The fermion sign problem constitutes a fundamental computational bottleneck across a plethora of research fields in physics, quantum chemistry and related disciplines. Recently, it has been suggested to alleviate the sign problem in \emph{ab initio} path integral Molecular Dynamics and path integral Monte Carlo (PIMC) calculations based on the simulation of fictitious identical particles that are represented by a continuous quantum statistics variable $\xi$ [\textit{J.~Chem.~Phys.}~\textbf{157}, 094112 (2022)]. This idea facilitated a host of applications including the interpretation of an x-ray scattering experiment with strongly compressed beryllium at the National Ignition Facility [\textit{Nature Commun.}~\textbf{16}, 5103 (2025)]. In the present work, we express the original isothermal $\xi$-extrapolation method as a special case of a truncated Taylor series expansion around the $\xi=0$ limit of distinguishable particles. We derive new PIMC estimators that allow us to evaluate the Taylor coefficients up to arbitrary order and we carry out extensive new PIMC simulations of the warm dense electron gas to systematically analyze the sign problem from this new perspective. This gives us important insights into the applicability of the $\xi$-extrapolation method for different levels of quantum degeneracy in terms of the Taylor series radius of convergence. Moreover, the direct PIMC evaluation of the $\xi$-derivatives, in principle, removes the necessity for simulations at different values of $\xi$ and can facilitate more efficient simulations that are designed to maximize compute time in those regions of the full permutation space that contribute most to the final Taylor estimate of the fermionic expectation value of interest.

Energy decomposition analysis (EDA) based on absolutely localized molecular orbitals provides detailed insights into intermolecular bonding by decomposing the total molecular binding energy into physically meaningful components. Here, we develop a neural network EDA model capable of predicting the electron delocalization energy component of water molecules, which captures the stabilization arising from charge transfer between occupied absolutely localized molecular orbitals of one molecule and the virtual orbitals of another. Exploiting the locality assumption of the electronic structure, our model enables accurate prediction of electron delocalization energies for molecular systems far beyond the size accessible to conventional density functional theory calculations, while maintaining its accuracy. We demonstrate the applicability of our approach by modeling hydration effects in large molecular complexes, specifically in metal-organic frameworks.

We present relativistic $N$-body simulations of a $\Lambda_{\rm s}$CDM - sign-switching cosmological constant (CC) - scenario under general relativity and compare its nonlinear matter power spectrum to $\Lambda$CDM at ${z = 15,\,2,\,1,\,0}$, using best-fit parameters from Planck-only and a combined ''full'' dataset. During the AdS-like CC (\Lambda_{\rm s}<0) phase, prior to the transition redshift $z_\dagger$, reduced Hubble friction dynamically enhances the growth of perturbations; after the switch, with dS-like CC (\Lambda_{\rm s}>0), the larger late-time expansion rate partly suppresses, but does not erase, the earlier amplification. Consequently, the ratio $P_{\Lambda_{\rm s}\rm CDM}/P_{\Lambda\rm CDM}$ exhibits a pronounced, redshift-dependent shape feature: a crest peaking at ${\sim 20-25\%}$ around ${k \simeq 1-3\,h\,\mathrm{Mpc}^{-1}}$ near the transition, which then migrates to larger physical scales and persists to ${z = 0}$ as a robust ${\sim 15-20\%}$ uplift at ${k \simeq 0.6-1.0\,h\,\mathrm{Mpc}^{-1}}$. These wavenumbers correspond to group/poor-cluster environments and lie within the sensitivity range of weak lensing, galaxy-galaxy lensing, cluster counts, and tSZ power, providing a concrete, falsifiable target that cannot be mimicked by a scale-independent change in $\sigma_8$ or $S_8$. The timing (earlier for Planck-only, later for the full dataset) and the amplitude of the crest align with the ''cosmic noon'' epoch (${z \simeq 1-2}$), offering a gravitational prior for the observed peak in the cosmic star-formation rate.

X-ray Thomson scattering (XRTS) constitutes an essential technique for diagnosing material properties under extreme conditions, such as high pressures and intense laser heating. Time-dependent density functional theory (TDDFT) is one of the most accurate available ab initio methods for modeling XRTS spectra, as well as a host of other dynamic material properties. However, strong thermal excitations, along with the need to account for variations in temperature and density as well as the finite size of the detector significantly increase the computational cost of TDDFT simulations compared to ambient conditions. In this work, we present a broadly applicable method for optimizing and enhancing the efficiency of TDDFT calculations. Our approach is based on a one-to-one mapping between the dynamic structure factor and the imaginary time density--density correlation function, which naturally emerges in Feynman's path integral formulation of quantum many-body theory. Specifically, we combine rigorous convergence tests in the imaginary time domain with a constraints-based noise attenuation technique to improve the efficiency of TDDFT modeling without the introduction of any significant bias. As a result, we can report a speed-up by up to an order of magnitude, thus potentially saving millions of CPU hours for modeling a single XRTS measurement of matter under extreme conditions.

Recognizing less salient features is the key for model compression. However, it has not been investigated in the revolutionary attention mechanisms. In this work, we propose a novel normalization-based attention module (NAM), which suppresses less salient weights. It applies a weight sparsity penalty to the attention modules, thus, making them more computational efficient while retaining similar performance. A comparison with three other attention mechanisms on both Resnet and Mobilenet indicates that our method results in higher accuracy. Code for this paper can be publicly accessed at this https URL.

Due to the recent announcement of the Frontier supercomputer, many scientific application developers are working to make their applications compatible with AMD architectures (CPU-GPU), which means moving away from the traditional CPU and NVIDIA-GPU systems. Due to the current limitations of profiling tools for AMD GPUs, this shift leaves a void in how to measure application performance on AMD GPUs. In this paper, we design an instruction roofline model for AMD GPUs using AMD's ROCProfiler and a benchmarking tool, BabelStream (the HIP implementation), as a way to measure an application's performance in instructions and memory transactions on new AMD hardware. Specifically, we create instruction roofline models for a case study scientific application, PIConGPU, an open source particle-in-cell (PIC) simulations application used for plasma and laser-plasma physics on the NVIDIA V100, AMD Radeon Instinct MI60, and AMD Instinct MI100 GPUs. When looking at the performance of multiple kernels of interest in PIConGPU we find that although the AMD MI100 GPU achieves a similar, or better, execution time compared to the NVIDIA V100 GPU, profiling tool differences make comparing performance of these two architectures hard. When looking at execution time, GIPS, and instruction intensity, the AMD MI60 achieves the worst performance out of the three GPUs used in this work.

We determine the resonating-valence-bond (RVB) state in graphene using real-space quantum Monte Carlo with correlated variational wave functions. Variational and diffusion quantum Monte Carlo (DMC) calculations with Jastrow-Slater-determinant and Jastrow-antisymmetrized-geminal-power ansatze are employed to evaluate the RVB pairing energy. Using a rectangular graphene sample that lacks $\pi/3$ rotational symmetry, we found that the single-particle energy gap near the Fermi level depends on the system size along the $x$-direction. The gap vanishes when the length satisfies $L_x=3n\sqrt{3}d$, where $n$ is an integer and $d$ is the carbon-carbon bond length, otherwise, the system, exhibits a finite gap. Our DMC results show no stable RVB pairing in the zero-gap case, whereas the opening of a finite gap near the Fermi level stabilizes the electron pairing. The DMC predicted absolute value of pairing energy at the thermodynamic limit for a finite-gap system is $\sim 0.48(1)$ mHa/atom. Our results reveal a feometry-driven electron pairing mechanism in the confined graphene nanostructure.

Julia is a mature general-purpose programming language, with a large ecosystem of libraries and more than 12000 third-party packages, which specifically targets scientific computing. As a language, Julia is as dynamic, interactive, and accessible as Python with NumPy, but achieves run-time performance on par with C/C++. In this paper, we describe the state of adoption of Julia in HEP, where momentum has been gathering over a number of years. HEP-oriented Julia packages can already, via this http URL, read HEP's major file formats, including TTree and RNTuple. Interfaces to some of HEP's major software packages, such as through this http URL, are available too. Jet reconstruction algorithms in Julia show excellent performance. A number of full HEP analyses have been performed in Julia. We show how, as the support for HEP has matured, developments have benefited from Julia's core design choices, which makes reuse from and integration with other packages easy. In particular, libraries developed outside HEP for plotting, statistics, fitting, and scientific machine learning are extremely useful. We believe that the powerful combination of flexibility and speed, the wide selection of scientific programming tools, and support for all modern programming paradigms and tools, make Julia the ideal choice for a future language in HEP.

Time-dependent density functional theory (TDDFT) is a widely used method to investigate electron dynamics under external time-dependent perturbations such as laser fields. In this work, we present a machine learning approach to accelerate electron dynamics simulations based on real time TDDFT using autoregressive neural operators as time-propagators for the electron density. By leveraging physics-informed constraints and featurization, and high-resolution training data, our model achieves superior accuracy and computational speed compared to traditional numerical solvers. We demonstrate the effectiveness of our model on a class of one-dimensional diatomic molecules under the influence of a range of laser parameters. This method has potential in enabling on-the-fly modeling of laser-irradiated molecules and materials by utilizing fast machine learning predictions in a large space of varying experimental parameters of the laser.

We present a physics-informed Bayesian analysis of equation of state constraints using observational data for masses, radii and tidal deformability of pulsars and a generic class of hybrid neutron star equation of state with color superconducting quark matter on the basis of a recently developed nonlocal chiral quark model. The nuclear matter phase is described within a relativistic density functional model of the DD2 class and the phase transition is obtained by a Maxwell construction. We find the region in the two-dimensional parameter space spanned by the vector meson coupling and the scalar diquark coupling, where three conditions are fulfilled: (1) the Maxwell construction can be performed, \mbox{(2) the maximum} mass of the hybrid neutron star is not smaller than \mbox{2.0 M$_\odot$} and (3) the onset density of the phase transition is not below the nuclear saturation density $n_0=0.15$ fm$^{-3}$. The result of this study shows that the favorable neutron star equation of state has low onset masses for the occurrence of a color superconducting quark matter core between 0.5-0.7 $M_\odot$ and maximum masses in the range 2.15-2.22 $M_\odot$. In the typical mass range of 1.2-2.0 $M_\odot$, the radii of these stars are between 11.9 and 12.4 km, almost independent of the mass. In principle, hybrid stars would allow for larger maximum masses than provided by the hadronic reference equation of state.

Accurate knowledge of the properties of hydrogen at high compression is crucial for astrophysics (e.g. planetary and stellar interiors, brown dwarfs, atmosphere of compact stars) and laboratory experiments, including inertial confinement fusion. There exists experimental data for the equation of state, conductivity, and Thomson scattering spectra. However, the analysis of the measurements at extreme pressures and temperatures typically involves additional model assumptions, which makes it difficult to assess the accuracy of the experimental data. rigorously. On the other hand, theory and modeling have produced extensive collections of data. They originate from a very large variety of models and simulations including path integral Monte Carlo (PIMC) simulations, density functional theory (DFT), chemical models, machine-learned models, and combinations thereof. At the same time, each of these methods has fundamental limitations (fermion sign problem in PIMC, approximate exchange-correlation functionals of DFT, inconsistent interaction energy contributions in chemical models, etc.), so for some parameter ranges accurate predictions are difficult. Recently, a number of breakthroughs in first principle PIMC and DFT simulations were achieved which are discussed in this review. Here we use these results to benchmark different simulation methods. We present an update of the hydrogen phase diagram at high pressures, the expected phase transitions, and thermodynamic properties including the equation of state and momentum distribution. Furthermore, we discuss available dynamic results for warm dense hydrogen, including the conductivity, dynamic structure factor, plasmon dispersion, imaginary-time structure, and density response functions. We conclude by outlining strategies to combine different simulations to achieve accurate theoretical predictions.

X-ray Thomson scattering (XRTS) constitutes an essential technique for diagnosing material properties under extreme conditions, such as high pressures and intense laser heating. Time-dependent density functional theory (TDDFT) is one of the most accurate available ab initio methods for modeling XRTS spectra, as well as a host of other dynamic material properties. However, strong thermal excitations, along with the need to account for variations in temperature and density as well as the finite size of the detector significantly increase the computational cost of TDDFT simulations compared to ambient conditions. In this work, we present a broadly applicable method for optimizing and enhancing the efficiency of TDDFT calculations. Our approach is based on a one-to-one mapping between the dynamic structure factor and the imaginary time density--density correlation function, which naturally emerges in Feynman's path integral formulation of quantum many-body theory. Specifically, we combine rigorous convergence tests in the imaginary time domain with a constraints-based noise attenuation technique to improve the efficiency of TDDFT modeling without the introduction of any significant bias. As a result, we can report a speed-up by up to an order of magnitude, thus potentially saving millions of CPU hours for modeling a single XRTS measurement of matter under extreme conditions.

We present relativistic $N$-body simulations of a $\Lambda_{\rm s}$CDM - sign-switching cosmological constant (CC) - scenario under general relativity and compare its nonlinear matter power spectrum to $\Lambda$CDM at ${z = 15,\,2,\,1,\,0}$, using best-fit parameters from Planck-only and a combined ''full'' dataset. During the AdS-like CC (\Lambda_{\rm s}<0) phase, prior to the transition redshift $z_\dagger$, reduced Hubble friction dynamically enhances the growth of perturbations; after the switch, with dS-like CC (\Lambda_{\rm s}>0), the larger late-time expansion rate partly suppresses, but does not erase, the earlier amplification. Consequently, the ratio $P_{\Lambda_{\rm s}\rm CDM}/P_{\Lambda\rm CDM}$ exhibits a pronounced, redshift-dependent shape feature: a crest peaking at ${\sim 20-25\%}$ around ${k \simeq 1-3\,h\,\mathrm{Mpc}^{-1}}$ near the transition, which then migrates to larger physical scales and persists to ${z = 0}$ as a robust ${\sim 15-20\%}$ uplift at ${k \simeq 0.6-1.0\,h\,\mathrm{Mpc}^{-1}}$. These wavenumbers correspond to group/poor-cluster environments and lie within the sensitivity range of weak lensing, galaxy-galaxy lensing, cluster counts, and tSZ power, providing a concrete, falsifiable target that cannot be mimicked by a scale-independent change in $\sigma_8$ or $S_8$. The timing (earlier for Planck-only, later for the full dataset) and the amplitude of the crest align with the ''cosmic noon'' epoch (${z \simeq 1-2}$), offering a gravitational prior for the observed peak in the cosmic star-formation rate.

The initialization and control of a long-lived spin population in lead halide perovskites are prerequisites for their use in spintronic applications. Here, we demonstrate circular polarization of the interlayer exciton emission in a (BA)2PbI4/WSe2 monolayer heterostructure. The helicity of this emission is controlled by tuning the energy of the excitation laser through the manifold of exciton resonances of the WSe2 monolayer, together with an emerging interlayer absorption feature of the heterostructure. Theoretical calculations show that this resonance arises from hybridized (BA)2PbI4/WSe2 states in the valence band. This hybrid character enables its observation in both linear absorption and ultrafast pump-probe spectroscopies, and plays a key role in controlling the sign of the helicity of the interlayer exciton emission. The tunable spin polarization demonstrated here, with the WSe2 monolayer effectively acting as a tunable spin filter, represents an important step toward the use of 2D perovskites in opto-spintronic applications.

We investigate a two-dimensional system of interacting Active Brownian Particles. Using the Martin-Siggia-Rose-Janssen-de Dominicis formalism, we built up the generating functional for correlation functions. We study in detail the hydrodynamic regime with a constant density stationary state. Our findings reveal that, within a small density fluctuations regime, an emergent $U(1)$ gauge symmetry arises, originated from the conservation of fluid vorticity. Consequently, the interaction between the orientational order parameter and density fluctuations can be cast into a gauge theory, where the concept of ``electric charge density" aligns with the local vorticity of the original fluid. We study in detail the case of a microscopic local two-body interaction. We show that, upon integrating out the gauge fields, the stationary states of the rotational degrees of freedom satisfy a non-local Frank free energy for a nematic fluid. We give explicit expressions for the splay and bend elastic constants as a function of the Péclet number (${\rm Pe}$) and the diffusion interaction constant ($k_d$).

We present a freeze-out approach to the formation of heavy elements in expanding nuclear matter. Applying concepts used in the description of heavy-ion collisions or ternary fission, we determine the abundances of heavy elements taking into account in-medium effects such as Pauli blocking and the Mott effect, which describes the dissolution of nuclei at high densities of nuclear matter. With this approach, we search for a universal primordial distribution in an equilibrium state from which the gross structure of the solar abundances of heavy elements freezes out via radioactive decay of the excited states. The universal primordial state is characterized by the Lagrangian parameters of temperature and chemical potentials of neutrons and protons. We show that such a state exists and determine a temperature of 5.266 MeV, a neutron chemical potential of 940.317 MeV and a proton chemical potential of 845.069 MeV, at a baryon number density of 0.013 fm$^{-3}$ and a proton fraction of 0.13. Heavy neutron-rich nuclei such as the hypothesized double-magic nucleus $^{358}$Sn appear in the primordial distribution and contribute to the observed abundances after fission. We discuss astrophysical scenarios for the realization of this universal primordial distribution for heavy element nucleosynthesis, including supernova explosions, neutron star mergers and the inhomogeneous Big Bang. The latter scenario may be of interest in the light of early massive objects observed with the James Webb Space Telescope and opens new perspectives to explain universality of the observed r-process patterns and the lack of observations of population III stars.

Electronic structure theory calculations offer an understanding of matter at the quantum level, complementing experimental studies in materials science and chemistry. One of the most widely used methods, density functional theory (DFT), maps a set of real interacting electrons to a set of fictitious non-interacting electrons that share the same probability density. Ensuring that the density remains the same depends on the exchange-correlation (XC) energy and, by a derivative, the XC potential. Inversions provide a method to obtain exact XC potentials from target electronic densities, in hopes of gaining insights into accuracy-boosting approximations. Neural networks provide a new avenue to perform inversions by learning the mapping from density to potential. In this work, we learn this mapping using physics-informed machine learning (PIML) methods, namely physics informed neural networks (PINNs) and Fourier neural operators (FNOs). We demonstrate the capabilities of these two methods on a dataset of one-dimensional atomic and molecular models. The capabilities of each approach are discussed in conjunction with this proof-of-concept presentation. The primary finding of our investigation is that the combination of both approaches has the greatest potential for inverting the Kohn-Sham equations at scale.

The performance gap between CPU and memory widens continuously. Choosing the best memory layout for each hardware architecture is increasingly important as more and more programs become memory bound. For portable codes that run across heterogeneous hardware architectures, the choice of the memory layout for data structures is ideally decoupled from the rest of a program. This can be accomplished via a zero-runtime-overhead abstraction layer, underneath which memory layouts can be freely exchanged. We present the Low-Level Abstraction of Memory Access (LLAMA), a C++ library that provides such a data structure abstraction layer with example implementations for multidimensional arrays of nested, structured data. LLAMA provides fully C++ compliant methods for defining and switching custom memory layouts for user-defined data types. The library is extensible with third-party allocators. Providing two close-to-life examples, we show that the LLAMA-generated AoS (Array of Structs) and SoA (Struct of Arrays) layouts produce identical code with the same performance characteristics as manually written data structures. Integrations into the SPEC CPU\textsuperscript{\textregistered} lbm benchmark and the particle-in-cell simulation PIConGPU demonstrate LLAMA's abilities in real-world applications. LLAMA's layout-aware copy routines can significantly speed up transfer and reshuffling of data between layouts compared with naive element-wise copying. LLAMA provides a novel tool for the development of high-performance C++ applications in a heterogeneous environment.

We present the Materials Learning Algorithms (MALA) package, a scalable machine learning framework designed to accelerate density functional theory (DFT) calculations suitable for large-scale atomistic simulations. Using local descriptors of the atomic environment, MALA models efficiently predict key electronic observables, including local density of states, electronic density, density of states, and total energy. The package integrates data sampling, model training and scalable inference into a unified library, while ensuring compatibility with standard DFT and molecular dynamics codes. We demonstrate MALA's capabilities with examples including boron clusters, aluminum across its solid-liquid phase boundary, and predicting the electronic structure of a stacking fault in a large beryllium slab. Scaling analyses reveal MALA's computational efficiency and identify bottlenecks for future optimization. With its ability to model electronic structures at scales far beyond standard DFT, MALA is well suited for modeling complex material systems, making it a versatile tool for advanced materials research.