Searcharxiv⌕ Search

arXiv subjects

Zhennan Zhang

Publications and source records attributed to Zhennan Zhang.

3 recordsLinked to original sources

CGTransformer: A Topology-Aware Autoregressive Framework for Crystal Generation and Inverse Design

Autoregressive language models offer a unified route to crystal generation and inverse design, but their performance depends on how periodic structures are serialized. Symmetry equivalent and closely related crystals can receive different atomic orders under coordinate-based serialization, introducing variability unrelated to their structural connectivity. Here, we present CGTransformer (Crystal Graph Transformer), an autoregressive generative framework that addresses this challenge via a graph-embedded sorting strategy. Using quotientgraph topology and barycentric embeddings, Element Symmetry Graph (ESG) converts periodic crystal networks into topology-guided one-dimensional sequences that retain key local and symmetry-related structural information. CGTransformer achieves high generation validity and the highest single shot crystal structure prediction (CSP) match rate among the compared MP-20 baselines (63.24%), while same backbone comparisons show improved structure recovery over coordinate-based ordering. Furthermore, we demonstrate property-conditioned inverse design for wide-bandgap semiconductors, where ESG achieves a much higher stable, unique and novel (S.U.N.) rate and stronger target-property alignment than the coordinate-based baseline. These results establish topology-guided sequence construction as an effective route to improving crystal generation and property-directed design at fixed model capacity.

physics.comp-ph↗

High-order tensor neural network for iteration-free structure relaxation

Structure relaxation is important for the discovery of new materials, yet conventional ab initio optimization remains a major bottleneck in high-throughput screening workflows. Machine learning potentials have accelerated relaxation by orders of magnitude, but they still rely on iterative optimization and high-quality DFT force labels. Here, we present HotRelax, a high-order tensor message-passing neural network for one-shot, end-to-end prediction of relaxed structures. Trained directly on paired unrelaxed and relaxed structures, HotRelax requires no DFT force labels and predicts relaxed structures in a single forward pass, without iterative inference or post-processing. Across five diverse datasets spanning 3D bulk crystals, 2D layered materials and catalysts, HotRelax shows strong performance relative to state-of-the-art end-to-end relaxation models, achieving lower prediction errors on several benchmarks while maintaining a compact model size and efficient inference. Extensive DFT calculations further show that the predicted structures are close in energy to their DFT-relaxed counterparts. When integrated into catalytic workflows, HotRelax also improves the accuracy and generalization of relaxed-state energy prediction models. Together, these results support HotRelax as an efficient and widely applicable framework for end-to-end structure relaxation, with strong potential to accelerate high-throughput materials discovery.

physics.comp-ph↗

GPU-MetaD: Full-Life-Cycle GPU Accelerated Metadynamics with Machine Learning Potentials

Large-scale molecular dynamics simulations with high accuracy have been increasingly popular for their capability to bridge the gap between atomistic modeling and mesoscale phenomena. Both machine learning potentials and enhanced sampling approaches offer substantial improvements in high-accuracy simulation efficiency, which can be further boosted through GPU acceleration. However, an efficient framework combining these advances for extending simulations to large systems and long timescales remains elusive. In this work, we proposed a full-life-cycle GPU accelerated metadynamics simulations package GPU-MetaD. Benchmarking across molecular, interface, and bulk systems demonstrates that GPU-MetaD efficiently handles diverse atomic systems and delivers an order-of-magnitude performance improvement. Building on this demonstrated capability, it enables ab-initio-level rare-event sampling for systems comprising millions of atoms on a typical single GPU. This capability allows us to reveal a previously unknown size-dependent two-step nucleation mechanism in gallium nitride (GaN), highlighting the potential of GPU-MetaD for uncovering complex rare events in realistic large-scale materials systems.

physics.comp-ph↗