Linh Nguyen

Markus Haltmeier, Michael Sandbichler, Thomas Berer et al.

Compressed sensing (CS) is a promising approach to reduce the number of measurements in photoacoustic tomography (PAT) while preserving high spatial resolution. This allows to increase the measurement speed and to reduce system costs. Instead of collecting point-wise measurements, in CS one uses various combinations of pressure values at different sensor locations. Sparsity is the main condition allowing to recover the photoacoustic (PA) source from compressive measurements. In this paper we introduce a new concept enabling sparse recovery in CS PAT. Our approach is based on the fact that the second time derivative applied to the measured pressure data corresponds to the application of the Laplacian to the original PA source. As typical PA sources consist of smooth parts and singularities along interfaces the Laplacian of the source is sparse (or at least compressible). To efficiently exploit the induced sparsity we develop a reconstruction framework to jointly recover the initial and the modified sparse source. Reconstruction results with simulated as well as experimental data are given.

Akshay Iyer, Linh Nguyen, Shweta Khushu

Wind energy is expected to be one of the leading ways to achieve the goals of the Paris Agreement but it in turn heavily depends on effective management of its operations and maintenance (O&M) costs. Blade failures account for one-third of all O&M costs thus making accurate detection of blade damages, especially cracks, very important for sustained operations and cost savings. Traditionally, damage inspection has been a completely manual process thus making it subjective, error-prone, and time-consuming. Hence in this work, we bring more objectivity, scalability, and repeatability in our damage inspection process, using deep learning, to miss fewer cracks. We build a deep learning model trained on a large dataset of blade damages, collected by our drone-based inspection, to correctly detect cracks. Our model is already in production and has processed more than a million damages with a recall of 0.96. We also focus on model interpretability using class activation maps to get a peek into the model workings. The model not only performs as good as human experts but also better in certain tricky cases. Thus, in this work, we aim to increase wind energy adoption by decreasing one of its major hurdles - the O\&M costs resulting from missing blade failures like cracks.

Melissa Pappy, Linh Nguyen, Suman Kumar et al.

Cybercrime has grown exponentially in both scale and sophistication, posing significant threats. As attack methods evolve rapidly, traditional classification schemes often fail to capture the complexity and diversity of modern threats. To address this gap, we introduce STRIKE,a Structured Taxonomy for Risk, Impact, Knowledge, and Emerging Threats, which provides a unified, multi-dimensional framework for categorizing cybercrimes. STRIKE spans both conventional and emerging domains, including ransomware, phishing, network intrusion, child sexual abuse material (CSAM), cryptojacking, deepfakes, and supply chain attacks. It organizes threats using criteria such as attack vectors, adversarial tactics, societal impact, detection techniques, and mitigation strategies. Alongside the taxonomy, we review recent advances in detection methodologies and present a response workflow to assist practitioners under active threat conditions. This work offers researchers, security professionals, and policymakers a practical foundation for threat analysis, comparative evaluation, and adaptive cyber defense.

Linh Nguyen, Akshay Iyer, Shweta Khushu

Wind energy's ability to compete with fossil fuels on a market level depends on lowering wind's high operational costs. Since damages on wind turbine blades are the leading cause for these operational problems, identifying blade damages is critical. However, recent works in visual identification of blade damages are still experimental and focus on optimizing the traditional machine learning metrics such as IoU. In this paper, we argue that pushing models to production long before achieving the "optimal" model performance can still generate real value for this use case. We discuss the performance of our damage's suggestion model in production and how this system works in coordination with humans as part of a commercialized product and how it can contribute towards lowering wind energy's operational costs.

Andy Nguyen, Danh Doan, Hoang Pham et al.

Memory-Augmented Generation (MAG) extends large language models with external memory to support long-context reasoning, but existing approaches universally treat memory as an external service that agents call into, delegating storage to separate pipelines of chunking, embedding, and graph extraction. This architectural separation means the system that stores knowledge does not understand it, leading to semantic drift between what the agent intended to remember and what the pipeline actually captured, loss of coordination context across agents, and fragile recovery after failures. In this paper, we propose ByteRover, an agent-native memory architecture that inverts the memory pipeline: the same LLM that reasons about a task also curates, structures, and retrieves knowledge. ByteRover represents knowledge in a hierarchical Context Tree, a file-based knowledge graph organized as Domain, Topic, Subtopic, and Entry, where each entry carries explicit relations, provenance, and an Adaptive Knowledge Lifecycle (AKL) with importance scoring, maturity tiers, and recency decay. Retrieval uses a 5-tier progressive strategy that resolves most queries at sub-100 ms latency without LLM calls, escalating to agentic reasoning only for novel questions. Experiments on LoCoMo and LongMemEval demonstrate that ByteRover achieves state-of-the-art accuracy on LoCoMo and competitive results on LongMemEval while requiring zero external infrastructure, no vector database, no graph database, no embedding service, with all knowledge stored as human-readable markdown files on the local filesystem.

Thao Hoang, Linh Nguyen, Khoi Do et al.

In the field of heart disease classification, two primary obstacles arise. Firstly, existing Electrocardiogram (ECG) datasets consistently demonstrate imbalances and biases across various modalities. Secondly, these time-series data consist of diverse lead signals, causing Convolutional Neural Networks (CNNs) to become overfitting to the one with higher power, hence diminishing the performance of the Deep Learning (DL) process. In addition, when facing an imbalanced dataset, performance from such high-dimensional data may be susceptible to overfitting. Current efforts predominantly focus on enhancing DL models by designing novel architectures, despite these evident challenges, seemingly overlooking the core issues, therefore hindering advancements in heart disease classification. To address these obstacles, our proposed approach introduces two straightforward and direct methods to enhance the classification tasks. To address the high dimensionality issue, we employ a Channel-wise Magnitude Equalizer (CME) on signal-encoded images. This approach reduces redundancy in the feature data range, highlighting changes in the dataset. Simultaneously, to counteract data imbalance, we propose the Inverted Weight Logarithmic Loss (IWL) to alleviate imbalances among the data. When applying IWL loss, the accuracy of state-of-the-art models (SOTA) increases up to 5% in the CPSC2018 dataset. CME in combination with IWL also surpasses the classification results of other baseline models from 5% to 10%.

Sarah Alnegheimish, Linh Nguyen, Laure Berti-Equille et al.

Recent studies have shown the ability of large language models to perform a variety of tasks, including time series forecasting. The flexible nature of these models allows them to be used for many applications. In this paper, we present a novel study of large language models used for the challenging task of time series anomaly detection. This problem entails two aspects novel for LLMs: the need for the model to identify part of the input sequence (or multiple parts) as anomalous; and the need for it to work with time series data rather than the traditional text input. We introduce sigllm, a framework for time series anomaly detection using large language models. Our framework includes a time-series-to-text conversion module, as well as end-to-end pipelines that prompt language models to perform time series anomaly detection. We investigate two paradigms for testing the abilities of large language models to perform the detection task. First, we present a prompt-based detection method that directly asks a language model to indicate which elements of the input are anomalies. Second, we leverage the forecasting capability of a large language model to guide the anomaly detection process. We evaluated our framework on 11 datasets spanning various sources and 10 pipelines. We show that the forecasting method significantly outperformed the prompting method in all 11 datasets with respect to the F1 score. Moreover, while large language models are capable of finding anomalies, state-of-the-art deep learning models are still superior in performance, achieving results 30% better than large language models.

Jeff Yang, Duy-Khanh Vu, Minh-Tien Nguyen et al.

This paper introduces layout-aware graph modeling for multimodal RAG. Different from traditional RAG methods that mostly deal with flat text chunks, the proposed method takes into account the relationship of multimodalities by using a graph structure. To do that, a graph modeling structure is defined based on document layout parsing. The structure of an input document is retained with the connection of text chunks, tables, and figures. This representation allows the method to handle complex questions that require information from multimodalities. To confirm the efficiency of the graph modeling, a flexible RAG pipeline is developed using robust components. Experimental results on four benchmark test sets confirm the contribution of the layout-aware modeling for performance improvement of the RAG pipeline.

Zhixin Pan, Ziyu Shu, Linh Nguyen et al.

The globalized semiconductor supply chain has made Hardware Trojans (HT) a significant security threat to embedded systems, necessitating the design of efficient and adaptable detection mechanisms. Despite promising machine learning-based HT detection techniques in the literature, they suffer from ad hoc feature selection and the lack of adaptivity, all of which hinder their effectiveness across diverse HT attacks. In this paper, we propose SAND, a selfsupervised and adaptive NAS-driven framework for efficient HT detection. Specifically, this paper makes three key contributions. (1) We leverage self-supervised learning (SSL) to enable automated feature extraction, eliminating the dependency on manually engineered features. (2) SAND integrates neural architecture search (NAS) to dynamically optimize the downstream classifier, allowing for seamless adaptation to unseen benchmarks with minimal fine-tuning. (3) Experimental results show that SAND achieves a significant improvement in detection accuracy (up to 18.3%) over state-of-the-art methods, exhibits high resilience against evasive Trojans, and demonstrates strong generalization.

Linh Nguyen, Chunhua Liu, Hong Yi Lin et al.

Code review is a crucial practice in software development. As code review nowadays is lightweight, various issues can be identified, and sometimes, they can be trivial. Research has investigated automated approaches to classify review comments to gauge the effectiveness of code reviews. However, previous studies have primarily relied on supervised machine learning, which requires extensive manual annotation to train the models effectively. To address this limitation, we explore the potential of using Large Language Models (LLMs) to classify code review comments. We assess the performance of LLMs to classify 17 categories of code review comments. Our results show that LLMs can classify code review comments, outperforming the state-of-the-art approach using a trained deep learning model. In particular, LLMs achieve better accuracy in classifying the five most useful categories, which the state-of-the-art approach struggles with due to low training examples. Rather than relying solely on a specific small training data distribution, our results show that LLMs provide balanced performance across high- and low-frequency categories. These results suggest that the LLMs could offer a scalable solution for code review analytics to improve the effectiveness of the code review process.

Linh Nguyen, Marcel Boersma, Erman Acar

Fraudulent activity in the financial industry costs billions annually. Detecting fraud, therefore, is an essential yet technically challenging task that requires carefully analyzing large volumes of data. While machine learning (ML) approaches seem like a viable solution, applying them successfully is not so easy due to two main challenges: (1) the sparsely labeled data, which makes the training of such approaches challenging (with inherent labeling costs), and (2) lack of explainability for the flagged items posed by the opacity of ML models, that is often required by business regulations. This article proposes SAGE-FIN, a semi-supervised graph neural network (GNN) based approach with Granger causal explanations for Financial Interaction Networks. SAGE-FIN learns to flag fraudulent items based on weakly labeled (or unlabelled) data points. To adhere to regulatory requirements, the flagged items are explained by highlighting related items in the network using Granger causality. We empirically validate the favorable performance of SAGE-FIN on a real-world dataset, Bipartite Edge-And-Node Attributed financial network (Elliptic++), with Granger-causal explanations for the identified fraudulent items without any prior assumption on the network structure.

Duy Nhat Phan, Sushant Jha, James P. Mavo et al.

Additive Manufacturing (AM) is transforming the manufacturing sector by enabling efficient production of intricately designed products and small-batch components. However, metal parts produced via AM can include flaws that cause inferior mechanical properties, including reduced fatigue response, yield strength, and fracture toughness. To address this issue, we leverage convolutional neural networks (CNN) to analyze thermal images of printed layers, automatically identifying anomalies that impact these properties. We also investigate various synthetic data generation techniques to address limited and imbalanced AM training data. Our models' defect detection capabilities were assessed using images of Nickel alloy 718 layers produced on a laser powder bed fusion AM machine and synthetic datasets with and without added noise. Our results show significant accuracy improvements with synthetic data, emphasizing the importance of expanding training sets for reliable defect detection. Specifically, Generative Adversarial Networks (GAN)-generated datasets streamlined data preparation by eliminating human intervention while maintaining high performance, thereby enhancing defect detection capabilities. Additionally, our denoising approach effectively improves image quality, ensuring reliable defect detection. Finally, our work integrates these models in the CLoud ADditive MAnufacturing (CLADMA) module, a user-friendly interface, to enhance their accessibility and practicality for AM applications. This integration supports broader adoption and practical implementation of advanced defect detection in AM processes.

Viet-Anh Le, Linh Nguyen, Truong X. Nghiem

This paper discusses the adaptive sampling problem in a nonholonomic mobile robotic sensor network for efficiently monitoring a spatial field. It is proposed to employ Gaussian process to model a spatial phenomenon and predict it at unmeasured positions, which enables the sampling optimization problem to be formulated by the use of the log determinant of a predicted covariance matrix at next sampling locations. The control, movement and nonholonomic dynamics constraints of the mobile sensors are also considered in the adaptive sampling optimization problem. In order to tackle the nonlinearity and nonconvexity of the objective function in the optimization problem we first exploit the linearized alternating direction method of multipliers (L-ADMM) method that can effectively simplify the objective function, though it is computationally expensive since a nonconvex problem needs to be solved exactly in each iteration. We then propose a novel approach called the successive convexified ADMM (SC-ADMM) that sequentially convexify the nonlinear dynamic constraints so that the original optimization problem can be split into convex subproblems. It is noted that both the L-ADMM algorithm and our SC-ADMM approach can solve the sampling optimization problem in either a centralized or a distributed manner. We validated the proposed approaches in 1000 experiments in a synthetic environment with a real-world dataset, where the obtained results suggest that both the L-ADMM and SC- ADMM techniques can provide good accuracy for the monitoring purpose. However, our proposed SC-ADMM approach computationally outperforms the L-ADMM counterpart, demonstrating its better practicality.

Daniel Obmann, Linh Nguyen, Johannes Schwab et al.

We propose a sparse reconstruction framework (aNETT) for solving inverse problems. Opposed to existing sparse reconstruction techniques that are based on linear sparsifying transforms, we train an autoencoder network $D \circ E$ with $E$ acting as a nonlinear sparsifying transform and minimize a Tikhonov functional with learned regularizer formed by the $\ell^q$-norm of the encoder coefficients and a penalty for the distance to the data manifold. We propose a strategy for training an autoencoder based on a sample set of the underlying image class such that the autoencoder is independent of the forward operator and is subsequently adapted to the specific forward model. Numerical results are presented for sparse view CT, which clearly demonstrate the feasibility, robustness and the improved generalization capability and stability of aNETT over post-processing networks.

Daniel Obmann, Linh Nguyen, Johannes Schwab et al.

We propose aNETT (augmented NETwork Tikhonov) regularization as a novel data-driven reconstruction framework for solving inverse problems. An encoder-decoder type network defines a regularizer consisting of a penalty term that enforces regularity in the encoder domain, augmented by a penalty that penalizes the distance to the data manifold. We present a rigorous convergence analysis including stability estimates and convergence rates. For that purpose, we prove the coercivity of the regularizer used without requiring explicit coercivity assumptions for the networks involved. We propose a possible realization together with a network architecture and a modular training strategy. Applications to sparse-view and low-dose CT show that aNETT achieves results comparable to state-of-the-art deep-learning-based reconstruction methods. Unlike learned iterative methods, aNETT does not require repeated application of the forward and adjoint models, which enables the use of aNETT for inverse problems with numerically expensive forward models. Furthermore, we show that aNETT trained on coarsely sampled data can leverage an increased sampling rate without the need for retraining.

Linh Nguyen, Jaime Valls Miro, Lei Shi et al.

Rapidly estimating the remaining wall thickness (RWT) is paramount for the non-destructive condition assessment evaluation of large critical metallic pipelines. A robotic vehicle with embedded magnetism-based sensors has been developed to traverse the inside of a pipeline and conduct inspections at the location of a break. However its sensing speed is constrained by the magnetic principle of operation, thus slowing down the overall operation in seeking dense RWT mapping. To ameliorate this drawback, this work proposes the partial scanning of the pipe and then employing Gaussian Processes (GPs) to infer RWT at the unseen pipe sections. Since GP prediction assumes to have normally distributed input data - which does correspond with real RWT measurements - Gaussian mixture (GM) models are proven in this work as fitting marginal distributions to effectively capture the probability of any RWT value in the inspected data. The effectiveness of the proposed approach is extensively validated from real-world data collected in collaboration with a water utility from a cast iron water main pipeline in Sydney, Australia.

Linh Nguyen, Jaime Valls Miro, Xiaojun Qiu

Knowing the geometry of a space is desirable for many applications, e.g. sound source localization, sound field reproduction or auralization. In circumstances where only acoustic signals can be obtained, estimating the geometry of a room is a challenging proposition. Existing methods have been proposed to reconstruct a room from the room impulse responses (RIRs). However, the sound source and microphones must be deployed in a feasible region of the room for it to work, which is impractical when the room is unknown. This work propose to employ a robot equipped with a sound source and four acoustic sensors, to follow a proposed path planning strategy to moves around the room to collect first image sources for room geometry estimation. The strategy can effectively drives the robot from a random initial location through the room so that the room geometry is guaranteed to be revealed. Effectiveness of the proposed approach is extensively validated in a synthetic environment, where the results obtained are highly promising.

Dung Tien Pham, Thai Van Nguyen, Hai Xuan Le et al.

The paper discusses an adaptive strategy to effectively control nonlinear manipulation motions of a dual arm robot (DAR) under system uncertainties including parameter variations, actuator nonlinearities and external disturbances. It is proposed that the control scheme is first derived from the dynamic surface control (DSC) method, which allows the robot's end-effectors to robustly track the desired trajectories. Moreover, since exactly determining the DAR system's dynamics is impractical due to the system uncertainties, the uncertain system parameters are then proposed to be adaptively estimated by the use of the radial basis function network (RBFN). The adaptation mechanism is derived from the Lyapunov theory, which theoretically guarantees stability of the closed-loop control system. The effectiveness of the proposed RBFN-DSC approach is demonstrated by implementing the algorithm in a synthetic environment with realistic parameters, where the obtained results are highly promising.

Linh Nguyen, Tsukasa Ishigaki

Recently multi-domain recommender systems have received much attention from researchers because they can solve cold-start problem as well as support for cross-selling. However, when applying into multi-domain items, although algorithms specifically addressing a single domain have many difficulties in capturing the specific characteristics of each domain, multi-domain algorithms have less opportunity to obtain similar features among domains. Because both similarities and differences exist among domains, multi-domain models must capture both to achieve good performance. Other studies of multi-domain systems merely transfer knowledge from the source domain to the target domain, so the source domain usually comes from external factors such as the search query or social network, which is sometimes impossible to obtain. To handle the two problems, we propose a model that can extract both homogeneous and divergent features among domains and extract data in a domain can support for other domain equally: a so-called Domain-to-Domain Translation Model (D2D-TM). It is based on generative adversarial networks (GANs), Variational Autoencoders (VAEs), and Cycle-Consistency (CC) for weight-sharing. We use the user interaction history of each domain as input and extract latent features through a VAE-GAN-CC network. Experiments underscore the effectiveness of the proposed system over state-of-the-art methods by a large margin.

Linh Nguyen, Sky Wang, Arunesh Sinha

Deep Neural Networks (DNNs) have been shown to be vulnerable against adversarial examples, which are data points cleverly constructed to fool the classifier. Such attacks can be devastating in practice, especially as DNNs are being applied to ever increasing critical tasks like image recognition in autonomous driving. In this paper, we introduce a new perspective on the problem. We do so by first defining robustness of a classifier to adversarial exploitation. Next, we show that the problem of adversarial example generation can be posed as learning problem. We also categorize attacks in literature into high and low perturbation attacks; well-known attacks like fast-gradient sign method (FGSM) and our attack produce higher perturbation adversarial examples while the more potent but computationally inefficient Carlini-Wagner (CW) attack is low perturbation. Next, we show that the dual approach of the attack learning problem can be used as a defensive technique that is effective against high perturbation attacks. Finally, we show that a classifier masking method achieved by adding noise to the a neural network's logit output protects against low distortion attacks such as the CW attack. We also show that both our learning and masking defense can work simultaneously to protect against multiple attacks. We demonstrate the efficacy of our techniques by experimenting with the MNIST and CIFAR-10 datasets.

Jacob Abernethy, Cyrus Anderson, Chengyu Dai et al.

Recovery from the Flint Water Crisis has been hindered by uncertainty in both the water testing process and the causes of contamination. In this work, we develop an ensemble of predictive models to assess the risk of lead contamination in individual homes and neighborhoods. To train these models, we utilize a wide range of data sources, including voluntary residential water tests, historical records, and city infrastructure data. Additionally, we use our models to identify the most prominent factors that contribute to a high risk of lead contamination. In this analysis, we find that lead service lines are not the only factor that is predictive of the risk of lead contamination of water. These results could be used to guide the long-term recovery efforts in Flint, minimize the immediate damages, and improve resource-allocation decisions for similar water infrastructure crises.

Julianne Chung, Linh Nguyen

Motion, e.g., due to patient movement or improper device calibration, is inevitable in many imaging modalities such as photoacoustic tomography (PAT) by a rotating system and can lead to undesirable motion artifacts in image reconstructions, if ignored. In this paper, we establish a hybrid-type model for PAT that incorporates motion in the model. We first introduce an approximate continuous model and establish two uniqueness results for simple parameterized motion models. Then we formulate the discrete problem of simultaneous motion estimation and image reconstruction as a separable nonlinear least squares problem and describe an automatic approach to detect and eliminate motion artifacts during the reconstruction process. Numerical examples validate our methods.