903 papers
Computational physics bridges the gap between abstract theory and real-world observation by using powerful computers to solve complex physical problems. This field allows scientists to simulate everything from the collision of subatomic particles to the swirling dynamics of galaxies, offering insights that traditional experiments alone cannot provide.
On Gist.Science, we continuously process every new preprint in this category from arXiv to make these breakthroughs accessible to everyone. Each entry is accompanied by both a clear, plain-language explanation and a detailed technical summary, ensuring that researchers and curious readers alike can grasp the significance of the latest findings without getting lost in dense equations.
Below are the latest papers in computational physics, curated to keep you at the forefront of this rapidly evolving discipline.
This paper introduces Neural-ISAM, a hybrid in-situ machine learning method that dynamically replaces pruned regions of adaptive tabulation databases with trained neural networks to significantly reduce memory requirements while maintaining accuracy in large-eddy simulations of complex turbulent flames.
jNO is a unified, JAX-native library that streamlines the training of neural operators and foundation models by integrating data-driven and physics-informed approaches into a single symbolic tracing system, enabling seamless transitions between operator regression, mesh-aware residual evaluation, and PDE-constrained optimization without code restructuring.
This paper introduces stochastic tensor contraction as a highly efficient computational primitive that reduces the cost of high-order tensor operations in ab initio quantum chemistry, specifically enabling coupled cluster theory to achieve chemical accuracy with mean-field scaling and significantly outperforming existing local correlation approximations in both speed and error.
This paper introduces CarCrashNet, a large-scale open-source benchmark comprising over 14,000 component-level and 825 full-vehicle crash simulations, alongside CrashSolver, a hierarchical neural solver designed to enable data-driven, AI-powered structural crash prediction and reproducible research in vehicle safety.
This paper presents a unified framework for solving diverse NP-hard combinatorial optimization problems on a Rydberg quantum annealer by mapping them to a QUBO formalism via local light-shifts encoding and an optimized quantum annealing protocol, while introducing a generalized hardness parameter to quantify problem complexity.
This study introduces a facet-resolved adsorption energy distribution framework utilizing machine-learned force fields to analyze 1.4 million adsorption sites across diverse alloy surfaces, thereby identifying specific compositions and orientations that optimize both activity and methanol selectivity for CO hydrogenation.
This paper introduces a physics-informed reduced-order operator learning framework that combines Equilibrium Neural Operators with QR-based discrete empirical interpolation to drastically reduce the computational cost of training and inference for 3D hyperelastic microstructure surrogate models while ensuring mechanical equilibrium and enabling accurate stress predictions.
This paper systematically compares Constrained Transport (CT) and Dedner's mixed divergence cleaning methods for astrophysical MHD simulations, revealing that the latter can produce significant artifacts and inaccuracies in scenarios involving localized magnetic fields or sudden timestep changes, thereby suggesting that CT is generally more accurate and reliable while proposing specific modifications to improve the robustness of divergence cleaning.
This paper introduces an uncertainty quantification toolkit extension to the KLIFF package within the OpenKIM framework, utilizing parallel-tempered Markov chain Monte Carlo to assess uncertainties arising from both parameter variations and functional form inadequacies in interatomic potentials, as demonstrated on a silicon Stillinger–Weber potential.
This contribution presents a scalable, convex optimization-based information alignment approach that leverages the Fisher information matrix to select minimal, high-quality training data for accurately predicting quantities of interest, thereby addressing data scarcity and parameter non-identifiability in diverse scientific modeling and active learning applications.