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 investigates the pursuit-evasion dynamics between a deterministic, self-steering pursuer and a stochastic, cognitive evader in two dimensions, revealing that the evader's capture time is significantly influenced by whether it adopts a high-risk backward maneuver or a forward-tumbling strategy with continuous adjustments depending on the pursuer's dominance.
This paper presents an extension of the EPW package to calculate electron-phonon coupling in magnetic materials using the local spin density approximation, revealing through validation on ferromagnetic iron and nickel that electron-phonon scattering is the dominant resistivity mechanism in iron but accounts for less than one-third of the resistivity in nickel.
This paper introduces MiAD, an equivariant joint diffusion model that utilizes a novel "mirage infusion" technique to dynamically alter the number of atoms during generation, thereby significantly improving the discovery of stable, unique, and novel crystalline materials compared to existing state-of-the-art approaches.
This paper establishes an exact mathematical correspondence between deep learning training and Hamilton-Jacobi initial-value problems, unifying neural network architectures, tropical algebra, viscous PDEs, and convex optimization under a single deformation parameter to derive precise theoretical insights into generalization, robustness, and attribution.
This paper proposes a hybrid quantum-classical variational algorithm utilizing polynomial approximations of strain energy density to solve nonlinear finite element problems for hyperelastic materials on near-term quantum devices, demonstrating its feasibility through numerical experiments on a one-dimensional Neo-Hookean model.
This paper introduces WF-Bench, a comprehensive benchmarking dataset and protocol that evaluates neural network wavefunction expressivity across diverse quantum many-body systems, revealing empirical scaling laws and establishing a unified framework for comparing architectures like Psiformer and Ferminet.
This paper reveals a largely system-independent power-law relationship between the coupling strength required to trigger extreme events in networked dynamical systems and the topological or spectral properties of their coupling structures.
This paper presents a synergistic framework combining coarse-grained simulations, experimental rheology, and machine learning to efficiently map the design space of DNA-based soft matter fluids, enabling the rational and accelerated discovery of materials with tailored bulk rheological properties.
This paper demonstrates that syntactic and semantic information in the DeepSeek-V3 LLM are partially linearly encoded and differentially distributed across layers, as evidenced by the ability to decouple these signals through the subtraction of averaged representation centroids.
GenSBI is a new open-source JAX library that implements flow matching, score matching, and denoising diffusion models with transformer-based architectures to provide a native, end-to-end simulation-based inference framework for researchers using JAX, achieving high accuracy and well-calibrated posteriors on standard benchmarks.