Wednesday, 4 November 2026: 0 new papers, 5 cross-lists and 9 replacements in Computational Physics (physics.comp-ph), in arXiv's order, with full abstracts. The same papers as arXiv's own listing, set as a newspaper; every title links to arXiv.
Charge equilibration assigns atomic charges by equalizing an atomic chemical potential built from electronegativity, hardness and the Coulomb interaction. Under one global constraint it lets charge flow without limit between molecules, which fluctuating-charge models prevent with one neutrality constraint per molecule. A metal needs the opposite: a global constraint and a hardness near zero, and there the charge operator becomes indefinite unless it includes each charge's own self-energy. Here, we present samQEq (solids-and-molecules QEq), a unified charge-equilibration package for LAMMPS that combines the two constructions in one energy functional and one matrix-free solve. An orthogonal projector imposes one neutrality constraint per molecule, and the finite-width self-energy of a Gaussian charge on designated conductor types, narrower than the conductor's own Gaussian, keeps the conductor's part of the operator positive definite at any hardness. On a gold slab carrying a water film, the operator stays definite down to the conductor limit of the model, the metal screens a point charge by a constant amount at every height beyond 4 {\AA}, and the film's charges stay bounded under dynamics. Reciprocal-space electrostatics is evaluated inside the matrix-vector product, and three distinct shielding kernels are run-time settings. We verify the implementation by reproducing a published fluctuating-charge water model without refitting, and the native ReaxFF charge-equilibration step to approximately $10^{-9}\,e$. On a single four-V100 node the largest tested systems run at approximately twice the throughput of a 128-core CPU node. samQEq thus provides one charge-equilibration layer for molecules, conductors and the interfaces between them.
Training neural operators on million-cell meshes can exceed GPU memory limits. Models are therefore trained on subsamples and deployed on whole meshes or selected regions. However, subsampling can distort boundary information, degrading accuracy. We ask when boundary conditioning under sampling stays consistent with whole-mesh training, and how to aggregate and inject boundary data. Representing the solution as an integral of boundary data against type-dependent measures, we show that subsampling distorts this conditioning in three ways: incomplete boundary coverage or a sample-dependent global state, face weighting by count rather than area, and compression of boundary inputs into a low-dimensional vector. This analysis yields four measurable consistency conditions checkable before training. We satisfy them with two backbone-agnostic modules. Boundary Quadrature Attention (BQA) aggregates all boundary faces, typed and area-weighted, into slice tokens that every query point reads at every layer. Anchored Global State (AGS) estimates the global state from anchors independent of the sampled points. Adding both to a scalable message-passing and slice-attention backbone yields SCANO. We evaluate the approach on controlled and public benchmarks and a data-centre thermal simulation dataset (DCTS) containing million-cell meshes. On both benchmarks, BQA improves every backbone, reducing whole-mesh error up to 15-fold. On all three datasets, AGS aligns regional with whole-mesh predictions, cutting regional error at least tenfold for most backbones on the public benchmark. Together, the modules nearly halve the best original backbone's whole-mesh error on both benchmarks. SCANO predicts a room-scale field in under one second, over 100 times faster than whole-room anchor self-attention.
The gas dynamics of high speed flows over several compression-expansion ramp geometries with large separation regions were investigated using a particle kinetic approach and data driven methods. These large separation regions allow study of the role of scaled angle predicted by triple deck theory on recirculation strength, separation size, and appearance of secondary recirculation regions, all of which lead to flow unsteadiness. Examining four different free stream conditions for the same geometry using the direct simulation Monte Carlo (DSMC) approach, we show that the stability of single recirculation in the separation region is more sensitive to wall temperature and free stream velocity than physical ramp angle. Stability in this context refers to appearance of secondary separation regions. In addition, configurations with similar types of recirculation regions from literature considered, we find that general trends regarding unsteadiness, such as appearance of secondary recirculation regions, mostly hold; however, the flows are too complex to be characterized by only scaled angle. Thus, stochastic flowfields generated by DSMC were investigated using data driven methods, namely spectral proper orthogonal decomposition and dynamic mode decomposition. We found that both methods yield similar frequency components and mode shapes for spatio-temporal coherent structures of unsteady DSMC flows. With the help of these structures, the main mechanisms of unsteadiness were identified as oscillations of the reattachment shock, Kelvin-Helmholtz like instability of the shear layer, and c-shaped mode of the separation bubble. This is the first time SPOD has been applied to a DSMC generated flowfield and compared with DMD.
Bragg Coherent Diffraction Imaging (BCDI) relies on phase retrieval (PR), a non-convex, ill-posed inverse problem whose conditioning is degraded by factors like noise, partial coherence, and strain or crystallographic defects in the imaged nanocrystal. Conventional approaches, including alternating projections and deep learning methods, represent the object on a dense voxel grid, which is prone to amplitude-phase mixing and becomes increasingly ill-posed under these factors, often preventing reliable three-dimensional mapping of the particle's shape and strain. Here we present ConvexAD, a gradient descent PR model that parametrizes the object shape as a convex polyhedron, built by the intersection of half-spaces and enforces flat amplitude within it by construction. This geometric regularization eliminates amplitude-phase mixing and reduces the number of trainable parameters, hence facilitating the convergence of the PR. Implemented in JAX for native GPU acceleration and forward-model flexibility, ConvexAD is validated on simulated and experimental BCDI data, where it outperforms conventional and deep learning assisted PR at high strain, low signal, and reduced coherence, extending the conditions under which quantitative BCDI reconstruction is achievable.
This work presents an optimization-based remap approach for arbitrary Lagrangian-Eulerian (ALE) hydrodynamics that preserves sharp features and physical bounds while conserving global quantities such as volume, mass, momentum, and total energy. Achieving these properties is the main challenge in remap methods. The presented approach is well-suited for high-order finite element (FE) discretizations and ALE methods that allow for large, purely Lagrangian displacements between successive remeshing and remapping steps. Recent advances in parallel, GPU-capable interpolation routines enable the initial transfer of all fields via direct physical-space interpolation, avoiding computationally costly and diffusive pseudo-transient PDE-based advection. The interpolation produces bounded fields with minimal numerical diffusion. However, it may violate the desired conservation of global properties. Thus, a subsequent optimization step restores conservation while preserving physical bounds. The presented method provides minimally diffusive remapping of finite element fields of arbitrary order. Moreover, it directly remaps quantities defined at integration points, eliminating the need for projection onto finite element spaces. The approach also mitigates typical challenges associated with synchronizing primal variables and conserved quantities in advection-based remap methods. The proposed optimization-based remap method is evaluated using standard two- and three-dimensional remap benchmarks and full ALE hydrodynamics simulations.