arXiv subjects

Zhizhong Zheng

Publications and source records attributed to Zhizhong Zheng.

5 recordsLinked to original sources

arXiv2026

A Quantization-Aware Training Based Lightweight Method for Neural Distinguishers

In 2019, Gohr pioneered the application of deep neural networks to differential cryptanalysis, developing DNN-based neural distinguisher classifiers to analyze the SPECK lightweight block cipher. Unlike traditional differential analysis, which relies on Boolean operations on 0-1 sequences, neural distinguishers extract continuous features, introducing 32-bit multiplications operations that increase complexity and potential redundancy. This study proposes a lightweight neural distinguisher based on quantization-aware training. Leveraging learnable step-size quantization, the model's weights are quantized to 1.58 bits, enabling the replacement of all convolutional multiplication operations with Boolean logic. Additionally, the ReLU activation function is reimplemented as a comparison-based indicator function. This transforms the original 32-bit multiplication-dependent architecture into a lightweight structure composed solely of Boolean operations, additions, and indicator functions. Experimental results confirm significant computational complexity reduction. Owing to a high proportion of zero-valued weights, the total operations amount to just 13.9% of Gohr's model. Critically, the most costly 32-bit multiplications are eliminated, with classification accuracy dropping by only 2.87%. When applied exclusively to the initial convolutional layer, the 128 1-by-1 convolutions are replaced with 4 Boolean operations on 16-bit sequences, incurring a negligible 0.3% accuracy loss.

cs.CR↗

arXiv2026

A one-step generation model with a Single-Layer Transformer: Layer number re-distillation of FreeFlow

Currently, Flow matching methods aim to compress the iterative generation process of diffusion models into a few or even a single step, with MeanFlow and FreeFlow being representative achievements of one-step generation based on Ordinary Differential Equations (ODEs). We observe that the 28-layer Transformer architecture of FreeFlow can be characterized as an Euler discretization scheme for an ODE along the depth axis, where the layer index serves as the discrete time step. Therefore, we distill the number of layers of the FreeFlow model, following the same derivation logic as FreeFlow, and propose SLT (Single-Layer Transformer), which uses a single shared DiT block to approximate the depth-wise feature evolution of the 28-layer teacher. During training, it matches the teacher's intermediate features at several depth patches, fuses those patch-level representations, and simultaneously aligns the teacher's final velocity prediction. Through distillation training, we compress the 28 independent Transformer Blocks of the teacher model DiT-XL/2 into a single Transformer Block, reducing the parameter count from 675M to 4.3M. Furthermore, leveraging its minimal parameters and rapid sampling speed, SLT can screen more candidate points in the noise space within the same timeframe, thereby selecting higher-quality initial points for the teacher model FreeFlow and ultimately enhancing the quality of generated images. Experimental results demonstrate that within a time budget comparable to two random samplings of the teacher model, our method performs over 100 noise screenings and produces a high-quality sample through the teacher model using the selected points. Quality fluctuations caused by low-quality initial noise under a limited number of FreeFlow sampling calls are effectively avoided, substantially improving the stability and average generation quality of one-step generation.

cs.CV↗

arXiv2024

Dual-Domain CLIP-Assisted Residual Optimization Perception Model for Metal Artifact Reduction

Metal artifacts in computed tomography (CT) imaging pose significant challenges to accurate clinical diagnosis. The presence of high-density metallic implants results in artifacts that deteriorate image quality, manifesting in the forms of streaking, blurring, or beam hardening effects, etc. Nowadays, various deep learning-based approaches, particularly generative models, have been proposed for metal artifact reduction (MAR). However, these methods have limited perception ability in the diverse morphologies of different metal implants with artifacts, which may generate spurious anatomical structures and exhibit inferior generalization capability. To address the issues, we leverage visual-language model (VLM) to identify these morphological features and introduce them into a dual-domain CLIP-assisted residual optimization perception model (DuDoCROP) for MAR. Specifically, a dual-domain CLIP (DuDoCLIP) is fine-tuned on the image domain and sinogram domain using contrastive learning to extract semantic descriptions from anatomical structures and metal artifacts. Subsequently, a diffusion model is guided by the embeddings of DuDoCLIP, thereby enabling the dual-domain prior generation. Additionally, we design prompt engineering for more precise image-text descriptions that can enhance the model's perception capability. Then, a downstream task is devised for the one-step residual optimization and integration of dual-domain priors, while incorporating raw data fidelity. Ultimately, a new perceptual indicator is proposed to validate the model's perception and generation performance. With the assistance of DuDoCLIP, our DuDoCROP exhibits at least 63.7% higher generalization capability compared to the baseline model. Numerical experiments demonstrate that the proposed method can generate more realistic image structures and outperform other SOTA approaches both qualitatively and quantitatively.

cs.CV↗

arXiv2020

Dual-energy CT Reconstruction from Dual Quarter Scans

Compared with conventional single-energy computed tomography (CT), dual-energy CT (DECT) provides better material differentiation but most DECT imaging systems require dual full-angle projection data at different X-ray spectra. Relaxing the requirement of data acquisition is a particularly attractive research to promote the applications of DECT in a wide range of imaging areas. In this work, we design a novel DECT imaging scheme with dual quarter scans and propose an efficient method to reconstruct the desired DECT images from dual limited-angle projection data, which enables DECT on imaging configurations with half-scan and largely reduces scanning angles and radiation doses. We first study the characteristics of image artifacts under dual quarter scans scheme, and find that the directional limited-angle artifacts of DECT images are complementarily distributed in image domain because the corresponding X-rays of high- and low-energy scans are orthogonal. Inspired by this finding, a fusion CT image is generated by integrating the limited-angle DECT images of dual quarter scans. This strategy largely reduces the limited-angle artifacts and preserves the image edges and inner structures. Utilizing the capability of neural network in the modeling of nonlinear problem, a novel Anchor network with single-entry double-out architecture is designed in this work to yield the desired DECT images from the generated fusion CT image. Experimental results on the simulated and real data verify the effectiveness of the proposed method.

physics.med-ph↗

arXiv2016

Efficient Image Reconstruction and Practical Decomposition for Dual-energy Computed Tomography

Dual-energy computed tomography (DECT) has shown great potential and promising applications in advanced imaging fields for its capabilities of material decomposition. However, image reconstructions and decompositions under sparse views dataset suffers severely from multi factors, such as insufficiencies of data, appearances of noise, and inconsistencies of observations. Under sparse views, conventional filtered back-projection type reconstruction methods fails to provide CT images with satisfying quality. Moreover, direct image decomposition is unstable and meet with noise boost even with full views dataset. This paper proposes an iterative image reconstruction algorithm and a practical image domain decomposition method for DECT. On one hand, the reconstruction algorithm is formulated as an optimization problem, which containing total variation regularization term and data fidelity term. The alternating direction method is utilized to design the corresponding algorithm which shows faster convergence speed compared with the existing ones. On the other hand, the image domain decomposition applies the penalized least square (PLS) estimation on decomposing the material mappings. The PLS includes linear combination term and the regularization term which enforces the smoothness on estimation images. The authors implement and evaluate the proposed joint method on real DECT projections and compare the method with typical and state-of-the-art reconstruction and decomposition methods. The experiments on dataset of an anthropomorphic head phantom show that our methods have advantages on noise suppression and edge reservation, without blurring the fine structures in the sinus area in the phantom. Compared to the existing approaches, our method achieves a superior performance on DECT imaging with respect to reconstruction accuracy and decomposition quality.

physics.med-ph↗