arXiv subjects

Wanrun Jiang

Publications and source records attributed to Wanrun Jiang.

13 recordsLinked to original sources

arXiv2025

LAMBench: A Benchmark for Large Atomistic Models

Large Atomistic Models (LAMs) have undergone remarkable progress recently, emerging as universal or fundamental representations of the potential energy surface defined by the first-principles calculations of atomistic systems. However, our understanding of the extent to which these models achieve true universality, as well as their comparative performance across different models, remains limited. This gap is largely due to the lack of comprehensive benchmarks capable of evaluating the effectiveness of LAMs as approximations to the universal potential energy surface. In this study, we introduce LAMBench, a benchmarking system designed to evaluate LAMs in terms of their generalizability, adaptability, and applicability. These attributes are crucial for deploying LAMs as ready-to-use tools across a diverse array of scientific discovery contexts. We benchmark ten state-of-the-art LAMs released prior to August 1, 2025, using LAMBench. Our findings reveal a significant gap between the current LAMs and the ideal universal potential energy surface. They also highlight the need for incorporating cross-domain training data, supporting multi-fidelity modeling, and ensuring the models' conservativeness and differentiability. As a dynamic and extensible platform, LAMBench is intended to continuously evolve, thereby facilitating the development of robust and generalizable LAMs capable of significantly advancing scientific research. The LAMBench code is open-sourced at https://github.com/deepmodeling/lambench, and an interactive leaderboard is available at https://www.aissquare.com/openlam?tab=Benchmark.

physics.comp-ph↗

arXiv2025

DeePMD-kit v3: A Multiple-Backend Framework for Machine Learning Potentials

In recent years, machine learning potentials (MLPs) have become indispensable tools in physics, chemistry, and materials science, driving the development of software packages for molecular dynamics (MD) simulations and related applications. These packages, typically built on specific machine learning frameworks such as TensorFlow, PyTorch, or JAX, face integration challenges when advanced applications demand communication across different frameworks. The previous TensorFlow-based implementation of DeePMD-kit exemplified these limitations. In this work, we introduce DeePMD-kit version 3, a significant update featuring a multi-backend framework that supports TensorFlow, PyTorch, JAX, and PaddlePaddle backends, and demonstrate the versatility of this architecture through the integration of other MLPs packages and of Differentiable Molecular Force Field. This architecture allows seamless backend switching with minimal modifications, enabling users and developers to integrate DeePMD-kit with other packages using different machine learning frameworks. This innovation facilitates the development of more complex and interoperable workflows, paving the way for broader applications of MLPs in scientific research.

physics.chem-ph↗

arXiv2023

DPA-2: a large atomic model as a multi-task learner

The rapid advancements in artificial intelligence (AI) are catalyzing transformative changes in atomic modeling, simulation, and design. AI-driven potential energy models have demonstrated the capability to conduct large-scale, long-duration simulations with the accuracy of ab initio electronic structure methods. However, the model generation process remains a bottleneck for large-scale applications. We propose a shift towards a model-centric ecosystem, wherein a large atomic model (LAM), pre-trained across multiple disciplines, can be efficiently fine-tuned and distilled for various downstream tasks, thereby establishing a new framework for molecular modeling. In this study, we introduce the DPA-2 architecture as a prototype for LAMs. Pre-trained on a diverse array of chemical and materials systems using a multi-task approach, DPA-2 demonstrates superior generalization capabilities across multiple downstream tasks compared to the traditional single-task pre-training and fine-tuning methodologies. Our approach sets the stage for the development and broad application of LAMs in molecular and materials simulation research.

physics.chem-ph↗

arXiv2022

DPA-1: Pretraining of Attention-based Deep Potential Model for Molecular Simulation

Machine learning assisted modeling of the inter-atomic potential energy surface (PES) is revolutionizing the field of molecular simulation. With the accumulation of high-quality electronic structure data, a model that can be pretrained on all available data and finetuned on downstream tasks with a small additional effort would bring the field to a new stage. Here we propose DPA-1, a Deep Potential model with a novel attention mechanism, which is highly effective for representing the conformation and chemical spaces of atomic systems and learning the PES. We tested DPA-1 on a number of systems and observed superior performance compared with existing benchmarks. When pretrained on large-scale datasets containing 56 elements, DPA-1 can be successfully applied to various downstream tasks with a great improvement of sample efficiency. Surprisingly, for different elements, the learned type embedding parameters form a $spiral$ in the latent space and have a natural correspondence with their positions on the periodic table, showing interesting interpretability of the pretrained DPA-1 model.

physics.chem-ph↗

arXiv2022

Extending the limit of molecular dynamics with ab initio accuracy to 10 billion atoms

High-performance computing, together with a neural network model trained from data generated with first-principles methods, has greatly boosted applications of \textit{ab initio} molecular dynamics in terms of spatial and temporal scales on modern supercomputers. Previous state-of-the-art can achieve $1-2$ nanoseconds molecular dynamics simulation per day for 100-million atoms on the entire Summit supercomputer. In this paper, we have significantly reduced the memory footprint and computational time by a comprehensive approach with both algorithmic and system innovations. The neural network model is compressed by model tabulation, kernel fusion, and redundancy removal. Then optimizations such as acceleration of customized kernel, tabulation of activation function, MPI+OpenMP parallelization are implemented on GPU and ARM architectures. Testing results of the copper system show that the optimized code can scale up to the entire machine of both Fugaku and Summit, and the corresponding system size can be extended by a factor of $134$ to an unprecedented $17$ billion atoms. The strong scaling of a $13.5$-million atom copper system shows that the time-to-solution can be 7 times faster, reaching $11.2$ nanoseconds per day. This work opens the door for unprecedentedly large-scale molecular dynamics simulations based on {\it ab initio} accuracy and can be potentially utilized in studying more realistic applications such as mechanical properties of metals, semiconductor devices, batteries, etc. The optimization techniques detailed in this paper also provide insight for relevant high-performance computing applications.

cs.DC↗

arXiv2021

DP Compress: a Model Compression Scheme for Generating Efficient Deep Potential Models

Machine-learning-based interatomic potential energy surface (PES) models are revolutionizing the field of molecular modeling. However, although much faster than electronic structure schemes, these models suffer from costly computations via deep neural networks to predict the energy and atomic forces, resulting in lower running efficiency as compared to the typical empirical force fields. Herein, we report a model compression scheme for boosting the performance of the Deep Potential (DP) model, a deep learning based PES model. This scheme, we call DP Compress, is an efficient post-processing step after the training of DP models (DP Train). DP Compress combines several DP-specific compression techniques, which typically speed up DP-based molecular dynamics simulations by an order of magnitude faster, and consume an order of magnitude less memory. We demonstrate that DP Compress is sufficiently accurate by testing a variety of physical properties of Cu, H2O, and Al-Cu-Mg systems. DP Compress applies to both CPU and GPU machines and is publicly available online.

physics.comp-ph↗

arXiv2021

Accurate Deep Potential model for the Al-Cu-Mg alloy in the full concentration space

Combining first-principles accuracy and empirical-potential efficiency for the description of the potential energy surface (PES) is the philosopher's stone for unraveling the nature of matter via atomistic simulation. This has been particularly challenging for multi-component alloy systems due to the complex and non-linear nature of the associated PES. In this work, we develop an accurate PES model for the Al-Cu-Mg system by employing Deep Potential (DP), a neural network based representation of the PES, and DP Generator (DP-GEN), a concurrent-learning scheme that generates a compact set of ab initio data for training. The resulting DP model gives predictions consistent with first-principles calculations for various binary and ternary systems on their fundamental energetic and mechanical properties, including formation energy, equilibrium volume, equation of state, interstitial energy, vacancy and surface formation energy, as well as elastic moduli. Extensive benchmark shows that the DP model is ready and will be useful for atomistic modeling of the Al-Cu-Mg system within the full range of concentration.

cond-mat.mtrl-sci↗

arXiv2018

Electronic quantum effect on hydrogen bond geometry in water dimer

During compression of a water dimer calculated with high-precision first-principles methods, the trends of H-bond and O-H bond lengths show quantum effect of the electronic structure. We found that the H-bond length keeps decreasing, while the O-H bond length increases up to the stable point and decreases beyond it when the water dimer is further compressed. The remarkable properties are different from those observed in most previous researches which can be understood and extrapolated through classical simulation. The observations can be explained by the decrease in orbital overlap and change in the exchange repulsion interaction between water monomers. The dominant interaction between water monomers changes from electrostatic interaction to exchange repulsion at the turning point of the O-H bond length when the O...O distance is decreased. These findings highlight the quantum effect on the hydrogen bond in extreme conditions and play an important role in the recognition of the hydrogen bond structure and mechanism.

physics.chem-ph↗

arXiv2017

Spin Polarization and Dispersion Effects in Emergence of Roaming Transition State for Nitrobenzene Isomerization

Since roaming was found as a new but common reaction path of isomerization, many of its properties, especially those of roaming transition state (TSR), have been studied recently on many systems. However, the mechanism of roaming is still not clear at the atomic level. In this work, we used first-principles calculations to illustrate the detailed structure of TSR in an internal isomerization process of nitrobenzene. The calculations distinctively show its nature of antiferromagnetic coupling between two roaming fragments. Moreover, the effect of dispersion is also revealed as an important issue for the stability of the TSR. Our work provides a new insight from the view of electronic structure towards the TSR and contributes to the basic understanding of the roaming systems.

physics.atm-clus↗

arXiv2017

The Synthesis and Electrical Transport of Ligand-Protected Au13 Clusters

The ligand-protected Au13 clusters have been synthesized by using meso-2,3-imercaptosuccinic acid as the reducing and stabilizing agent. Transmission electron microscopic analysis shows a size distribution of 0.6nm. Optical spectrum shows an absorbance peak at 390 nm. The electrical transport measurement devices are fabricated using the electro-migration method. Coulomb blockade is observed at the temperature of 1.6 K,revealing the formation of the tunneling junction. The Coulomb oscillation on-off ratio is nearly 5. Three peaks are extracted in the dI-dV data and attributed to the energy levels of Au13 clusters, gapped by about 60 meV. First principle calculations are carried out to interpret the energy diagram.

cond-mat.mes-hall↗

arXiv2016

Electronic structure contribution to hydrogen bonding interaction of a water dimer

Hydrogen bond (H-bond) covalency has recently been observed in ice and liquid water, while the penetrating molecular orbitals (MOs) in the H-bond region of most typical water dimer system, (H2O)2, have also been discovered. However, obtaining the quantitative contribution of these MOs to the H-bond interaction is still problematic. In this work, we introduced the orbital-resolved electron density projected integral (EDPI) along the H-bond to approach this problem. The calculations show that, surprisingly, the electronic occupied orbital (HOMO-4) of (H2O)2 accounts for about 40% of the electron density at the bond critical point. Moreover, the charge transfer analysis visualizes the electron accumulating effect of the orbital interaction within the H-bond between water molecules, supporting its covalent-like character. Our work expands the classical understanding of H-bond with specific contributions from certain MOs, and will also advance further research into such covalency and offer quantitative electronic structure insights into intermolecular systems.

physics.atm-clus↗

arXiv2016

The self-consistent charge density functional tight-binding (DFTB) theory study of carbon adatoms using tuned Hubbard U parameters

The self-consistent charge density functional tight-binding (DFTB) theory is a useful tool for realizing the electronic structures of large molecular complex systems. In this study, we analyze the electronic structure of C61, formed by fullerene C60 with a carbon adatom, using the fully localized limit and pseudo self-interaction correction methods of DFTB to adjust the Hubbard U parameter (DFTB+U). The results show that both the methods used to adjust U can significantly reduce the molecular orbital energy of occupied states localized on the defect carbon atom and improve the gap between highest occupied molecular orbital(HOMO) and lowest unoccupied molecular orbital(LUMO) of C61. This work will provide a methodological reference point for future DFTB calculations of the electronic structures of carbon materials.

cond-mat.mtrl-sci↗

arXiv2015

Covalent features in the hydrogen bond of a water dimer: molecular orbital analysis

The covalent-like characteristics of hydrogen bonds offer a new perspective on intermolecular interactions. Here, using density functional theory and post-Hartree-Fock methods, we reveal that there are two bonding molecular orbitals (MOs) crossing the O and H atoms of the hydrogen-bond in water dimer. Energy decomposition analysis also shows a non-negligible contribution of the induction term. These results illustrate the covalent-like character of the hydrogen bond between water molecules, which contributes to the essential understanding of ice, liquid water, related materials, and life sciences.

physics.atm-clus↗