Research Digest — 2026-03-21¶
Papers¶
1. An SO(3)-Equivariant Reciprocal-Space Neural Potential for Long-Range Interactions¶
Source: arXiv (physics.chem-ph), Mar 19, 2026 · 📅 2026-03-19 · ↗ Open paper
EquiEwald performs equivariant message passing in reciprocal space using learned k-space filters, embedding an Ewald-inspired formulation inside an SO(3)-equivariant neural network. This captures long-range electrostatics without sacrificing the physical consistency that makes modern equivariant MLIPs powerful. SO(3)-equivariant architecture with reciprocal-space message passing, equivariant k-space filters, and equivariant inverse Fourier transform. Benchmarked on periodic and aperiodic systems against ab initio reference data. First MLIP architecture to combine true reciprocal-space long-range treatment with SO(3) equivariance in a unified framework. Previous long-range extensions either broke equivariance or sacrificed energy-force consistency.
Relevance to DENG.Group
Directly relevant to Yanhao Deng's MLIP work. Long-range electrostatics are critical for solid electrolytes where Li⁺ migration depends on the electrostatic potential landscape. Current MLIPs (MACE, NequIP) use local environments only — this paper introduces a principled reciprocal-space extension that captures anisotropic multipolar correlations while preserving SO(3) equivariance and energy-force consistency. Test EquiEwald for halide electrolyte MLIPs (Yanhao) — if it captures long-range electrostatic effects better than local-only MACE, it could improve ionic conductivity predictions, especially for charged defects and interfaces. Also relevant for Umang's electrolyte/electrode interface work.
2. Design Space of Self-Consistent Electrostatic Machine Learning Interatomic Potentials¶
Source: arXiv (physics.chem-ph), Mar 16, 2026 · 📅 2026-03-16 · ↗ Open paper
Existing electrostatic MLIPs are coarse-grained approximations to DFT. By making this explicit, the paper identifies the full design space of self-consistent electrostatic MLIPs, reveals equivalences between previously proposed models, and identifies fundamental limitations. Theoretical framework viewing electrostatic MLIPs as DFT coarse-graining. Shared charge density representation within MACE architecture. Benchmarked on metal-water interfaces and charged vacancies in SiO₂. A unifying theoretical framework that maps all existing electrostatic MLIP approaches onto a common design space. Shows that current approaches fail on metal-water interfaces and charged vacancies — systems where self-consistency is essential.
Relevance to DENG.Group
From Gábor Csányi's group (creators of MACE). Directly relevant to Yanhao's work — provides a systematic framework for incorporating electrostatics into MLIPs using the MACE architecture. Addresses the same challenge as EquiEwald but from the charge-density coarse-graining perspective. Use the framework to design electrostatic extensions for the halide MLIPs (Yanhao). The charged vacancy benchmark is directly analogous to charged point defects in halide electrolytes (Yan Li / Mengke Li's work). Combined with EquiEwald, there are now two complementary approaches to long-range MLIPs worth benchmarking.
3. Revealing Hydroxide Ion Transport Mechanisms in Anion-Exchange Membranes from MLIP Simulations¶
Source: arXiv (cond-mat.mtrl-sci), Mar 14, 2026 · 📅 2026-03-14 · ↗ Open paper
Fine-tuned MLIPs enable large-scale MD simulations (>10 ns, >10 nm) of hydroxide ion transport in a commercial anion-exchange membrane. Water content controls the percolation transition from isolated clusters to connected hydrogen-bond networks. Fine-tuned MLIP for membrane systems. Large-scale MD at varying water content. Computed diffusion coefficients and activation energies, validated against experiment. First MLIP-based study of ion transport in commercial anion-exchange membranes at realistic length and time scales.
Relevance to DENG.Group
Methodologically relevant to Naibing Wu's polymer electrolyte work. Demonstrates the fine-tuned MLIP workflow for ion transport in a complex heterogeneous membrane — a system analogous to polymer electrolytes. The fine-tuning workflow for membrane systems is directly transferable to Naibing's polymer electrolyte work. The percolation analysis parallels what's needed for understanding Li⁺ transport in PEO-based and composite polymer electrolytes.
4. GPUMDkit: A User-Friendly Toolkit for GPUMD and NEP¶
Source: arXiv (cond-mat.mtrl-sci), Mar 18, 2026 · 📅 2026-03-18 · ↗ Open paper
Comprehensive toolkit that automates format conversion, structure sampling, active learning, property calculation, and data visualization for GPUMD and NEP. Modular Python toolkit wrapping GPUMD and NEP workflows. Lowers the barrier to using GPUMD/NEP for MLIP development and MD simulation. Active learning integration could save significant setup time.
Relevance to DENG.Group
Practical tool for the group. GPUMD/NEP is one of the fastest MLIP MD engines available. GPUMDkit streamlines the entire workflow from training data preparation to trajectory analysis. Tim (research engineer) could evaluate GPUMDkit as an alternative/complement to LAMMPS for MLIP-driven MD. Benchmarking NEP vs MACE for halide electrolyte ionic conductivity predictions would be informative.