arXiv subjects

Yunqing Chen

Publications and source records attributed to Yunqing Chen.

5 recordsLinked to original sources

arXiv2026

Unveiling the Potential of Quantization with MXFP4: Strategies for Quantization Error Reduction

Large Language Models (LLMs) have intensified the need for low-precision formats that enable efficient, large-scale inference. The Open Compute Project (OCP) Microscaling (MX) standard is attractive due to its favorable hardware efficiency, but its 4-bit variant (MXFP4) lags behind NVIDIA's NVFP4 in accuracy, limiting adoption. We introduce two software-only techniques, Overflow-Aware Scaling (OAS) and Macro Block Scaling (MBS), that improve MXFP4 quantization fidelity without requiring hardware changes. OAS reduces overall errors by increasing effective dynamic range under power-of-two block scaling, while MBS allocates higher-precision scaling at a coarser granularity to better preserve outliers. Across multiple LLMs and standard downstream benchmarks, OAS and MBS reduce the end-to-end accuracy gap between MXFP4 and NVFP4 from about 10% to below 1% on average, while incurring modest GEMM overhead (6.2% on average). These results re-establish MXFP4 as a practical alternative to NVFP4, enabling near-NVFP4 accuracy while retaining MX's hardware-efficiency advantages (e.g., 12% relative area savings in tensor cores).

cs.AR↗

arXiv2025

Latent Space Consistency for Sparse-View CT Reconstruction

Computed Tomography (CT) is a widely utilized imaging modality in clinical settings. Using densely acquired rotational X-ray arrays, CT can capture 3D spatial features. However, it is confronted with challenged such as significant time consumption and high radiation exposure. CT reconstruction methods based on sparse-view X-ray images have garnered substantial attention from researchers as they present a means to mitigate costs and risks. In recent years, diffusion models, particularly the Latent Diffusion Model (LDM), have demonstrated promising potential in the domain of 3D CT reconstruction. Nonetheless, due to the substantial differences between the 2D latent representation of X-ray modalities and the 3D latent representation of CT modalities, the vanilla LDM is incapable of achieving effective alignment within the latent space. To address this issue, we propose the Consistent Latent Space Diffusion Model (CLS-DM), which incorporates cross-modal feature contrastive learning to efficiently extract latent 3D information from 2D X-ray images and achieve latent space alignment between modalities. Experimental results indicate that CLS-DM outperforms classical and state-of-the-art generative models in terms of standard voxel-level metrics (PSNR, SSIM) on the LIDC-IDRI and CTSpine1K datasets. This methodology not only aids in enhancing the effectiveness and economic viability of sparse X-ray reconstructed CT but can also be generalized to other cross-modal transformation tasks, such as text-to-image synthesis. We have made our code publicly available at https://anonymous.4open.science/r/CLS-DM-50D6/ to facilitate further research and applications in other domains.

eess.IV↗

arXiv2022

Model Pruning Based on Quantified Similarity of Feature Maps

Convolutional Neural Networks (CNNs) has been applied in numerous Internet of Things (IoT) devices for multifarious downstream tasks. However, with the increasing amount of data on edge devices, CNNs can hardly complete some tasks in time with limited computing and storage resources. Recently, filter pruning has been regarded as an effective technique to compress and accelerate CNNs, but existing methods rarely prune CNNs from the perspective of compressing high-dimensional tensors. In this paper, we propose a novel theory to find redundant information in three-dimensional tensors, namely Quantified Similarity between Feature Maps (QSFM), and utilize this theory to guide the filter pruning procedure. We perform QSFM on datasets (CIFAR-10, CIFAR-100 and ILSVRC-12) and edge devices, demonstrate that the proposed method can find the redundant information in the neural networks effectively with comparable compression and tolerable drop of accuracy. Without any fine-tuning operation, QSFM can compress ResNet-56 on CIFAR-10 significantly (48.7% FLOPs and 57.9% parameters are reduced) with only a loss of 0.54% in the top-1 accuracy. For the practical application of edge devices, QSFM can accelerate MobileNet-V2 inference speed by 1.53 times with only a loss of 1.23% in the ILSVRC-12 top-1 accuracy.

cs.CV↗

arXiv2009

Water Vapor: An Extraordinary Terahertz Wave Source under Optical Excitation

In modern terahertz (THz) sensing and imaging spectroscopy, water is considered a nemesis to be avoided due to strong absorption in the THz frequency range. Here we report the first experimental demonstration and theoretical implications of using femtosecond laser pulses to generate intense broadband THz emission from water vapor. When we focused an intense laser pulse in water vapor contained in a gas cell or injected from a gas jet nozzle, an extraordinarily strong THz field from optically excited water vapor is observed. Water vapor has more than 50% greater THz generation efficiency than dry nitrogen. It had previously been assumed that the nonlinear generation of THz waves in this manner primarily involves a free-electron plasma, but we show that the molecular structure plays an essential role in the process. In particular, we found that THz wave generation from H2O vapor is significantly stronger than that from D2O vapor. Vibronic activities of water cluster ions, occurring naturally in water vapor, may possibly contribute to the observed isotope effect along with rovibrational contributions from the predominant monomers.

physics.optics↗

arXiv2008

Measurement of the Optical Absorption Spectra of Epitaxial Graphene from Terahertz to Visible

We present experimental results on the optical absorption spectra of epitaxial graphene from the visible to the terahertz (THz) frequency range. In the THz range, the absorption is dominated by intraband processes with a frequency dependence similar to the Drude model. In the near IR range, the absorption is due to interband processes and the measured optical conductivity is close to the theoretical value of $e^{2}/4\hbar$. We extract values for the carrier densities, the number of carbon atom layers, and the intraband scattering times from the measurements.

cond-mat.mtrl-sci↗