arXiv subjects

Quang Hung Nguyen

Publications and source records attributed to Quang Hung Nguyen.

4 recordsLinked to original sources

arXiv2026

Polarized Electron Scattering from Light Nuclei at High Energies

We present a theoretical approach to investigate the scattering of polarized electrons from light nuclei using the multipole expansion for the scattering cross section within the framework of the unified electroweak theory. Scattering processes corresponding to different electron polarizations are analyzed and compared with the unpolarized electron scattering investigated earlier. Besides, the contribution of both polarized and unpolarized terms to the scattering cross section is examined. Numerical calculations for stable $^{6,7}$Li and unstable $^7$Be nuclei using the Weinberg-Salam model show that the longitudinal polarization and weak interaction are not explicitly correlated when electrons scatter at $ θ\simeq 0^{\circ}$ across all energy scales. A strong correlation emerges at the other scattering angles when the electron energy exceeds 10 GeV. This study provides additional information about nuclear structure and uncovers the role of electron polarization and its correlation with the weak interaction in each process, thus offering a more complete picture of electron-nucleus scattering.

nucl-th↗

arXiv2025

FinStat2SQL: A Text2SQL Pipeline for Financial Statement Analysis

Despite the advancements of large language models, text2sql still faces many challenges, particularly with complex and domain-specific queries. In finance, database designs and financial reporting layouts vary widely between financial entities and countries, making text2sql even more challenging. We present FinStat2SQL, a lightweight text2sql pipeline enabling natural language queries over financial statements. Tailored to local standards like VAS, it combines large and small language models in a multi-agent setup for entity extraction, SQL generation, and self-correction. We build a domain-specific database and evaluate models on a synthetic QA dataset. A fine-tuned 7B model achieves 61.33\% accuracy with sub-4-second response times on consumer hardware, outperforming GPT-4o-mini. FinStat2SQL offers a scalable, cost-efficient solution for financial analysis, making AI-powered querying accessible to Vietnamese enterprises.

cs.AI↗

arXiv2025

Optimizing MV CBCT Imaging Protocols Using NTCP and Secondary Cancer Risk: A Multi-Site Study in Breast, Pelvic, and Head & Neck Radiotherapy

Purpose: To evaluate the cumulative radiobiological impact of daily Megavoltage Cone-Beam Computed Tomography (MV-CBCT) imaging dose based on Normal Tissue Complication Probability (NTCP) and Excess Absolute Risk (EAR) of secondary malignancies among radiotherapy patients treated for breast, pelvic, and head and neck cancers. This study investigated whether MV-CBCT imaging dose warrants protocol personalization according to patient age, anatomical treatment site, and organ-specific radiosensitivity. Methods: This retrospective study included cohorts of breast (n=30), pelvic (n=17), and head and neck (n=20) cancer patients undergoing radiotherapy with daily MV-CBCT. Imaging plans using two common protocols (5 MU and 10 MU per fraction) were analyzed. NTCP values were estimated using logistic and Lyman-Kutcher-Burman (LKB) models, while EAR was calculated using Schneider's Organ Equivalent Dose (OED)-based model. Statistical analysis used paired t-tests, and results were further stratified by age (under 40, 40 to 60, over 60 years). Results: In breast cancer patients, NTCP for lung increased significantly under the 10 MU protocol (p<0.001). EAR was elevated in younger breast patients (under 40 years), with some exceeding 15 cases per 10,000 person-years. In pelvic and head and neck groups, NTCP and EAR remained low (under 1 percent), with no clinically meaningful differences between protocols. Across all sites, younger age correlated with higher secondary cancer risk. Conclusion: Daily 10 MU MV-CBCT presents minimal additional risk in pelvic and head and neck radiotherapy. For breast cancer patients under 40, however, it significantly increases secondary cancer risk and lung NTCP. Personalized imaging protocols are recommended based on age, treatment site, and radiosensitivity.

physics.med-ph↗

arXiv2025

NEU-ESC: A Comprehensive Vietnamese dataset for Educational Sentiment analysis and topic Classification toward multitask learning

In the field of education, understanding students' opinions through their comments is crucial, especially in the Vietnamese language, where resources remain limited. Existing educational datasets often lack domain relevance and student slang. To address these gaps, we introduce NEU-ESC, a new Vietnamese dataset for Educational Sentiment Classification and Topic Classification, curated from university forums, which offers more samples, richer class diversity, longer texts, and broader vocabulary. In addition, we explore multitask learning using encoder-only language models (BERT), in which we showed that it achieves performance up to 83.7% and 79.8% accuracy for sentiment and topic classification tasks. We also benchmark our dataset and model with other datasets and models, including Large Language Models, and discuss these benchmarks. The dataset is publicly available at: https://huggingface.co/datasets/hung20gg/NEU-ESC.

cs.CL↗