arXiv subjects

Shiwei Zhu

Publications and source records attributed to Shiwei Zhu.

9 recordsLinked to original sources

arXiv2026

Beating micromagnetic limits on skyrmion stability by long-range frustration

Skyrmion stability is commonly assumed to scale with skyrmion size or exchange stiffness within micromagnetic models. Here, we demonstrate that long-range exchange frustration can break this paradigm, enhancing the collapse energy barrier without increasing skyrmion size or magnetic energy scale. By mapping the continuum model onto a spin-lattice Hamiltonian, we find that skyrmions with identical micromagnetic parameters can exhibit significantly different energy barriers, depending on their underlying atomistic exchange interactions. We attribute this behavior to saddle point textures, whose pronounced noncollinearity captures long-range frustration beyond the micromagnetic approximation. We further develop an exchange optimization framework to predict that long-range frustration can double the energy barrier in physically realistic conditions, possibly valid for ultrathin films or van der Waals magnets. These results hold across different lattice symmetries, revealing an intrinsic limitation of micromagnetics and establishing long-range frustration engineering as a promising route toward highly stable nanoscale skyrmions.

cond-mat.mes-hall↗

arXiv2026

Robustness Analysis of USmorph: II. Optimizing Feature Extraction, Dimensionality Reduction, and Clustering for Unsupervised Galaxy Morphology Classification

We conduct a systematic robustness analysis of the unsupervised machine learning module within the hybrid framework \texttt{USmorph}. This module automatically discovers morphological structures from large-scale galaxy images, forming the foundation of the complete classification workflow. We evaluate five pre-trained models for feature extraction and identify an ImageNet-pretrained AlexNet as the most effective for capturing discriminative morphological features. UMAP is chosen for dimensionality reduction due to its optimal balance between preserving high-dimensional structure and computational efficiency. To enhance clustering stability, we propose a Bagging-based multi-cluster voting scheme, which significantly improves label consistency and cluster purity. We compare the convergence, scalability, and quality of five clustering algorithms, finding that the Bagging voting scheme has the best performance with the combination of K-means, Birch, and Agg. A bagging clustering number of $K=16$ is used to achieve the optimal balance between classification granularity and manual validation efficiency. Our tests show that: (1) the t-distributed stochastic neighbor embedding (t-SNE) reveals clear, compact cluster boundaries in low-dimensional space with strong feature separability; (2) the morphology classification results align with galaxy evolution theory, showing physically plausible distributions of different types in parameter space. These results demonstrate the technical robustness and scientific credibility of \texttt{USmorph}, establishing it as a reliable method for automated morphological classification in future large-scale surveys such as the China Space Station Telescope (CSST) mission.

astro-ph.GA↗

arXiv2026

Extended saddle points govern long-lived antiskyrmions

Achieving long-lived nanoscale magnetic solitons remains a central challenge, as their lifetimes typically decrease rapidly with temperature. Here, we demonstrate that anisotropic Dzyaloshinskii-Moriya interaction (aDMI) enables spatially extended saddle points (SPs) that fundamentally alter thermally activated decay. In contrast to conventional localized SPs, these extended configurations completely suppress the entropic contribution to the activation rate, rendering the lifetimes effectively temperature independent. To establish this mechanism, we develop a first-principles method based on spin spirals to compute DMI beyond the isotropic approximation, resolving its full directional dependence for arbitrary nearest neighbors. We apply this method to oxidized Fe$_3$GeTe$_2$ (FGT-O), an experimentally accessible van der Waals magnet. Oxygen adsorption simultaneously breaks inversion symmetry and lowers the in-plane crystalline symmetry, thereby generating a sizable aDMI. We demonstrate that aDMI stabilizes nanoscale antiskyrmions with energy barriers exceeding 120 meV at low external magnetic fields. Crucially, extended SPs enhance the lifetime in FGT-O by more than five orders of magnitude at room temperature compared to conventional ultrathin-film skyrmion systems. We further show that aDMI is not the only route to such extended SPs and identify the general conditions under which they emerge, establishing a general route to soliton decay pathways with temperature-independent prefactors. Our results uncover a new paradigm for enhancing soliton stability through transition-state geometry rather than energy-barrier height.

cond-mat.mes-hall↗

arXiv2025

Dual-coding contrastive learning based on ConvNeXt and ViT models for morphological classification of galaxies in COSMOS-Web

In our previous works, we proposed a machine learning framework named \texttt{USmorph} for efficiently classifying galaxy morphology. In this study, we propose a self-supervised method called contrastive learning to upgrade the unsupervised machine learning (UML) part of the \texttt{USmorph} framework, aiming to improve the efficiency of feature extraction in this step. The upgraded UML method primarily consists of the following three aspects. (1) We employ a Convolutional Autoencoder to denoise galaxy images and the Adaptive Polar Coordinate Transformation to enhance the model's rotational invariance. (2) A pre-trained dual-encoder convolutional neural network based on ConvNeXt and ViT is used to encode the image data, while contrastive learning is then applied to reduce the dimension of the features. (3) We adopt a Bagging-based clustering model to cluster galaxies with similar features into distinct groups. By carefully dividing the redshift bins, we apply this model to the rest-frame optical images of galaxies in the COSMOS-Web field within the redshift range of $0.5 < z < 6.0$. Compared to the previous algorithm, the improved UML method successfully classifies 73\% galaxies. Using the GoogleNet algorithm, we classify the morphology of the remaining 27\% galaxies. To validate the reliability of our updated algorithm, we compared our classification results with other galaxy morphological parameters and found a good consistency with galaxy evolution. Benefiting from its higher efficiency, this updated algorithm is well-suited for application in future China Space Station Telescope missions.

astro-ph.GA↗

arXiv2025

An updated efficient galaxy morphology classification model based on ConvNeXt encoding with UMAP dimensionality reduction

We present an enhanced unsupervised machine learning (UML) module within our previous \texttt{USmorph} classification framework featuring two components: (1) hierarchical feature extraction via a pre-trained ConvNeXt convolutional neural network (CNN) with transfer learning, and (2) nonlinear manifold learning using Uniform Manifold Approximation and Projection (UMAP) for topology-aware dimensionality reduction. This dual-stage design enables efficient knowledge transfer from large-scale visual datasets while preserving morphological pattern geometry through UMAP's neighborhood preservation. We apply the upgraded UML on I-band images of 99,806 COSMOS galaxies at redshift $0.2 10^{9}~M_{\odot}$. Our classification results align well with galaxy evolution theory. This improved algorithm significantly enhances galaxy morphology classification efficiency, making it suitable for large-scale sky surveys such as those planned with the China Space Station Telescope (CSST).

astro-ph.GA↗

arXiv2025

Robustness Analysis of USmorph: I. Generalization Efficiency of Unsupervised Strategies and Supervised Learning in Galaxy Morphological Classification

We conduct a systematic robustness analysis of the hybrid machine learning framework \texttt{USmorph}, which integrates unsupervised and supervised learning for galaxy morphological classification. Although \texttt{USmorph} has already been applied to nearly 100,000 $I$-band galaxy images in the COSMOS field ($0.2 < z < 1.2$, $I_{\mathrm{mag}} < 25$), the stability of its core modules has not been quantitatively assessed. Our tests show that the convolutional autoencoder (CAE) achieves the best performance in preserving structural information when adopting an intermediate network depth, $5\times5$ convolutional kernels, and a 40-dimensional latent representation. The adaptive polar coordinate transform (APCT) effectively enhances rotational invariance and improves the robustness of downstream tasks. In the unsupervised stage, a bagging clustering number of $K=50$ provides the optimal trade-off between classification granularity and labeling efficiency. For supervised learning, we employ GoogLeNet, which exhibits stable performance without overfitting. We validate the reliability of the final classifications through two independent tests: (1) the t-distributed stochastic neighbor embedding (t-SNE) visualization reveals clear clustering boundaries in the low-dimensional space; and (2) the morphological classifications are consistent with theoretical expectations of galaxy evolution, with both true and false positives showing unbiased distributions in the parameter space. These results demonstrate the strong robustness of the \texttt{USmorph} algorithm, providing guidance for its future application to the China Space Station Telescope (CSST) mission.

astro-ph.GA↗

arXiv2025

Universal two-stage dynamics and phase control in skyrmion formation

We uncover a universal two-stage dynamics during skyrmion formation and establish its connection to equilibrium phases through the introduction of a chiral correlation $\chi$. Stage I involves stripe coarsening governed by the exchange-to-DMI ratio $J'$, while stage II entails stripe contraction driven by the synergy between $J'$ and the anisotropy-to-DMI ratio $K'$. The magnetic field-to-DMI ratio $B'$ influences both stages. By combining symbolic regression with neural networks, we model the competition and cooperation among these parameters and derive a skyrmion formation criterion, $0.58 K'J' + \mu B'J' > 1$. Our model disentangles their distinct roles: $J'$ sets the stripe width, $K'$ primarily controls the skyrmion size, and $B'$ strongly affects the topological charge. This approach provides a general framework for predicting and controlling magnetic phases in chiral magnets.

cond-mat.mes-hall↗

arXiv2025

Strongly enhanced lifetime of higher-order bimerons and antibimerons

Magnetic bimerons, similar to skyrmions, are topologically nontrivial spin textures characterized by topological charge $Q$. Most studies so far have focused on low-$Q$ solitons ($|Q| \leq 1$), such as skyrmions, bimerons, and vortices. Here, we present the first calculations of the lifetimes of {ring-like} high-$Q$ bimerons and demonstrate that they are fundamentally more stable than high-$Q$ skyrmions over a wide range of temperature. To obtain realistic results, our chosen system is an experimentally feasible van der Waals interface, Fe$_3$GeTe$_2$/Cr$_2$Ge$_2$Te$_6$. We show that the lifetimes of high-$Q$ (anti)bimerons can exceed the lifetime of those with $|Q|=1$ by 3 orders of magnitude. Remarkably, this trend remains valid even when extrapolated to room temperature (RT), as the lifetimes are dominated by entropy rather than energy barriers. This contrasts with high-$Q$ skyrmions, whose lifetimes fall with $|Q|$ near RT. We attribute this fundamental difference between skyrmions and bimerons to their distinct magnetic texture symmetries, which lead to different entropy-dominated lifetimes.

cond-mat.mes-hall↗

arXiv2025

Prompt Mechanisms in Medical Imaging: A Comprehensive Survey

Deep learning offers transformative potential in medical imaging, yet its clinical adoption is frequently hampered by challenges such as data scarcity, distribution shifts, and the need for robust task generalization. Prompt-based methodologies have emerged as a pivotal strategy to guide deep learning models, providing flexible, domain-specific adaptations that significantly enhance model performance and adaptability without extensive retraining. This systematic review critically examines the burgeoning landscape of prompt engineering in medical imaging. We dissect diverse prompt modalities, including textual instructions, visual prompts, and learnable embeddings, and analyze their integration for core tasks such as image generation, segmentation, and classification. Our synthesis reveals how these mechanisms improve task-specific outcomes by enhancing accuracy, robustness, and data efficiency and reducing reliance on manual feature engineering while fostering greater model interpretability by making the model's guidance explicit. Despite substantial advancements, we identify persistent challenges, particularly in prompt design optimization, data heterogeneity, and ensuring scalability for clinical deployment. Finally, this review outlines promising future trajectories, including advanced multimodal prompting and robust clinical integration, underscoring the critical role of prompt-driven AI in accelerating the revolution of diagnostics and personalized treatment planning in medicine.

eess.IV↗