arXiv subjects

Tri Nguyen

Publications and source records attributed to Tri Nguyen.

At least 37 records · Page 2Linked to original sources

arXiv2025

DualFit: A Two-Stage Virtual Try-On via Warping and Synthesis

Virtual Try-On technology has garnered significant attention for its potential to transform the online fashion retail experience by allowing users to visualize how garments would look on them without physical trials. While recent advances in diffusion-based warping-free methods have improved perceptual quality, they often fail to preserve fine-grained garment details such as logos and printed text elements that are critical for brand integrity and customer trust. In this work, we propose DualFit, a hybrid VTON pipeline that addresses this limitation by two-stage approach. In the first stage, DualFit warps the target garment to align with the person image using a learned flow field, ensuring high-fidelity preservation. In the second stage, a fidelity-preserving try-on module synthesizes the final output by blending the warped garment with preserved human regions. Particularly, to guide this process, we introduce a preserved-region input and an inpainting mask, enabling the model to retain key areas and regenerate only where necessary, particularly around garment seams. Extensive qualitative results show that DualFit achieves visually seamless try-on results while faithfully maintaining high-frequency garment details, striking an effective balance between reconstruction accuracy and perceptual realism.

cs.CV↗

arXiv2025

LLM-as-a-Fuzzy-Judge: Fine-Tuning Large Language Models as a Clinical Evaluation Judge with Fuzzy Logic

Clinical communication skills are critical in medical education, and practicing and assessing clinical communication skills on a scale is challenging. Although LLM-powered clinical scenario simulations have shown promise in enhancing medical students' clinical practice, providing automated and scalable clinical evaluation that follows nuanced physician judgment is difficult. This paper combines fuzzy logic and Large Language Model (LLM) and proposes LLM-as-a-Fuzzy-Judge to address the challenge of aligning the automated evaluation of medical students' clinical skills with subjective physicians' preferences. LLM-as-a-Fuzzy-Judge is an approach that LLM is fine-tuned to evaluate medical students' utterances within student-AI patient conversation scripts based on human annotations from four fuzzy sets, including Professionalism, Medical Relevance, Ethical Behavior, and Contextual Distraction. The methodology of this paper started from data collection from the LLM-powered medical education system, data annotation based on multidimensional fuzzy sets, followed by prompt engineering and the supervised fine-tuning (SFT) of the pre-trained LLMs using these human annotations. The results show that the LLM-as-a-Fuzzy-Judge achieves over 80\% accuracy, with major criteria items over 90\%, effectively leveraging fuzzy logic and LLM as a solution to deliver interpretable, human-aligned assessment. This work suggests the viability of leveraging fuzzy logic and LLM to align with human preferences, advances automated evaluation in medical education, and supports more robust assessment and judgment practices. The GitHub repository of this work is available at https://github.com/2sigmaEdTech/LLMAsAJudge

cs.AI↗

arXiv2025

Multiphysics Modeling of SNAP 10A/2 Space Reactor with Cardinal

The SNAP 10/A nuclear-powered satellite was launched into space in 1965. The present work discusses the development of a coupled neutronic-thermal hydraulics model with the high-fidelity multiphysics code Cardinal for SNAP 10/A. A comparison of neutronic benchmarks between OpenMC and MCNP5 shows a difference of 304+/-70 pcm. The primary source of difference is that MCNP5 uses a combination of cross-section libraries ENDF/B-VI and ENDF/B-V for minor isotopes. At the same time, the present work utilizes ENDF/B-VIII.0. A comparison against experimental results showed a difference of 355 pcm. A coupled model of the SNAP 10/A reactor is then developed in Cardinal. Temperature results show reasonable agreement with reference calculations.

physics.comp-ph↗

arXiv2025

Fuzzy Logic -- Based Scheduling System for Part-Time Workforce

This paper explores the application of genetic fuzzy systems to efficiently generate schedules for a team of part-time student workers at a university. Given the preferred number of working hours and availability of employees, our model generates feasible solutions considering various factors, such as maximum weekly hours, required number of workers on duty, and the preferred number of working hours. The algorithm is trained and tested with availability data collected from students at the University of Cincinnati. The results demonstrate the algorithm's efficiency in producing schedules that meet operational criteria and its robustness in understaffed conditions.

cs.NE↗

arXiv2025

Jailbreak Detection in Clinical Training LLMs Using Feature-Based Predictive Models

Jailbreaking in Large Language Models (LLMs) threatens their safe use in sensitive domains like education by allowing users to bypass ethical safeguards. This study focuses on detecting jailbreaks in 2-Sigma, a clinical education platform that simulates patient interactions using LLMs. We annotated over 2,300 prompts across 158 conversations using four linguistic variables shown to correlate strongly with jailbreak behavior. The extracted features were used to train several predictive models, including Decision Trees, Fuzzy Logic-based classifiers, Boosting methods, and Logistic Regression. Results show that feature-based predictive models consistently outperformed Prompt Engineering, with the Fuzzy Decision Tree achieving the best overall performance. Our findings demonstrate that linguistic-feature-based models are effective and explainable alternatives for jailbreak detection. We suggest future work explore hybrid frameworks that integrate prompt-based flexibility with rule-based robustness for real-time, spectrum-based jailbreak monitoring in educational LLMs.

cs.CL↗

arXiv2025

NWP-based deep learning for tropical cyclone intensity prediction

Global artificial intelligence (AI) models are rapidly advancing and beginning to outperform traditional numerical weather prediction (NWP) models across metrics, yet predicting regional extreme weather such as tropical cyclone (TC) intensity presents unique spatial and temporal challenges that global AI models cannot capture. This study presents a new approach to train deep learning (DL) models specifically for regional extreme weather prediction. By leveraging physics-based NWP models to generate high-resolution data, we demonstrate that DL models can better predict or downscale TC intensity and structure when fine-scale processes are properly accounted for. Furthermore, by training DL models on different resolution outputs from physics-based simulations, we highlight the critical role of fine-scale processes in larger storm-scale dynamics, an aspect that current climate datasets used to train most global DL models cannot fully represent. These findings underscore the challenges in predicting or downscaling extreme weather with data-driven models, thus proposing the new role of NWP models as data generators for training DL models in the future AI model development for weather applications.

physics.ao-ph↗

arXiv2025

A2VIS: Amodal-Aware Approach to Video Instance Segmentation

Handling occlusion remains a significant challenge for video instance-level tasks like Multiple Object Tracking (MOT) and Video Instance Segmentation (VIS). In this paper, we propose a novel framework, Amodal-Aware Video Instance Segmentation (A2VIS), which incorporates amodal representations to achieve a reliable and comprehensive understanding of both visible and occluded parts of objects in a video. The key intuition is that awareness of amodal segmentation through spatiotemporal dimension enables a stable stream of object information. In scenarios where objects are partially or completely hidden from view, amodal segmentation offers more consistency and less dramatic changes along the temporal axis compared to visible segmentation. Hence, both amodal and visible information from all clips can be integrated into one global instance prototype. To effectively address the challenge of video amodal segmentation, we introduce the spatiotemporal-prior Amodal Mask Head, which leverages visible information intra clips while extracting amodal characteristics inter clips. Through extensive experiments and ablation studies, we show that A2VIS excels in both MOT and VIS tasks in identifying and tracking object instances with a keen understanding of their full shape.

cs.CV↗

arXiv2025

Momentum Contrastive Learning with Enhanced Negative Sampling and Hard Negative Filtering

Contrastive learning has become pivotal in unsupervised representation learning, with frameworks like Momentum Contrast (MoCo) effectively utilizing large negative sample sets to extract discriminative features. However, traditional approaches often overlook the full potential of key embeddings and are susceptible to performance degradation from noisy negative samples in the memory bank. This study addresses these challenges by proposing an enhanced contrastive learning framework that incorporates two key innovations. First, we introduce a dual-view loss function, which ensures balanced optimization of both query and key embeddings, improving representation quality. Second, we develop a selective negative sampling strategy that emphasizes the most challenging negatives based on cosine similarity, mitigating the impact of noise and enhancing feature discrimination. Extensive experiments demonstrate that our framework achieves superior performance on downstream tasks, delivering robust and well-structured representations. These results highlight the potential of optimized contrastive mechanisms to advance unsupervised learning and extend its applicability across domains such as computer vision and natural language processing

cs.LG↗

arXiv2024

Amodal Instance Segmentation with Diffusion Shape Prior Estimation

Amodal Instance Segmentation (AIS) presents an intriguing challenge, including the segmentation prediction of both visible and occluded parts of objects within images. Previous methods have often relied on shape prior information gleaned from training data to enhance amodal segmentation. However, these approaches are susceptible to overfitting and disregard object category details. Recent advancements highlight the potential of conditioned diffusion models, pretrained on extensive datasets, to generate images from latent space. Drawing inspiration from this, we propose AISDiff with a Diffusion Shape Prior Estimation (DiffSP) module. AISDiff begins with the prediction of the visible segmentation mask and object category, alongside occlusion-aware processing through the prediction of occluding masks. Subsequently, these elements are inputted into our DiffSP module to infer the shape prior of the object. DiffSP utilizes conditioned diffusion models pretrained on extensive datasets to extract rich visual features for shape prior estimation. Additionally, we introduce the Shape Prior Amodal Predictor, which utilizes attention-based feature maps from the shape prior to refine amodal segmentation. Experiments across various AIS benchmarks demonstrate the effectiveness of our AISDiff.

cs.CV↗

arXiv2024

Exascale Simulations of Fusion and Fission Systems

We discuss pioneering heat and fluid flow simulations of fusion and fission energy systems with NekRS on exascale computing facilities, including Frontier and Aurora. The Argonne-based code, NekRS, is a highly-performant open-source code for the simulation of incompressible and low-Mach fluid flow, heat transfer, and combustion with a particular focus on turbulent flows in complex domains. It is based on rapidly convergent high-order spectral element discretizations that feature minimal numerical dissipation and dispersion. State-of-the-art multilevel preconditioners, efficient high-order time-splitting methods, and runtime-adaptive communication strategies are built on a fast OCCA-based kernel library, libParanumal, to provide scalability and portability across the spectrum of current and future high-performance computing platforms. On Frontier, Nek5000/RS has achieved an unprecedented milestone in breaching over 1 trillion degrees of freedom with the spectral element methods for the simulation of the CHIMERA fusion technology testing platform. We also demonstrate for the first time the use of high-order overset grids at scale.

cs.CE↗

arXiv2024

How DREAMS are made: Emulating Satellite Galaxy and Subhalo Populations with Diffusion Models and Point Clouds

The connection between galaxies and their host dark matter (DM) halos is critical to our understanding of cosmology, galaxy formation, and DM physics. To maximize the return of upcoming cosmological surveys, we need an accurate way to model this complex relationship. Many techniques have been developed to model this connection, from Halo Occupation Distribution (HOD) to empirical and semi-analytic models to hydrodynamic. Hydrodynamic simulations can incorporate more detailed astrophysical processes but are computationally expensive; HODs, on the other hand, are computationally cheap but have limited accuracy. In this work, we present NeHOD, a generative framework based on variational diffusion model and Transformer, for painting galaxies/subhalos on top of DM with an accuracy of hydrodynamic simulations but at a computational cost similar to HOD. By modeling galaxies/subhalos as point clouds, instead of binning or voxelization, we can resolve small spatial scales down to the resolution of the simulations. For each halo, NeHOD predicts the positions, velocities, masses, and concentrations of its central and satellite galaxies. We train NeHOD on the TNG-Warm DM suite of the DREAMS project, which consists of 1024 high-resolution zoom-in hydrodynamic simulations of Milky Way-mass halos with varying warm DM mass and astrophysical parameters. We show that our model captures the complex relationships between subhalo properties as a function of the simulation parameters, including the mass functions, stellar-halo mass relations, concentration-mass relations, and spatial clustering. Our method can be used for a large variety of downstream applications, from galaxy clustering to strong lensing studies.

astro-ph.GA↗

arXiv2024

FLORAH: A generative model for halo assembly histories

The mass assembly history (MAH) of dark matter halos plays a crucial role in shaping the formation and evolution of galaxies. MAHs are used extensively in semi-analytic and empirical models of galaxy formation, yet current analytic methods to generate them are inaccurate and unable to capture their relationship with the halo internal structure and large-scale environment. This paper introduces FLORAH, a machine-learning framework for generating assembly histories of ensembles of dark matter halos. We train FLORAH on the assembly histories from the GUREFT and VSMDPL N-body simulations and demonstrate its ability to recover key properties such as the time evolution of mass and concentration. We obtain similar results for the galaxy stellar mass versus halo mass relation and its residuals when we run the Santa Cruz semi-analytic model on FLORAH-generated assembly histories and halo formation histories extracted from an N-body simulation. We further show that FLORAH also reproduces the dependence of clustering on properties other than mass (assembly bias), which is not captured by other analytic methods. By combining multiple networks trained on a suite of simulations with different redshift ranges and mass resolutions, we are able to construct accurate main progenitor branches (MPBs) with a wide dynamic mass range from $z=0$ up to an ultra-high redshift $z \approx 20$, currently far beyond that of a single N-body simulation. FLORAH is the first step towards a machine learning-based framework for planting full merger trees; this will enable the exploration of different galaxy formation scenarios with great computational efficiency at unprecedented accuracy.

astro-ph.GA↗

arXiv2024

Study of Turbulence and Pressure Recovery in the Heat Pipe Vapor Flow Using the Spectral-Element Method

Heat pipes can efficiently and passively remove heat in nuclear microreactors. Nevertheless, the flow dynamics within heat pipes present a significant challenge in designing and optimizing them for nuclear energy applications. This work aims to explore the vapor core of heat pipes through comprehensive two- and three-dimensional simulations, with a primary focus on modeling the pressure recovery observed in the condenser section. The goal is to establish improved correlations for one-dimensional heat pipe codes. The simulations are validated against experimental data from a vapor pipe documented in the literature. The turbulence model is employed in the two-dimensional simulations through the open-source spectral-element code Nek5000. This model provides insights into pressure recovery within heat pipes with low computational cost. In addition, Large Eddy Simulations (LES) are used to capture turbulent flow features in a three-dimensional vapor pipe model, utilizing the code NekRS. Using LES is crucial for comprehending the impact of laminar-to-turbulent transition on pressure recovery. A simulation framework is created to model the heat pipe's vapor core, laying the foundation for an enhanced understanding of heat pipe behavior. The ultimate goal is to improve and optimize heat pipe designs, provide data to validate lower-fidelity approaches and enhance their performance in nuclear reactors.

physics.flu-dyn↗

arXiv2024

Wait or Not to Wait: Evaluating Trade-Offs between Speed and Precision in Blockchain-based Federated Aggregation

This paper presents a fully coupled blockchain-assisted federated learning architecture that effectively eliminates single points of failure by decentralizing both the training and aggregation tasks across all participants. Our proposed system offers a high degree of flexibility, allowing participants to select shared models and customize the aggregation for local needs, thereby optimizing system performance, including accurate inference results. Notably, the integration of blockchain technology in our work is to promote a trustless environment, ensuring transparency and non-repudiation among participants when abnormalities are detected. To validate the effectiveness, we conducted real-world federated learning deployments on a private Ethereum platform, using two different models, ranging from simple to complex neural networks. The experimental results indicate comparable inference accuracy between centralized and decentralized federated learning settings. Furthermore, our findings indicate that asynchronous aggregation is a feasible option for simple learning models. However, complex learning models require greater training model involvement in the aggregation to achieve high model quality, instead of asynchronous aggregation. With the implementation of asynchronous aggregation and the flexibility to select models, participants anticipate decreased aggregation time in each communication round, while experiencing minimal accuracy trade-off.

cs.DC↗

arXiv2024

Characterising ultra-high-redshift dark matter halo demographics and assembly histories with the GUREFT simulations

Dark matter halo demographics and assembly histories are a manifestation of cosmological structure formation and have profound implications for the formation and evolution of galaxies. In particular, merger trees provide fundamental input for several modelling techniques, such as semi-analytic models (SAMs), sub-halo abundance matching (SHAM), and decorated halo occupation distribution models (HODs). Motivated by the new ultra-high-redshift (z > 10) frontier enabled by JWST, we present a new suite of Gadget at Ultrahigh Redshift with Extra-Fine Timesteps (GUREFT) dark matter-only cosmological simulations that are carefully designed to capture halo merger histories and structural properties in the ultra-z universe. The simulation suite consists of four 1024^3-particle simulations with box sizes of 5, 15, 35, and 90 Mpc h-1, each with 170 snapshots stored between 40 > z > 6. With the unprecedented number of available snapshots and strategically chosen dynamic range covered by these boxes, gureft uncovers the emerging dark matter halo populations and their assembly histories in the earliest epochs of cosmic history. In this work, we present the halo mass functions between z ~ 20 to 6 down to log(Mvir/Msun) ~ 5, and show that at high redshift, these robust halo mass functions can differ substantially from commonly used analytic approximations or older fitting functions in the literature. We also present key physical properties of the ultra-z halo population, such as concentration and spin, as well as their mass growth and merger rates, and again provide updated fitting functions.

astro-ph.CO↗

arXiv2024

Introducing the DREAMS Project: DaRk mattEr and Astrophysics with Machine learning and Simulations

We introduce the DREAMS project, an innovative approach to understanding the astrophysical implications of alternative dark matter models and their effects on galaxy formation and evolution. The DREAMS project will ultimately comprise thousands of cosmological hydrodynamic simulations that simultaneously vary over dark matter physics, astrophysics, and cosmology in modeling a range of systems -- from galaxy clusters to ultra-faint satellites. Such extensive simulation suites can provide adequate training sets for machine-learning-based analyses. This paper introduces two new cosmological hydrodynamical suites of Warm Dark Matter, each comprised of 1024 simulations generated using the Arepo code. One suite consists of uniform-box simulations covering a $(25~h^{-1}~{\rm M}_\odot)^3$ volume, while the other consists of Milky Way zoom-ins with sufficient resolution to capture the properties of classical satellites. For each simulation, the Warm Dark Matter particle mass is varied along with the initial density field and several parameters controlling the strength of baryonic feedback within the IllustrisTNG model. We provide two examples, separately utilizing emulators and Convolutional Neural Networks, to demonstrate how such simulation suites can be used to disentangle the effects of dark matter and baryonic physics on galactic properties. The DREAMS project can be extended further to include different dark matter models, galaxy formation physics, and astrophysical targets. In this way, it will provide an unparalleled opportunity to characterize uncertainties on predictions for small-scale observables, leading to robust predictions for testing the particle physics nature of dark matter on these scales.

astro-ph.GA↗

arXiv2024

ShapeFormer: Shape Prior Visible-to-Amodal Transformer-based Amodal Instance Segmentation

Amodal Instance Segmentation (AIS) presents a challenging task as it involves predicting both visible and occluded parts of objects within images. Existing AIS methods rely on a bidirectional approach, encompassing both the transition from amodal features to visible features (amodal-to-visible) and from visible features to amodal features (visible-to-amodal). Our observation shows that the utilization of amodal features through the amodal-to-visible can confuse the visible features due to the extra information of occluded/hidden segments not presented in visible display. Consequently, this compromised quality of visible features during the subsequent visible-to-amodal transition. To tackle this issue, we introduce ShapeFormer, a decoupled Transformer-based model with a visible-to-amodal transition. It facilitates the explicit relationship between output segmentations and avoids the need for amodal-to-visible transitions. ShapeFormer comprises three key modules: (i) Visible-Occluding Mask Head for predicting visible segmentation with occlusion awareness, (ii) Shape-Prior Amodal Mask Head for predicting amodal and occluded masks, and (iii) Category-Specific Shape Prior Retriever aims to provide shape prior knowledge. Comprehensive experiments and extensive ablation studies across various AIS benchmarks demonstrate the effectiveness of our ShapeFormer. The code is available at: \url{https://github.com/UARK-AICV/ShapeFormer}

cs.CV↗

arXiv2024

Large language models in 6G security: challenges and opportunities

The rapid integration of Generative AI (GenAI) and Large Language Models (LLMs) in sectors such as education and healthcare have marked a significant advancement in technology. However, this growth has also led to a largely unexplored aspect: their security vulnerabilities. As the ecosystem that includes both offline and online models, various tools, browser plugins, and third-party applications continues to expand, it significantly widens the attack surface, thereby escalating the potential for security breaches. These expansions in the 6G and beyond landscape provide new avenues for adversaries to manipulate LLMs for malicious purposes. We focus on the security aspects of LLMs from the viewpoint of potential adversaries. We aim to dissect their objectives and methodologies, providing an in-depth analysis of known security weaknesses. This will include the development of a comprehensive threat taxonomy, categorizing various adversary behaviors. Also, our research will concentrate on how LLMs can be integrated into cybersecurity efforts by defense teams, also known as blue teams. We will explore the potential synergy between LLMs and blockchain technology, and how this combination could lead to the development of next-generation, fully autonomous security solutions. This approach aims to establish a unified cybersecurity strategy across the entire computing continuum, enhancing overall digital security infrastructure.

cs.CR↗