arXiv subjects

Chi Ding

Publications and source records attributed to Chi Ding.

At least 19 recordsLinked to original sources

arXiv2026

Cartesian tensor equivariant machine-learning force field for spin-dependent atomistic simulations

Magnetic materials exhibit an intricate coupling between atomic structure and spin degrees of freedom, posing a fundamental challenge for atomistic simulations across experimentally relevant length and time scales. Here we introduce HotPP-Spin, a spin-dependent extension of HotPP for magnetic machine learning interatomic potentials, built on Cartesian tensor equivariant message passing. Atomic magnetic moments are treated as explicit axial-vector degrees of freedom, while spatial-inversion and time-reversal parities are propagated through the tensor couplings. This construction provides a unified representation of exchange-dominated and spin-orbit-induced interactions without imposing predefined analytical interaction forms. A scalar spin-dependent potential energy surface yields energy-conserving atomic forces and magnetic effective fields through differentiation. Benchmarks spanning collinear magnetism, noncollinear magnetism, and spin-orbit-coupling-induced magnetic anisotropy show that HotPP-Spin accurately describes magnetic energy landscapes, magnetic forces, and magnetic-order-dependent energy-volume relations within the same general framework. For H-phase monolayer VSe\(_2\), stochastic spin-dynamics simulations using the learned magnetic effective fields locate the finite-size magnetic ordering crossover at 415--435~K, in close numerical agreement with the reported experimental value of \(418.5\pm7.8\)~K. These results establish Cartesian tensor message passing as a general route for connecting first-principles magnetic energetics with large-scale atomistic simulations of coupled structural and spin phenomena.

physics.comp-ph↗

arXiv2026

High-dimensional Multi-objective Bayesian Optimization with Learned Variable Interactions

Multi-objective Bayesian optimization (MOBO) is effective in identifying the Pareto fronts for expensive black-box problems. However, most current MOBO approaches are limited to low-dimensional decision space due to its exponential sampling complexity. This paper presents decision variable interaction analysis-based MOBO, ViaMOBO, a generic framework for expensive multi-objective problems with high-dimensional decision space. The key idea of ViaMOBO is that it utilizes a variable interaction analysis model to determine whether the decision space can be completely or partially divided, and then performs local Bayesian optimization in the divided decision subspaces. Through the variable analysis model, it can be derived whether the objectives in black-box problems are separable, partially separable, or non-separable based on the potential independent or interdependent relationships among decision variables without any strong assumptions. We compare ViaMOBO with the state-of-the-art MOBO methods on both synthetic and real-world benchmarks. The experimental results demonstrate that ViaMOBO outperforms other related MOBO baselines in approximating the Pareto front of high-dimensional expensive multi-objective problems.

cs.LG↗

arXiv2026

Large-eddy simulation of the FDA benchmark blood pump: validation against experiments and implications for turbulent flow mechanisms

This study presents a systematic validation and comparative assessment of computational fluid dynamics (CFD) strategies for centrifugal blood pump simulations using the U.S. Food and Drug Administration benchmark model. A scale-resolving large eddy simulation (LES) with transient sliding-interface (SI) coupling is evaluated and compared against Reynolds-averaged Navier-Stokes (RANS) approaches employing both multiple reference frame and SI formulations. Numerical predictions are validated through direct comparison with particle image velocimetry measurements under two representative operating conditions. The results indicate that LES with transient rotor-stator coupling achieves consistently improved agreement with experimental velocity fields compared with RANS-based methods, particularly in the diffuser region where strong intermittency and wall-bounded turbulence are present. In contrast, RANS-based approaches exhibit noticeable discrepancies in these regions. A mesh sensitivity study and an assessment of temporal averaging effects are conducted for LES. The quality of the LES results is further quantified using three complementary metrics, demonstrating that a mesh resolution of approximately 80 million cells achieves a well-resolved LES regime. Building on the validated scale-resolving simulations, detailed analyses of vortical structures, turbulent kinetic energy distributions, and velocity energy spectra are performed to characterize the internal flow physics of the pump. This study demonstrates that scale-resolving, transient simulation approaches are essential for accurately capturing the highly unsteady, turbulence-dominated flow features in ventricular assist devices and provides practical guidance for future high-fidelity hemodynamic and hemocompatibility studies.

physics.flu-dyn↗

arXiv2026

NEPMaker: Active learning of neuroevolution machine learning potential for large cells

Machine learning potentials (MLPs) achieve near first-principles accuracy but often fail for atomic environments outside the training distribution. Active learning can mitigate this limitation; however, its application to large-scale simulations is hindered by the prohibitive cost of labeling entire configurations. Here, we develop a D-optimality-driven active learning framework for the neuroevolution potential (NEP) implemented within the GPUMD package, named NEPMaker. Extrapolative atomic environments are identified on-the-fly and embedded into locally periodic structures, where boundary atoms are optimized to remain close to the training distribution. This strategy enables large-scale simulations to directly contribute to dataset construction, significantly reducing extrapolation errors while improving model robustness and transferability. The proposed framework provides a scalable route for constructing reliable machine learning potentials in complex materials systems, including those involving defects, interfaces, and phase transitions.

physics.comp-ph↗

arXiv2026

Transferable Optimization Network for Cross-Domain Image Reconstruction

We develop a novel transfer learning framework to tackle the challenge of limited training data in image reconstruction problems. The proposed framework consists of two training steps, both of which are formed as bi-level optimizations. In the first step, we train a powerful universal feature-extractor that is capable of learning important knowledge from large, heterogeneous data sets in various domains. In the second step, we train a task-specific domain-adapter for a new target domain or task with only a limited amount of data available for training. Then the composition of the adapter and the universal feature-extractor effectively explores feature which serve as an important component of image regularization for the new domains, and this leads to high-quality reconstruction despite the data limitation issue. We apply this framework to reconstruct under-sampled MR images with limited data by using a collection of diverse data samples from different domains, such as images of other anatomies, measurements of various sampling ratios, and even different image modalities, including natural images. Experimental results demonstrate a promising transfer learning capability of the proposed method.

cs.CV↗

arXiv2026

GPUTB-2:An efficient E(3) network method for learning high-precision orthogonal Hamiltonian

Although equivariant neural networks have become a cornerstone for learning electronic Hamiltonians, the intrinsic non-orthogonality of linear combinations of atomic orbitals (LCAO) basis sets poses a fundamental challenge. The computational cost of Hamiltonian orthogonalization scales as O(N^3), which severely hinders electronic structure calculations for large-scale systems containing hundreds of thousands to millions of atoms. To address this issue, we develop GPUTB-2, a framework that learns implicitly orthogonality-preserving Hamiltonians by training directly on electronic band structures. Benefiting from an E(3)-equivariant network accelerated by Gaunt tensor product and SO(2) tensor product layers, GPUTB-2 achieves significantly higher accuracy than GPUTB across multiple benchmark systems. Moreover, GPUTB-2 accurately predicts large-scale electronic structures, including transport properties of temperature-perturbed SnSe and the band structures of magic-angle twisted bilayer graphene. By further integrating this framework with the linear-scaling quantum transport (LSQT) method, we investigate the electronic properties of million-atom amorphous graphene and uncover pressure-induced electronic structure transitions in more complex amorphous silicon. Together, these results establish GPUTB-2 as a high-accuracy and scalable approach for predicting orthogonal Hamiltonians.

cond-mat.mtrl-sci↗

arXiv2026

Disentangling Hardness from Noise: An Uncertainty-Driven Model-Agnostic Framework for Long-Tailed Remote Sensing Classification

Long-Tailed distributions are pervasive in remote sensing due to the inherently imbalanced occurrence of grounded objects. However, a critical challenge remains largely overlooked, i.e., disentangling hard tail data samples from noisy ambiguous ones. Conventional methods often indiscriminately emphasize all low-confidence samples, leading to overfitting on noisy data. To bridge this gap, building upon Evidential Deep Learning, we propose a model-agnostic uncertainty-aware framework termed DUAL, which dynamically disentangles prediction uncertainty into Epistemic Uncertainty (EU) and Aleatoric Uncertainty (AU). Specifically, we introduce EU as an indicator of sample scarcity to guide a reweighting strategy for hard-to-learn tail samples, while leveraging AU to quantify data ambiguity, employing an adaptive label smoothing mechanism to suppress the impact of noise. Extensive experiments on multiple datasets across various backbones demonstrate the effectiveness and generalization of our framework, surpassing strong baselines such as TGN and SADE. Ablation studies provide further insights into the crucial choices of our design.

cs.CV↗

arXiv2025

Half-explicit Runge-Kutta integrators for variational multiscale turbulence modeling: Toward higher-order accuracy in space and time

The residual-based variational multiscale (VMS) formulation has achieved remarkable success in large-eddy simulation of turbulent flows. However, its temporal discretization has largely remained limited to second-order implicit schemes. The present work aims at advancing this direction through the introduction of Runge-Kutta (RK) schemes within the VMS framework in a mathematically consistent manner. Guided by the Rothe method, the half-explicit RK scheme is employed as its accuracy is theoretically guaranteed for index-2 differential-algebraic equations. Owing to the explicit treatment of the nonlinear term, the resulting spatial problem exhibits a structure analogous to that of the Darcy equation. Following the philosophy of the VMS analysis, a subgrid-scale model is derived without invoking linearization based on perturbation series and related assumptions. The analysis further reveals that the parameter in the subgrid model is independent of the spatial mesh size. Fourier analysis demonstrates that the Rothe method, compared with the conventional vertical method of lines, provides improved dissipation and dispersion properties and exhibits a larger stability region for convection-dominated regimes. In the Taylor-Green vortex benchmark, the proposed schemes demonstrate superior performance as a large-eddy simulation model, achieving higher fidelity in predicting the kinetic energy evolution, energy spectra, and vortex structures than the conventional VMS formulation. Simulations of the open cavity flow further show that the proposed schemes can accurately capture the periodic limit cycle caused by the supercritical Hopf bifurcation, confirming its effectiveness and fidelity for highly sensitive flow instability problems.

physics.flu-dyn↗

arXiv2025

GPUTB: Efficient Machine Learning Tight-Binding Method for Large-Scale Electronic Properties Calculations

The high computational cost of ab-initio methods limits their application in predicting electronic properties at the device scale. Therefore, an efficient method is needed to map the atomic structure to the electronic structure quickly. Here, we develop GPUTB, a GPU-accelerated tight-binding (TB) machine learning framework. GPUTB employs atomic environment descriptors, enabling the model parameters to incorporate environmental dependence. This allows the model to transfer to different basis, xc-functionals, and allotropes easily. Combined with the linear scaling quantum transport method, we have calculated the electronic density of states for up to 100 million atoms in pristine graphene. Trained on finite-temperature structures, the model can be easily extended to millions of atom finite-temperature systems. Furthermore, GPUTB can also successfully describe h-BN/graphene heterojunction systems, demonstrating its capability to handle complex material with high precision. We accurately reproduce the relationship between carrier concentration and room temperature mobility in graphene to verify the framework's accuracy. Therefore, our GPUTB framework presents a delicate balance between computational accuracy and efficiency, providing a powerful computational tool for investing electronic properties for large systems with millions of atoms.

cond-mat.mtrl-sci↗

arXiv2025

Hot-Ham: an accurate and efficient E(3)-equivariant machine-learning electronic structures calculation framework

The combinations of machine learning with ab initio methods have attracted much attention for their potential to resolve the accuracy-efficiency dilemma and facilitate calculations for large-scale systems. Recently, equivariant message passing neural networks (MPNNs) that explicitly incorporate symmetry constraints have demonstrated promise for interatomic potential and density functional theory (DFT) Hamiltonian predictions. However, the high-order tensors used to represent node and edge information are coupled through the Clebsch-Gordan tensor product (CGTP), leading to steep increases in computational complexity and seriously hindering the performance of equivariant MPNNs. Here, we develop High-order Tensor machine-learning Hamiltonian (Hot-Ham), an E(3) equivariant MPNN framework that combines two advanced technologies local coordinate transformation and Gaunt tensor product (GTP) to efficiently model DFT Hamiltonians. These two innovations significantly reduce the complexity of tensor products from O(L^6) to O(L^3) or O(L^2 log^2 L) for the max tensor order L, and enhance the performance of MPNNs. Benchmarks on several public datasets demonstrate its state-of-the-art accuracy with relatively few parameters, and the applications to multilayer twisted moir\'e systems, heterostructures and allotropes showcase its generalization ability and high efficiency. Our Hot-Ham method provides a new perspective for developing efficient equivariant neural networks and would be a promising approach for investigating the electronic properties of large-scale materials systems.

physics.comp-ph↗

arXiv2025

RubikSQL: Lifelong Learning Agentic Knowledge Base as an Industrial NL2SQL System

We present RubikSQL, a novel NL2SQL system designed to address key challenges in real-world enterprise-level NL2SQL, such as implicit intents and domain-specific terminology. RubikSQL frames NL2SQL as a lifelong learning task, demanding both Knowledge Base (KB) maintenance and SQL generation. RubikSQL systematically builds and refines its KB through techniques including database profiling, structured information extraction, agentic rule mining, and Chain-of-Thought (CoT)-enhanced SQL profiling. RubikSQL then employs a multi-agent workflow to leverage this curated KB, generating accurate SQLs. RubikSQL achieves SOTA performance on both the KaggleDBQA and BIRD Mini-Dev datasets. Finally, we release the RubikBench benchmark, a new benchmark specifically designed to capture vital traits of industrial NL2SQL scenarios, providing a valuable resource for future research.

cs.DB↗

arXiv2025

Trusted Fake Audio Detection Based on Dirichlet Distribution

With the continuous development of deep learning-based speech conversion and speech synthesis technologies, the cybersecurity problem posed by fake audio has become increasingly serious. Previously proposed models for defending against fake audio have attained remarkable performance. However, they all fall short in modeling the trustworthiness of the decisions made by the models themselves. Based on this, we put forward a plausible fake audio detection approach based on the Dirichlet distribution with the aim of enhancing the reliability of fake audio detection. Specifically, we first generate evidence through a neural network. Uncertainty is then modeled using the Dirichlet distribution. By modeling the belief distribution with the parameters of the Dirichlet distribution, an estimate of uncertainty can be obtained for each decision. Finally, the predicted probabilities and corresponding uncertainty estimates are combined to form the final opinion. On the ASVspoof series dataset (i.e., ASVspoof 2019 LA, ASVspoof 2021 LA, and DF), we conduct a number of comparison experiments to verify the excellent performance of the proposed model in terms of accuracy, robustness, and trustworthiness.

cs.SD↗

arXiv2025

Sketch-1-to-3: One Single Sketch to 3D Detailed Face Reconstruction

3D face reconstruction from a single sketch is a critical yet underexplored task with significant practical applications. The primary challenges stem from the substantial modality gap between 2D sketches and 3D facial structures, including: (1) accurately extracting facial keypoints from 2D sketches; (2) preserving diverse facial expressions and fine-grained texture details; and (3) training a high-performing model with limited data. In this paper, we propose Sketch-1-to-3, a novel framework for realistic 3D face reconstruction from a single sketch, to address these challenges. Specifically, we first introduce the Geometric Contour and Texture Detail (GCTD) module, which enhances the extraction of geometric contours and texture details from facial sketches. Additionally, we design a deep learning architecture with a domain adaptation module and a tailored loss function to align sketches with the 3D facial space, enabling high-fidelity expression and texture reconstruction. To facilitate evaluation and further research, we construct SketchFaces, a real hand-drawn facial sketch dataset, and Syn-SketchFaces, a synthetic facial sketch dataset. Extensive experiments demonstrate that Sketch-1-to-3 achieves state-of-the-art performance in sketch-based 3D face reconstruction.

cs.CV↗

arXiv2024

LAMA: Stable Dual-Domain Deep Reconstruction For Sparse-View CT

Inverse problems arise in many applications, especially tomographic imaging. We develop a Learned Alternating Minimization Algorithm (LAMA) to solve such problems via two-block optimization by synergizing data-driven and classical techniques with proven convergence. LAMA is naturally induced by a variational model with learnable regularizers in both data and image domains, parameterized as composite functions of neural networks trained with domain-specific data. We allow these regularizers to be nonconvex and nonsmooth to extract features from data effectively. We minimize the overall objective function using Nesterov's smoothing technique and residual learning architecture. It is demonstrated that LAMA reduces network complexity, improves memory efficiency, and enhances reconstruction accuracy, stability, and interpretability. Extensive experiments show that LAMA significantly outperforms state-of-the-art methods on popular benchmark datasets for Computed Tomography.

cs.CV↗

arXiv2024

Fast Deep Predictive Coding Networks for Videos Feature Extraction without Labels

Brain-inspired deep predictive coding networks (DPCNs) effectively model and capture video features through a bi-directional information flow, even without labels. They are based on an overcomplete description of video scenes, and one of the bottlenecks has been the lack of effective sparsification techniques to find discriminative and robust dictionaries. FISTA has been the best alternative. This paper proposes a DPCN with a fast inference of internal model variables (states and causes) that achieves high sparsity and accuracy of feature clustering. The proposed unsupervised learning procedure, inspired by adaptive dynamic programming with a majorization-minimization framework, and its convergence are rigorously analyzed. Experiments in the data sets CIFAR-10, Super Mario Bros video game, and Coil-100 validate the approach, which outperforms previous versions of DPCNs on learning rate, sparsity ratio, and feature clustering accuracy. Because of DCPN's solid foundation and explainability, this advance opens the door for general applications in object recognition in video without labels.

cs.CV↗

arXiv2023

Enhancement of superconducting transition temperature and exotic stoichiometries in Lu-S system under high pressure

Binary metal sulfides are potential material family for exploring high Tc superconductors under high pressure. In this work, we study the crystal structures, electronic structures and superconducting properties of the Lu-S system in the pressure range from 0 GPa to 200 GPa, combining crystal structure predictions with ab-initio calculations. We predict 14 new structures, encompassing 7 unidentified stoichiometries. Within the S-rich structures, the formation of S atom cages is beneficial for superconductivity, with the superconducting transition temperature 25.86 K and 25.30 K for LuS6-C2/m at 70 GPa and LuS6-R-3m at 90 GPa, respectively. With the Lu/(Lu+S) ratio increases, the Lu-d electrons participate more in the electronic properties at the Fermi energy, resulting in the coexistence of superconductivity and topological non-triviality of LuS2-Cmca, as well as the superconductivity of predicted Lu-rich compounds. Our calculation is helpful for understanding the exotic properties in transition metal sulfides system under high pressure, providing possibility in designing novel superconductors for future experimental and theoretical works.

cond-mat.supr-con↗

arXiv2023

Superconducting Ternary Hydridies in Ca-U-H under High Pressure

The research on hydrogen-rich ternary compounds attract tremendous attention for it paves new route to room-temperature superconductivity at lower pressures. Here, we study the crystal structures, electronic structures, and superconducting properties of the ternary Ca-U-H system, combining crystal structure predictions with ab-initio calculations under high pressure. We found four dynamically stable structures with hydrogen clathrate cages: CaUH12-Cmmm, CaUH12-Fd-3m, Ca2UH18-P-3m1, and CaU3H32-Pm-3m. Among them, the Ca2UH18-P-3m1 and CaU3H32-Pm-3m are likely to be synthesized below 1 megabar. The f electrons in U atoms make dominant contribution to the electronic density of states around the Fermi energy. The electron-phonon interaction calculations reveal that phonon softening in the mid-frequency region can enhance the electron-phonon coupling significantly. The Tc value of Ca2UH18-P-3m1 is estimated to be 57.5-65.8 K at 100 GPa. Our studies demonstrate that introducing actinides into alkaline-earth metal hydrides provides possibility in designing novel superconducting ternary hydrides.

cond-mat.supr-con↗

arXiv2023

Learned Alternating Minimization Algorithm for Dual-domain Sparse-View CT Reconstruction

We propose a novel Learned Alternating Minimization Algorithm (LAMA) for dual-domain sparse-view CT image reconstruction. LAMA is naturally induced by a variational model for CT reconstruction with learnable nonsmooth nonconvex regularizers, which are parameterized as composite functions of deep networks in both image and sinogram domains. To minimize the objective of the model, we incorporate the smoothing technique and residual learning architecture into the design of LAMA. We show that LAMA substantially reduces network complexity, improves memory efficiency and reconstruction accuracy, and is provably convergent for reliable reconstructions. Extensive numerical experiments demonstrate that LAMA outperforms existing methods by a wide margin on multiple benchmark CT datasets.

cs.CV↗