arXiv subjects

Wei-Lun Chen

Publications and source records attributed to Wei-Lun Chen.

4 recordsLinked to original sources

arXiv2026

Accelerating iterative linear equation solver using modified domain-wall fermion matrix in lattice QCD simulations

Lattice simulations of Quantum Chromodynamics (QCD) enable one to calculate the low-energy properties of the strong interaction among quarks and gluons based on the first principle. The most time-consuming part of the numerical simulations of lattice QCD is typically solving a linear equation for the quark matrix. In particular, a discretized quark formulation called the domain-wall fermion operator requires a high numerical cost, while retaining the lattice version of the chiral symmetry to good precision. The domain-wall operator is defined on a five-dimensional (5D) space extending the four-dimensional (4D) spacetime with an extra fifth coordinate. After solving the linear equation in 5D space, the result vector is projected onto the original 4D space. There is a variant of the domain-wall operator that improves the convergence of the 5D linear equation while unchanging the 4D solution vector. In this paper, we examine how this variant of the domain-wall operator accelerates the iterative linear equation solver in practical setups. We also measure the eigenvalues of the operator and compare the condition number with the convergence of the solver. We use a generic lattice QCD code set Bridge++ that is planned to be released including the improved form of the domain-wall operator examined in this work with code for the GPU.

hep-lat↗

arXiv2025

Transfer Learning for Keypoint Detection in Low-Resolution Thermal TUG Test Images

This study presents a novel approach to human keypoint detection in low-resolution thermal images using transfer learning techniques. We introduce the first application of the Timed Up and Go (TUG) test in thermal image computer vision, establishing a new paradigm for mobility assessment. Our method leverages a MobileNetV3-Small encoder and a ViTPose decoder, trained using a composite loss function that balances latent representation alignment and heatmap accuracy. The model was evaluated using the Object Keypoint Similarity (OKS) metric from the COCO Keypoint Detection Challenge. The proposed model achieves better performance with AP, AP50, and AP75 scores of 0.861, 0.942, and 0.887 respectively, outperforming traditional supervised learning approaches like Mask R-CNN and ViTPose-Base. Moreover, our model demonstrates superior computational efficiency in terms of parameter count and FLOPS. This research lays a solid foundation for future clinical applications of thermal imaging in mobility assessment and rehabilitation monitoring.

cs.CV↗

arXiv2024

MECG-E: Mamba-based ECG Enhancer for Baseline Wander Removal

Electrocardiogram (ECG) is an important non-invasive method for diagnosing cardiovascular disease. However, ECG signals are susceptible to noise contamination, such as electrical interference or signal wandering, which reduces diagnostic accuracy. Various ECG denoising methods have been proposed, but most existing methods yield suboptimal performance under very noisy conditions or require several steps during inference, leading to latency during online processing. In this paper, we propose a novel ECG denoising model, namely Mamba-based ECG Enhancer (MECG-E), which leverages the Mamba architecture known for its fast inference and outstanding nonlinear mapping capabilities. Experimental results indicate that MECG-E surpasses several well-known existing models across multiple metrics under different noise conditions. Additionally, MECG-E requires less inference time than state-of-the-art diffusion-based ECG denoisers, demonstrating the model's functionality and efficiency.

eess.SP↗

arXiv2024

Bridging the Gap: Integrating Pre-trained Speech Enhancement and Recognition Models for Robust Speech Recognition

Noise robustness is critical when applying automatic speech recognition (ASR) in real-world scenarios. One solution involves the used of speech enhancement (SE) models as the front end of ASR. However, neural network-based (NN-based) SE often introduces artifacts into the enhanced signals and harms ASR performance, particularly when SE and ASR are independently trained. Therefore, this study introduces a simple yet effective SE post-processing technique to address the gap between various pre-trained SE and ASR models. A bridge module, which is a lightweight NN, is proposed to evaluate the signal-level information of the speech signal. Subsequently, using the signal-level information, the observation addition technique is applied to effectively reduce the shortcomings of SE. The experimental results demonstrate the success of our method in integrating diverse pre-trained SE and ASR models, considerably boosting the ASR robustness. Crucially, no prior knowledge of the ASR or speech contents is required during the training or inference stages. Moreover, the effectiveness of this approach extends to different datasets without necessitating the fine-tuning of the bridge module, ensuring efficiency and improved generalization.

cs.SD↗