arXiv subjects

Loc Nguyen

Publications and source records attributed to Loc Nguyen.

8 recordsLinked to original sources

arXiv2026

Consensus-based Agentic Large Language Model Framework for Harmonized Tariff Schedule Code Classification

Accurate Harmonized Tariff Schedule (HTS) code classification is essential for customs clearance, duty assessment, trade statistics, and regulatory compliance in maritime logistics. However, exact HTS classification remains challenging because product descriptions are often short, incomplete, or ambiguous, while correct classification depends on hierarchical tariff structures, legal notes, and jurisdiction-specific rules. This paper proposes an agentic large language model (LLM) framework for Canadian 10-digit HTS code classification in smart-port and maritime logistics environments. The framework integrates multi-agent information retrieval, semantic retrieval over official tariff documents, evidence-grounded reasoning, consensus-based validation, element-wise voting across hierarchical code components, confidence estimation, and human-in-the-loop escalation. We evaluate the framework on a private dataset of 3,300 domain-expert-labeled product records collected from logistics and delivery contexts. Experimental results show that exact 10-digit classification remains difficult even for advanced LLMs, with performance decreasing from coarse chapter-level prediction to fine-grained tariff and statistical suffix assignment. These findings demonstrate the need for evidence-grounded, uncertainty-aware, and human-centered classification workflows rather than fully autonomous single-step prediction. The proposed framework supports more interpretable, accountable, and compliance-oriented HTS classification for maritime logistics and smart-port operations. Our code is available at https://github.com/Analytics-Everywhere-Lab/hts.

cs.AI↗

arXiv2026

Inverse initial data for nonlinear Schr\"odinger equation via Carleman estimates and the contraction principle

We study an inverse initial-data problem for a nonlinear Schr\"odinger equation in which the initial wave field is reconstructed from lateral measurements. Our approach combines a Legendre-polynomial-exponential-time dimensional reduction with a Carleman-based contraction principle. First, we expand the solution in a weighted Legendre basis in time and truncate the expansion to obtain a coupled nonlinear elliptic system for the spatial coefficients. Next, we solve this reduced system by constructing a contraction map on a suitable admissible set. This contraction map admits a unique fixed point, which is the limit of the corresponding Picard iteration. We also establish a stability estimate showing that this fixed point remains close to the exact reduced solution in the noisy-data case. Finally, we present numerical experiments in two space dimensions for several different geometries and nonlinear exponents. The numerical results show that the proposed method accurately reconstructs the main features of the initial wave field and remains stable even when the boundary data contain noise.

math.NA↗

arXiv2026

PolypSteer: Counterfactual Endoscopic Synthesis via Training-Free Activation Steering

Generative diffusion models are increasingly used for medical imaging data augmentation, but text prompting cannot produce causal training data. Re-prompting rerolls the entire generation trajectory, altering anatomy, texture, and background. Inversion-based editing methods introduce reconstruction error that causes structural drift. We propose PolypSteer, a training-free activation-steering framework for endoscopic synthesis. PolypSteer identifies a pathology vector for each contrastive prompt pair in the cross-attention layers of a diffusion transformer. At inference time, it steers image activations along this vector, generating counterfactual pairs from scratch where the only difference is the steered concept. All other structure is preserved by construction. We evaluate PolypSteer across three experiments on Kvasir v3 and HyperKvasir. On counterfactual generation across three clinical concept pairs, PolypSteer achieves flip rates of 0.800, 0.925, and 0.950, outperforming the best inversion-based baseline in both concept flip rate and structural preservation. On dye disentanglement, PolypSteer achieves 75% dye removal against 20% (PnP) and 10% (h-Edit). On downstream polyp detection, augmenting with PolypSteer counterfactual pairs achieves ViT AUC of 0.9755 versus 0.9083 for quantity-matched re-prompting, confirming that counterfactual structure drives the gain. Code is at link https://github.com/UARK-AICV/PolypSteer

cs.CV↗

arXiv2024

DANA: Domain-Aware Neurosymbolic Agents for Consistency and Accuracy

Large Language Models (LLMs) have shown remarkable capabilities, but their inherent probabilistic nature often leads to inconsistency and inaccuracy in complex problem-solving tasks. This paper introduces DANA (Domain-Aware Neurosymbolic Agent), an architecture that addresses these issues by integrating domain-specific knowledge with neurosymbolic approaches. We begin by analyzing current AI architectures, including AutoGPT, LangChain ReAct and OpenAI's ChatGPT, through a neurosymbolic lens, highlighting how their reliance on probabilistic inference contributes to inconsistent outputs. In response, DANA captures and applies domain expertise in both natural-language and symbolic forms, enabling more deterministic and reliable problem-solving behaviors. We implement a variant of DANA using Hierarchical Task Plans (HTPs) in the open-source OpenSSA framework. This implementation achieves over 90\% accuracy on the FinanceBench financial-analysis benchmark, significantly outperforming current LLM-based systems in both consistency and accuracy. Application of DANA in physical industries such as semiconductor shows that its flexible architecture for incorporating knowledge is effective in mitigating the probabilistic limitations of LLMs and has potential in tackling complex, real-world problems that require reliability and precision.

cs.AI↗

arXiv2022

The Carleman-Newton method to globally reconstruct a source term for nonlinear parabolic equation

We propose to combine the Carleman estimate and the Newton method to solve an inverse source problem for nonlinear parabolic equations from lateral boundary data. The stability of this inverse source problem is conditionally logarithmic. Hence, numerical results due to the conventional least squares optimization might not be reliable. In order to enhance the stability, we approximate this problem by truncating the high frequency terms of the Fourier series that represents the solution to the governing equation. By this, we derive a system of nonlinear elliptic PDEs whose solution consists of Fourier coefficients of the solution to the parabolic governing equation. We solve this system by the Carleman-Newton method. The Carleman-Newton method is a newly developed algorithm to solve nonlinear PDEs. The strength of the Carleman-Newton method includes (1) no good initial guess is required and (2) the computational cost is not expensive. These features are rigorously proved. Having the solutions to this system in hand, we can directly compute the solution to the proposed inverse problem. Some numerical examples are displayed.

math.NA↗

arXiv2022

The Carleman-based contraction principle to reconstruct the potential of nonlinear hyperbolic equations

We develop an efficient and convergent numerical method for solving the inverse problem of determining the potential of nonlinear hyperbolic equations from lateral Cauchy data. In our numerical method we construct a sequence of linear Cauchy problems whose corresponding solutions converge to a function that can be used to efficiently compute an approximate solution to the inverse problem of interest. The convergence analysis is established by combining the contraction principle and Carleman estimates. We numerically solve the linear Cauchy problems using a quasi-reversibility method. Numerical examples are presented to illustrate the efficiency of the method.

math.NA↗

arXiv2021

Combinations of Jaccard with Numerical Measures for Collaborative Filtering Enhancement: Current Work and Future Proposal

Collaborative filtering (CF) is an important approach for recommendation system which is widely used in a great number of aspects of our life, heavily in the online-based commercial systems. One popular algorithms in CF is the K-nearest neighbors (KNN) algorithm, in which the similarity measures are used to determine nearest neighbors of a user, and thus to quantify the dependency degree between the relative user/item pair. Consequently, CF approach is not just sensitive to the similarity measure, yet it is completely contingent on selection of that measure. While Jaccard - as one of those commonly used similarity measures for CF tasks - concerns the existence of ratings, other numerical measures such as cosine and Pearson concern the magnitude of ratings. Particularly speaking, Jaccard is not a dominant measure, but it is long proven to be an important factor to improve any measure. Therefore, in our continuous efforts to find the most effective similarity measures for CF, this research focuses on proposing new similarity measure via combining Jaccard with several numerical measures. The combined measures would take the advantages of both existence and magnitude. Experimental results on, Movie-lens dataset, showed that the combined measures are preeminent outperforming all single measures over the considered evaluation metrics.

cs.IR↗

arXiv2012

Quantitative thermo-acoustic imaging: An exact reconstruction formula

This paper aims to mathematically advance the field of quantitative thermo-acoustic imaging. Given several electromagnetic data sets, we establish for the first time an analytical formula for reconstructing the absorption coefficient from thermal energy measurements. Since the formula involves derivatives of the given data up to the third order, it is unstable in the sense that small measurement noises may cause large errors. However, in the presence of measurement noise, the obtained formula, together with a noise regularization technique, provides a good initial guess for the true absorption coefficient. We finally correct the errors by deriving a reconstruction formula based on the least square solution of an optimal control problem and prove that this optimization step reduces the errors occurring and enhances the resolution.

math.AP↗