arXiv subjects

Wenjun Liu

Publications and source records attributed to Wenjun Liu.

At least 37 records · Page 2Linked to original sources

arXiv2024

BugBlitz-AI: An Intelligent QA Assistant

The evolution of software testing from manual to automated methods has significantly influenced quality assurance (QA) practices. However, challenges persist in post-execution phases, particularly in result analysis and reporting. Traditional post-execution validation phases require manual intervention for result analysis and report generation, leading to inefficiencies and potential development cycle delays. This paper introduces BugBlitz-AI, an AI-powered validation toolkit designed to enhance end-to-end test automation by automating result analysis and bug reporting processes. BugBlitz-AI leverages recent advancements in artificial intelligence to reduce the time-intensive tasks of manual result analysis and report generation, allowing QA teams to focus more on crucial aspects of product quality. By adopting BugBlitz-AI, organizations can advance automated testing practices and integrate AI into QA processes, ensuring higher product quality and faster time-to-market. The paper outlines BugBlitz-AI's architecture, discusses related work, details its quality enhancement strategies, and presents results demonstrating its effectiveness in real-world scenarios.

cs.SE↗

arXiv2024

Extremely long transverse optical needle focus for reflective metalens enabled by monolayer MoS$_2$

Line-scan mode facilitates fast-speed and high-throughput imaging with developing a suitable optical transverse needle focus. Metasurface with periodic structures such as diffractive rings, ellipses, and gratings could enable discrete focus evolving into line focus under momentum conservation, but still face the challenge of extremely low light power utilization brought by inevitably multiple high-order diffractions. In addition, the designed focus requires the selection of particular optical functional materials. High dielectric constants in atomic transition metal dichalcogenides make significant phase modulation by bringing phase singularity at zero-reflection possible. However, no light power is available for use at zero-reflection and a balance between phase and amplitude modulation is needed. In this work, above issues are simultaneously solved by designing a monolayer MoS2 based Fresnel strip structure. An optical needle primary focus with a transverse length of 40 μm (~80 λ) is obtained, which is the longest value recorded so far, together with a sub-diffraction-limited lateral spot and a broad working wavelength range. This specially developed structure not only concentrates light power in primary diffraction by breaking restriction of momentum conservation, but also guarantees a consistent phase across different strips. The novel optical manipulation way provided here together with the longer focus length for flat optics will show promising applications in biology, oncology, nanofabrication, energy harvesting, and optical information processing.

physics.optics↗

arXiv2024

Toward Short-Term Glucose Prediction Solely Based on CGM Time Series

The global diabetes epidemic highlights the importance of maintaining good glycemic control. Glucose prediction is a fundamental aspect of diabetes management, facilitating real-time decision-making. Recent research has introduced models focusing on long-term glucose trend prediction, which are unsuitable for real-time decision-making and result in delayed responses. Conversely, models designed to respond to immediate glucose level changes cannot analyze glucose variability comprehensively. Moreover, contemporary research generally integrates various physiological parameters (e.g. insulin doses, food intake, etc.), which inevitably raises data privacy concerns. To bridge such a research gap, we propose TimeGlu -- an end-to-end pipeline for short-term glucose prediction solely based on CGM time series data. We implement four baseline methods to conduct a comprehensive comparative analysis of the model's performance. Through extensive experiments on two contrasting datasets (CGM Glucose and Colas dataset), TimeGlu achieves state-of-the-art performance without the need for additional personal data from patients, providing effective guidance for real-world diabetic glucose management.

cs.AI↗

arXiv2024

Accurate and Data-Efficient Micro-XRD Phase Identification Using Multi-Task Learning: Application to Hydrothermal Fluids

Traditional analysis of highly distorted micro-X-ray diffraction (μ-XRD) patterns from hydrothermal fluid environments is a time-consuming process, often requiring substantial data preprocessing and labeled experimental data. This study demonstrates the potential of deep learning with a multitask learning (MTL) architecture to overcome these limitations. We trained MTL models to identify phase information in μ-XRD patterns, minimizing the need for labeled experimental data and masking preprocessing steps. Notably, MTL models showed superior accuracy compared to binary classification CNNs. Additionally, introducing a tailored cross-entropy loss function improved MTL model performance. Most significantly, MTL models tuned to analyze raw and unmasked XRD patterns achieved close performance to models analyzing preprocessed data, with minimal accuracy differences. This work indicates that advanced deep learning architectures like MTL can automate arduous data handling tasks, streamline the analysis of distorted XRD patterns, and reduce the reliance on labor-intensive experimental datasets.

cond-mat.mtrl-sci↗

arXiv2024

Dose and compositional dependence of irradiation-induced property change in FeCr

Ferritic/martensitic steels will be used as structural components in next generation nuclear reactors. Their successful operation relies on an understanding of irradiation-induced defect behaviour in the material. In this study, Fe and FeCr alloys (3-12%Cr) were irradiated with 20 MeV Fe-ions at 313 K to doses ranging between 0.00008 dpa to 6.0 dpa. This dose range covers six orders of magnitude, spanning low, transition and high dose regimes. Lattice strain and hardness in the irradiated material were characterised with micro-beam Laue X-ray diffraction and nanoindentation, respectively. Irradiation hardening was observed even at very low doses (0.00008 dpa) and showed a monotonic increase with dose up to 6.0 dpa. Lattice strain measurements of samples at 0.0008 dpa allow the calculation of equivalent Frenkel pair densities and corrections to the Norgett-Robinson-Torrens (NRT) model for Fe and FeCr alloys at low dose. NRT efficiency for FeCr is 0.2, which agrees with literature values for high irradiation energy. Lattice strain increases up to 0.8 dpa and then decreases when the damage dose is further increased. The strains measured in this study are lower and peak at a larger dose than predicted by atomistic simulations. This difference can be explained by taking temperature and impurities into account.

cond-mat.mtrl-sci↗

arXiv2023

SAIC: Integration of Speech Anonymization and Identity Classification

Speech anonymization and de-identification have garnered significant attention recently, especially in the healthcare area including telehealth consultations, patient voiceprint matching, and patient real-time monitoring. Speaker identity classification tasks, which involve recognizing specific speakers from audio to learn identity features, are crucial for de-identification. Since rare studies have effectively combined speech anonymization with identity classification, we propose SAIC - an innovative pipeline for integrating Speech Anonymization and Identity Classification. SAIC demonstrates remarkable performance and reaches state-of-the-art in the speaker identity classification task on the Voxceleb1 dataset, with a top-1 accuracy of 96.1%. Although SAIC is not trained or evaluated specifically on clinical data, the result strongly proves the model's effectiveness and the possibility to generalize into the healthcare area, providing insightful guidance for future work.

cs.SD↗

arXiv2023

GENOME: GenerativE Neuro-symbOlic visual reasoning by growing and reusing ModulEs

Recent works have shown that Large Language Models (LLMs) could empower traditional neuro-symbolic models via programming capabilities to translate language into module descriptions, thus achieving strong visual reasoning results while maintaining the model's transparency and efficiency. However, these models usually exhaustively generate the entire code snippet given each new instance of a task, which is extremely ineffective. We propose generative neuro-symbolic visual reasoning by growing and reusing modules. Specifically, our model consists of three unique stages, module initialization, module generation, and module execution. First, given a vision-language task, we adopt LLMs to examine whether we could reuse and grow over established modules to handle this new task. If not, we initialize a new module needed by the task and specify the inputs and outputs of this new module. After that, the new module is created by querying LLMs to generate corresponding code snippets that match the requirements. In order to get a better sense of the new module's ability, we treat few-shot training examples as test cases to see if our new module could pass these cases. If yes, the new module is added to the module library for future reuse. Finally, we evaluate the performance of our model on the testing set by executing the parsed programs with the newly made visual modules to get the results. We find the proposed model possesses several advantages. First, it performs competitively on standard tasks like visual question answering and referring expression comprehension; Second, the modules learned from one task can be seamlessly transferred to new tasks; Last but not least, it is able to adapt to new visual reasoning tasks by observing a few training examples and reusing modules.

cs.CV↗

arXiv2023

Continuous data assimilation for the three dimensional primitive equations with magnetic field

In this paper, the problem of continuous data assimilation of three dimensional primitive equations with magnetic field in thin domain is studied. We establish the well-posedness of the assimilation system and prove that the $H^2$-strong solution of the assimilation system converges exponentially to the reference solution in the sense of $L^2$ as $t\rightarrow \infty$. We also study the sensitivity analysis of the assimilation system and prove that a sequence of solutions of the difference quotient equation converge to the unique solution of the formal sensitivity equation.

math.AP↗

arXiv2023

Deformation Localisation in Ion-Irradiated FeCr

Irradiation-induced ductility loss is a major concern facing structural steels in next-generation nuclear reactors. Currently, the mechanisms for this are unclear but crucial to address for the design of reactor components. Here, the deformation characteristics around nanoindents in Fe and Fe10Cr irradiated with Fe ions to $\sim$1 displacement-per-atom at 313 K are non-destructively studied. Deformation localisation in the irradiated materials is evident from the increased pile-up height and slip step formation, measured by atomic force microscopy. From 3D X-ray Laue diffraction, measurements of lattice rotation and strain fields near the indent site show a large confinement, over 85%, of plasticity in the irradiated material. We find that despite causing increased irradiation hardening, Cr content has little effect on the irradiation-induced changes in pile-up topography and deformation fields. The results demonstrate that varying Cr content in steels has limited impact on retaining strain hardening capacity and reducing irradiation-induced embrittlement.

cond-mat.mtrl-sci↗

arXiv2023

The horizontal magnetic primitive equations approximation of the anisotropic MHD equations in a thin 3D domain

In this paper, we give a rigorous justification of the deviation of the primitive equations with only horizontal viscosity and magnetic diffusivity (PEHM) as the small aspect ratio limit of the incompressible three-dimensional scaled horizontal viscous MHD (SHMHD) equations. Choosing an aspect ratio parameter $\varepsilon \in(0,\infty)$, we consider the case that if the horizontal and vertical viscous coefficients are of $μ= O(1)$ and $ν= O({\varepsilon ^α})$, and the orders of magnetic diffusion coefficients $k$ and $σ$ are $k = O(1)$ and $σ= O({\varepsilon ^α})$, with $α> 2$, then the limiting system is the PEHM as $\varepsilon$ goes to zero. For ${H^1}$-initial data, we prove that the global weak solutions of the SHMHD equations converge strongly to the local-in-time strong solutions of the PEHM, as $\varepsilon$ tends to zero. For ${H^1}$-initial data with additional regularity $({\partial _z}{\tilde A_0},{\partial _z}{\tilde B_0}) \in {L^p}(Ω)(2 2$, the case $α=2$ has been investigated by Du and Li in \cite{2023-Du-Li}, in which they consider the PEM and the rate of global-in-time convergences is of the order $O(\varepsilon)$.

math.AP↗

arXiv2023

Machine Learning Automated Approach for Enormous Synchrotron X-Ray Diffraction Data Interpretation

Manual analysis of XRD data is usually laborious and time consuming. The deep neural network (DNN) based models trained by synthetic XRD patterns are proved to be an automatic, accurate, and high throughput method to analysis common XRD data collected from solid sample in ambient environment. However, it remains unknown that whether synthetic XRD based models are capable to solve u-XRD mapping data for in-situ experiments involving liquid phase exhibiting lower quality with significant artifacts. In this study, we collected u-XRD mapping data from an LaCl3-calcite hydrothermal fluid system and trained two categories of models to solve the experimental XRD patterns. The models trained by synthetic XRD patterns show low accuracy (as low as 64%) when solving experimental u-XRD mapping data. The accuracy of the DNN models was significantly improved (90% or above) when training them with the dataset containing both synthetic and small number of labeled experimental u-XRD patterns. This study highlighted the importance of labeled experimental patterns on the training of DNN models to solve u-XRD mapping data from in-situ experiments involving liquid phase.

cond-mat.dis-nn↗

arXiv2023

Digital autofocusing of a coded-aperture Laue diffraction microscope

To provide optimal depth resolution with a coded-aperture Laue diffraction microscope, an accurate position of the coded-aperture and its scanning geometry need to be known. However, finding the geometry by trial and error is a time-consuming and often challenging process because of the large number of parameters involved. In this paper, we propose an optimization approach to automate the focusing process after data is collected. We demonstrate the robustness and efficiency of the proposed approach with experimental data taken at a synchrotron facility.

eess.IV↗

arXiv2022

Asymptotic behavior for 2D stochastic Navier-Stokes equations with memory in unbounded domains

We consider a stochastic model which describes the motion of a 2D incompressible fluid in a unbounded domain with viscosity and memory effects. This model is different from the classical stochastic Navier-Stokes-Voigt equations due to the absence of the Voigt term $ -αΔu_{t}$, and has a much weaker dissipation than the usual Navier-Stokes-Voigt model since only the memory viscoelasticity is present. We are interested in the global well-posedness and long-time behaviors of this model. We first investigate the well-posedness by using the classical Faedo-Galerkin method. Unlike the general method of energy estimate, we then split the solution into two parts and get the low-order and high-order uniform estimates, respectively. Based on the uniform estimates of far-field values of solutions, we further prove the existence and uniqueness of random attractors in unbounded domains with a constructed compact subspace corresponding to memory. Finally, we give the upper semicontinuity of the attractors when stochastic perturbation approaches to zero.

math.AP↗

arXiv2022

Higher-order regularity of solutions to the large scale moist atmosphere system

In this paper, we study the higher-order regularity of solutions to the large scale moist atmosphere system through the way of $p$-strong solutions. On the basis of the well-posedness results of strong solutions, we first improve the regularity of solutions in the vertical direction, and then improve the regularity in the horizontal direction. Thus we obtain the well-posedness of solutions in $H^{2}$ space.

math.AP↗

arXiv2022

Global well-posedness of large scale moist atmosphere system with only horizontal viscosity in the dynamic equation

In order to find a better physical model to describe the large-scale cloud-water transformation and rainfall, we consider a moist atmosphere model consisting of the primitive equations with only horizontal viscosity in the dynamic equation and a set of humidity equations describing water vapor, rain water and cloud condensates. To overcome difficulties caused by the absence of vertical viscosity in the dynamic equation, we get the local existence of $v$ in $H^{1}$ space by combining the viscous elimination method and the $z-$weak solution method and using the generalized Bihari-Lasalle inequality. And then, we get the global existence of $v$ under higher regularity assumption of initial data. In turn, the existence of quasi-strong and strong solutions to the whole system is obtained. By introducing two new unknown quantities appropriately and utilizing the monotone operator theory to overcome difficulties caused by the Heaviside function in the source terms, we get the uniqueness of solutions.

math.AP↗

arXiv2022

CMOS based high-resolution dynamic X-ray imaging with inorganic perovskite

High-resolution dynamic X-ray detector is crucial for time-resolved digital radiography (DR) imaging and fast 3D medical computed tomography (CT) imaging. Recently, perovskites have become promising alternatives to conventional semi-conductor materials, e.g., Si, a-Se and CdTe, for direct X-ray detection. However, the feasibility of their combination with high-speed pixelated complementary metal-oxide-semiconductor (CMOS) arrays remains unknown. This work originally reports an innovative direct-conversion X-ray detector fabricated with 300 micrometer thick inorganic perovskite film printed on a tailored CMOS array. In-house measurements demonstrate that the CsPbBr3 film has excellent optoelectric properties of an electron mobility-lifetime product of 3.40x10$^{-5}$ cm$^2$ V$^{-1}$, and the X-ray detector exhibits high sensitivity of 9341uC Gy$_{\rm air}^{-1}$ cm$^{-2}$, and low detection limit of 588 nGy$_{\rm air}^{-1}$. This CMOS X-ray imaging detector achieves a high spatial resolution up to 5.5 lp/mm (close to the resolution limit of 6.0 lp/mm), and >300 frame per second (fps) readout speed. DR image of a resolution pattern phantom and a anesthesia mice, CT images of a biological specimen are acquired for the first time.

physics.app-ph↗

arXiv2022

Refinements for Bragg coherent X-ray diffraction imaging: Electron backscatter diffraction alignment and strain field computation

Bragg coherent X-ray diffraction imaging (BCDI) allows the three-dimensional (3D) measurement of lattice strain along the scattering vector for specific microcrystals. If at least three linearly independent reflections are measured, the 3D variation of the full lattice strain tensor within the microcrystal can be recovered. However, this requires knowledge of the crystal orientation, which is typically attained via estimates based on crystal geometry or synchrotron micro-beam Laue diffraction measurements. Here, we present an alternative method to determine the crystal orientation for BCDI measurements, by using electron backscatter diffraction (EBSD) to align Fe-Ni and Co-Fe alloy microcrystals on three different substrates. The orientation matrix is calculated from EBSD Euler angles and compared to the orientation determined using micro-beam Laue diffraction. The average angular mismatch between the orientation matrices is less than ~6 degrees, which is reasonable for the search for Bragg reflections. We demonstrate the use of an orientation matrix derived from EBSD to align and measure five reflections for a single Fe-Ni microcrystal using multi-reflection BCDI. Using this dataset, a refined strain field computation based on the gradient of the complex exponential of the phase is developed. This approach is shown to increase accuracy, especially in the presence of dislocations. Our results demonstrate the feasibility of using EBSD to pre-align BCDI samples and the application of more efficient approaches to determine the lattice strain tensor with greater accuracy.

cond-mat.mtrl-sci↗

arXiv2022

Depth-resolved Laue microdiffraction with coded-apertures

We introduce a rapid data acquisition and reconstruction method to image the crystalline structure of materials and associated strain and orientations at micrometer resolution using Laue diffraction. Our method relies on scanning a coded-aperture across the diffracted x-ray beams from a broadband illumination, and a reconstruction algorithm to resolve Laue microdiffraction patterns as a function of depth along the incident illumination path. This method provides a rapid access to full diffraction information at sub-micrometer volume elements in bulk materials. Here we present the theory as well as the experimental validation of this imaging approach.

cond-mat.mtrl-sci↗