Search arXiv⌕ Search

arXiv subjects

Khoa Nguyen

Publications and source records attributed to Khoa Nguyen.

At least 19 recordsLinked to original sources

Structural Conditional Independence Test via Neural Quantile Resolution

Conditional independence testing is difficult when dependence is weak, localized, or expressed through distributional changes rather than a global mean shift. Existing tests usually return a scalar statistic or p-value, which can detect rejection but gives limited resolution about where the conditional law changes. We propose QRCIT, a quantile-resolution conditional independence test that compares P(X | Z) and P(X | Y, Z) through an amortized quantile deformation. A lightweight residual quantile-flow model first learns the baseline conditional quantile law and then learns the additional deformation induced by Y, yielding both a calibrated global test statistic and an inspectable deformation field over context and quantile level. We also introduce the Structural Dependency Benchmark (SDB), which evaluates real-world-inspired dependence structures, such as context-local, tail-only, sign-reversing, moving-ridge, and mean-neutral patterns. Across synthetic PNL and SDB benchmarks, QRCIT improves detection of structured alternatives while maintaining empirical Type-I control. Increasing quantile resolution consistently improves merged-SDB power, and the local-rescue strategy sharpens detection when dependence is concentrated in structural regions. On Sachs, METABRIC breast cancer, and CCLE/PLX4720, QRCIT provides competitive real-data performance and interpretable deformation patterns. These results support quantile-resolved conditional-law deformation as a practical framework for testing and diagnosing structured conditional dependence.

cs.LG↗

Accelerated Algorithm for Sparse Regularized Partial Optimal Transport

Partial Optimal Transport (POT) extends the classical optimal transport problem by relaxing the strict mass conservation constraint, enabling its use in a wide range of real-world applications. In many of these settings, sparse transport plans are preferred for their interpretability and computational benefits. While smooth and strongly convex regularizers - such as quadratic or elastic net - have been vastly used in various machine learning applications to induce sparsity and accelerate computation, they have received less algorithmic attention compared to entropic approaches for computational POT. In this paper, we propose a new optimization framework that leverages these regularizers through a penalty-based reformulation, enabling efficient gradient-based updates while preserving the structure of the original problem. Our method accommodates a broad class of regularizers that promote structured and sparse transport plans. Building on this formulation, we design an accelerated first-order algorithm that alternates between smooth updates and simple projection steps. Through empirical benchmarks on color transfer, domain adaptation, and point cloud registration, our approach consistently outperforms established baselines - achieving lower transport cost, higher sparsity, and faster convergence - making it a practical and scalable solution for modern transport problems.

cs.LG↗

Moment-Matching Probabilistic Data Association for Optimization-Based SLAM

Optimization-based simultaneous localization and mapping (SLAM) makes it possible to reduce accumulated navigation errors of sensing platforms by returning to known areas (loop closure). In this paper, we present an approach to combine probabilistic data association (PDA) with optimization-based SLAM. Instead of associating a single measurement with each landmark, we follow the PDA paradigm from the multiobject tracking community. In particular, in a processing stage performed in addition to the nonlinear least-squares solver of optimization-based SLAM, our method (i) assigns multiple measurements to landmarks probabilistically, (ii) computes the mean and covariance of landmark distributions via moment matching by taking multiple measurement-to-landmark associations into account, and (iii) establishes a virtual landmark measurement and a corresponding linear-Gaussian measurement model that leads to the mean and covariance matrix as moment-matching PDA in (ii). By converting the PDA update step into an equivalent linear-Gaussian measurement update step, PDA can be performed effectively within any optimization-based SLAM method. Our preliminary numerical evaluation in a scenario with false negatives and false positives indicates that incremental smoothing and mapping 2 (iSAM2), combined with the proposed PDA approach, can improve agent localization performance compared to conventional iSAM2.

cs.RO↗

GeoQ: Geometry-Aware Conditional Quantile Error Estimation for Scientific Surrogate Models

Neural-network surrogate models are increasingly used to accelerate scientific simulations, but their deployment in extrapolative and autoregressive settings requires input-dependent estimates of prediction error. In this work, we introduce GeoQ (Geometry-Aware Conditional Quantile Error Estimation), a non-intrusive calibration framework for estimating surrogate error at individual query points. GeoQ represents the error at a query point as an anchor-averaged calibration error plus a learned nonnegative correction. This correction is modeled as an upper conditional quantile of the anchor-relative error increment, using geometry-based features that encode representation-space displacement and local support density. A cross-fitting procedure generates approximately out-of-sample calibration tuples, while a feature-space k-nearest-neighbor support score identifies regions \textcolor{black}{where the learned error model is supported by calibration data}. We evaluate GeoQ on scalar regression, chaotic dynamics, medium-range weather forecasting, and Richtmyer-Meshkov instability prediction. The results demonstrate that geometry-aware conditional quantile modeling provides a practical and non-intrusive approach for validity-aware error estimation in scientific surrogate models.

cs.LG↗

FedA2L: Adaptive layer-wise learning rate adjustment in decentralized federated learning

Decentralized intelligence systems with heterogeneous devices and limited coordination increasingly rely on decentralized federated learning (DFL). However, DFL suffers from convergence inefficiency under data heterogeneity due to the use of a uniform learning rate (LR) that ignores layer-specific optimization needs. Foundational layers are responsible for maintaining network consensus, while specialized layers adapt to local data characteristics, leading to conflicting gradients and degraded performance under non-IID conditions. To address this fundamental tension, this work introduces FedA2L, a method that dynamically adjusts layer-wise LRs based on model divergence signals. By leveraging local update intensity and network consensus constraints, FedA2L seamlessly integrates into existing DFL protocols without additional communication or coordination. Extensive evaluations across DFL algorithms, various model architectures, and datasets demonstrate that FedA2L achieves up to 4.94 times faster convergence than vanilla DFL and reduces communication rounds by up to 59% compared to scheduler-based baselines. Furthermore, FedA2L exhibits resilience to severe data heterogeneity, larger network sizes, and sparse topologies, reducing communication overhead and establishing it as a versatile optimization tool for resource-constrained or large-scale distributed learning in edge and IoT deployments. The code is released at https://github.com/nclabteam/FedA2L.

cs.LG↗

MEmilio -- A high performance Modular EpideMIcs simuLatIOn software for multi-scale and comparative simulations of infectious disease dynamics

Epidemic and pandemic preparedness with rapid outbreak response rely on timely, trustworthy evidence. Mathematical models are crucial for supporting timely and reliable evidence generation for public health decision-making with models spanning approaches from compartmental and metapopulation models to detailed agent-based simulations. Yet, the accompanying software ecosystem remains fragmented across model types, spatial resolutions, and computational targets, making models harder to compare, extend, and deploy at scale. Here we present MEmilio, a modular, high-performance framework for epidemic simulation that harmonizes the specification and execution of diverse dynamic epidemiological models within a unified and harmonized architecture. MEmilio couples an efficient C++ simulation core with coherent model descriptions and a user-friendly Python interface, enabling workflows that run on laptops as well as high-performance computing systems. Standardized representations of space, demography, and mobility support straightforward adaptations in resolution and population size, facilitating systematic inter-model comparisons and ensemble studies. The framework integrates readily with established tools for uncertainty quantification and parameter inference, supporting a broad range of applications from scenario exploration to calibration. Finally, strict software-engineering practices, including extensive unit and continuous integration testing, promote robustness and minimize the risk of errors as the framework evolves. By unifying implementations across modeling paradigms, MEmilio aims to lower barriers to reuse and generalize models, enable principled comparisons of implicit assumptions, and accelerate the development of novel approaches that strengthen modeling-based outbreak preparedness.

q-bio.PE↗

On $U(\mathfrak{h})$-free modules of finite rank over $\mathfrak{sl}(2)$

We study $\mathfrak{sl}(2)$-modules that are free of finite rank over $U(\mathfrak h)$, where $\mathfrak h$ is a fixed Cartan subalgebra of $\mathfrak{sl}(2)$. These modules form a natural class of non-weight modules. The coherent families obtained from this class via the weighting functor are identified. We also study a distinguished class of indecomposable $U(\mathfrak h)$-free modules defined in terms of Jordan blocks and give a recursive description of their socle filtrations. Finally, we apply the general results to exponential modules arising from the first Weyl algebra and obtain simplicity criteria for these modules.

math.RT↗

Environmental Sound Deepfake Detection Using Deep-Learning Framework

In this paper, we propose a deep-learning framework for Environmental Sound Deepfake Detection (ESDD) - the task of identifying whether the sound scene and sound event in an input audio recording is fake or real. To this end, we first conduct extensive experiments to explore how individual spectrograms, a wide range of network architectures, and pre-trained models affect the performance of an ESDD model. The experimental results on the benchmark datasets of EnvSDD indicate that detecting deepfake audio of sound scenes and detecting deepfake audio of sound events should be considered as individual tasks. We also show that fine-tuning a pre-trained model is more effective than training a model from scratch for ESDD. Ultimately, our best model, which fine-tunes the pre-trained BEATs model using the proposed two-phase training strategy, achieves an Accuracy of 0.98, F1 score of 0.95, and AUC score of 0.99 on the Test subset of the EnvSDD dataset. Our best model also achieves an Accuracy of 0.86, F1 score of 0.80, and AUC of 0.93 when evaluated cross-dataset on the ESD-Challenge-TestSet dataset.

cs.SD↗

A Unified Subject Map for 130 Years of Physics

More than a century of physics is recorded in the American Physical Society (APS) archive, but the corpus cannot be analyzed as a single, time-resolved object because its subject metadata are fragmented across eras with no shared vocabulary. We close this gap by using a frontier large language model to retrospectively assign the modern Physics Subject Headings (PhySH) to the historical archive, yielding a unified subject map for every APS paper from 1893 to 2025. The resulting map not only reproduces century-scale disciplinary arcs but also resolves the fine-grained lifecycles of individual ideas, materials, techniques, and discoveries across a vocabulary of over 3,000 PhySH Concepts. The map turns a fragmented archive into a quantitative substrate for systematic search and for data-driven studies of how physics evolves.

physics.hist-ph↗

Towards multi-modal forgery representation learning for AI-generated video detection and localization

Recent advances in generative AI have democratized video creation at scale. AI-generated videos, including partially manipulated clips across visual and audio channels, pose escalating risks of semantic distortion and misuse, which motivates the need for reliable detection tools. Most existing AI-generated video detectors remain limited by single- or partial-modality of data modeling and the lack of fine-grained temporal forgery localization. To address these challenges, our primary novelty introduces a core architecture that jointly integrates an LMM semantic branch with a spatio-temporal (ST) visual branch and a multi-scale partial-spoof (PS) audio branch. This multi-modal approach enables simultaneous detection and fine-grained temporal localization of partially manipulated AI-generated video forgeries. Extensive experiments show that this approach outperforms existing state-of-the-art methods.

cs.CV↗

Program Structure-aware Language Models: Targeted Software Testing beyond Textual Semantics

Recent advances in large language models for test case generation have improved branch coverage via prompt-engineered mutations. However, they still lack principled mechanisms for steering models toward specific high-risk execution branches, limiting their effectiveness for discovering subtle bugs and security vulnerabilities. We propose GLMTest, the first program structure-aware LLM framework for targeted test case generation that seamlessly integrates code property graphs and code semantics using a graph neural network and a language model to condition test case generation on execution branches. This structured conditioning enables controllable and branch-targeted test case generation, thereby potentially enhancing bug and security risk discovery. Experiments on real-world projects show that GLMTest built on a Qwen2.5-Coder-7B-Instruct model improves branch accuracy from 27.4% to 50.2% on TestGenEval benchmark compared with state-of-the-art LLMs, i.e., Claude-Sonnet-4.5 and GPT-4o-mini.

cs.SE↗

Material Identification using Multi-Modal Intrinsic Radiation and Radiography

We investigate multi-modal material identification for special nuclear material (SNM) configurations using a combination of X-ray radiography, high-resolution γ-ray spectroscopy, and neutron multiplicity measurements. We consider a Beryllium Reflected Plutonium sphere (BeRP) ball surrounded by one or two concentric shielding shells of unknown composition whose radii are assumed known from radiography. High-purity germanium (HPGe) spectra are reduced to net counts in selected Pu-239 photo-peaks, while neutron multiplicity information is summarized by Feynman variances Y2 and Y3 computed from factorial moments of the neutron counting statistics. Using synthetic data generated with the Gamma Detector Response and Analysis Software (GADRAS) for a range of shielding materials and thicknesses, we cast the material identification problem as a supervised multi-class classification task over all admissible shell-material combinations. We demonstrate that a random forest classifier trained on combined gamma and neutron features achieves almost perfect identification accuracy for single-shell cases, and substantial performance gains for more challenging double-shell configurations relative to gamma-only classification. Alternative statistical and machine-learning formulations for this multi-class problem are examined along with examination of the impact of model-mismatch between the forward model and the test cases as given by variations in the statistical noise. Opportunities for extending the approach to more complex geometries and experimental data are also discussed.

physics.ins-det↗

Neural Autoregressive Flows for Markov Boundary Learning

Recovering Markov boundary -- the minimal set of variables that maximizes predictive performance for a response variable -- is crucial in many applications. While recent advances improve upon traditional constraint-based techniques by scoring local causal structures, they still rely on nonparametric estimators and heuristic searches, lacking theoretical guarantees for reliability. This paper investigates a framework for efficient Markov boundary discovery by integrating conditional entropy from information theory as a scoring criterion. We design a novel masked autoregressive network to capture complex dependencies. A parallelizable greedy search strategy in polynomial time is proposed, supported by analytical evidence. We also discuss how initializing a graph with learned Markov boundaries accelerates the convergence of causal discovery. Comprehensive evaluations on real-world and synthetic datasets demonstrate the scalability and superior performance of our method in both Markov boundary discovery and causal discovery tasks.

cs.LG↗

"I Felt Bad After We Ignored Her": Understanding How Interface-Driven Social Prominence Shapes Group Discussions with GenAI

Recent advancements in the conversational and social capabilities of generative AI (GenAI) have sparked interest in its role as an agent capable of actively participating in human-AI group discussions. Despite this momentum, we don't fully understand how GenAI shapes conversational dynamics or how the interface design impacts its influence on the group. In this paper, we introduce interface-driven social prominence as a design lens for collaborative GenAI systems. We then present a GenAI-based conversational agent that can actively engage in spoken dialogue during video calls and design three distinct collaboration modes that vary the social prominence of the agent by manipulating its presence in the shared space and the degree of control users have over its participation. A mixed-methods within-subjects study, in which 18 dyads engaged in realistic discussions with a GenAI agent, offers empirical insights into how communication patterns and the collective negotiation of GenAI's influence shift based on how it is embedded into the collaborative experience. Based on these findings, we outline design implications for supporting the coordination and critical engagement required in human-AI groups.

cs.HC↗

Optimizing Task Scheduling in Fog Computing with Deadline Awareness

The rise of Internet of Things (IoT) devices has led to the development of numerous time-sensitive applications that require quick responses and low latency. Fog computing has emerged as a solution for processing these IoT applications, but it faces challenges such as resource allocation and job scheduling. Therefore, it is crucial to determine how to assign and schedule tasks on Fog nodes. This work aims to schedule tasks in IoT while minimizing the total energy consumption of nodes and enhancing the Quality of Service (QoS) requirements of IoT tasks, taking into account task deadlines. This paper classifies Fog nodes into two categories based on their traffic level: low and high. It schedules short-deadline tasks on low-traffic nodes using an Improved Golden Eagle Optimization (IGEO) algorithm, an enhancement that utilizes genetic operators for discretization. Long-deadline tasks are processed on high-traffic nodes using reinforcement learning (RL). This combined approach is called the Reinforcement Improved Golden Eagle Optimization (RIGEO) algorithm. Experimental results demonstrate that RIGEO achieves up to a 29% reduction in energy consumption, up to an 86% improvement in response time, and up to a 19% reduction in deadline violations compared to state-of-the-art algorithms.

cs.DC↗

A family of simple $U(\mathfrak{h})$-free modules of rank 2 over $\mathfrak{sl} (2)$

We study simple $\mathfrak{sl}(2)$-modules over $\mathbb C$ that are free of finite rank as $U(\mathfrak h)$-modules, where $\mathfrak h$ is a Cartan subalgebra of $\mathfrak{sl}(2)$. Our main result is an explicit classification of the scalar-type simple modules of rank $2$. We also give a criterion for when two such modules are isomorphic. Both the classification and the isomorphism problem reduce to twisted conjugacy classes in $\mbox{GL}_2({\mathbb C}[h])$ and rely on Cohn's standard form.

math.RT↗

NOIR: Privacy-Preserving Generation of Code with Open-Source LLMs

Although boosting software development performance, large language model (LLM)-powered code generation introduces intellectual property and data security risks rooted in the fact that a service provider (cloud) observes a client's prompts and generated code, which can be proprietary in commercial systems. To mitigate this problem, we propose NOIR, the first framework to protect the client's prompts and generated code from the cloud. NOIR uses an encoder and a decoder at the client to encode and send the prompts' embeddings to the cloud to get enriched embeddings from the LLM, which are then decoded to generate the code locally at the client. Since the cloud can use the embeddings to infer the prompt and the generated code, NOIR introduces a new mechanism to achieve indistinguishability, a local differential privacy protection at the token embedding level, in the vocabulary used in the prompts and code, and a data-independent and randomized tokenizer on the client side. These components effectively defend against reconstruction and frequency analysis attacks by an honest-but-curious cloud. Extensive analysis and results using open-source LLMs show that NOIR significantly outperforms existing baselines on benchmarks, including the Evalplus (MBPP and HumanEval, Pass@1 of 76.7 and 77.4), and BigCodeBench (Pass@1 of 38.7, only a 1.77% drop from the original LLM) under strong privacy against attacks.

cs.CR↗

Publicly Verifiable Secret Sharing: Generic Constructions and Lattice-Based Instantiations in the Standard Model

Publicly verifiable secret sharing (PVSS) allows a dealer to share a secret among a set of shareholders so that the secret can be reconstructed later from any set of qualified participants. In addition, any public verifier should be able to check the correctness of the sharing and reconstruction process. PVSS has been demonstrated to yield various applications, such as e-voting, distributed key generation, decentralized random number generation protocols, and multi-party computation. Although many concrete PVSS protocols have been proposed, their security is either proven in the random oracle model or relies on quantum-vulnerable assumptions such as factoring or discrete logarithm. In this work, we put forward a generic construction for PVSS that can be instantiated in the standard model under the Learning With Errors (LWE) assumption. Our instantiation provides the first post-quantum PVSS in the standard model, with a reasonable level of asymptotic efficiency.

cs.CR↗