and inorganic molecules as well as organic solids using only the molecular structure as input, providing a streamlined tool for incorporating van der Waals interactions into molecular simulations on top of semiempirical methods and machine learning force fields.

A hybrid framework combining parameterized quantum circuits with transformer-based neural quantum states is proposed to construct variational wavefunctions for quantum many-body systems.

A data-driven linear foundation model identifies a compact variational subspace for quantum embedding, substantially reducing the cost of solving embedding Hamiltonians and enabling faster simulations of strongly correlated systems.

A novel clustering approach demonstrates that similarity-based representation learning combined with density-based aggregation is a promising strategy for point cloud segmentation in highly granular particle detectors.

A field-aware equivariant interatomic potential integrates electric bias into foundation-model simulations across chemical space, learning a differentiable electric enthalpy for polarization, polarizability, and Born effective charges.

A convolutional neural network is utilized to autonomously tune a minimal realization of a Kitaev chain toward a Poor Man’s Majorana sweet spot.