Regularization for Unconditional Image Diffusion Models via Shifted Data Augmentation by Kensuke Nakamura, Bong-Soo Sohn, Simon Korman, Byung-Woo Hong

“…This dual-path structure enables effective regularization while suppressing leakage through a trajectory shift in the diffusion process. We implement SDA with image rotation as its most basic and interpretable configuration. …”
Get full text

GODiff: Region-Specific Semantic Editing With CLIP-Guided Diffusion Models by Junho Kim, Eunwoo Kim, Kyungjae Lee

“…Furthermore, by optimizing the guidance in real-time during the generative process, GODiff more accurately reflects the intent of the text prompt and produces more natural and consistent results. …”
Get full text

ProT-GFDM: A generative fractional diffusion model for protein generation by Xiao Liang, Wentao Ma, Eric Paquet, Herna Viktor, Wojtek Michalowski

“…Unlike classical methods that rely on standard Brownian motion, ProT-GFDM employs a fractional stochastic process with superdiffusive properties to improve the capture of long-range dependencies in protein structures. …”
Get full text

SECONDGRAM: Self-conditioned diffusion with gradient manipulation for longitudinal MRI imputation by Brandon Theodorou, Anant Dadu, Mike Nalls, Faraz Faghri, Jimeng Sun

“…The implications extend beyond healthcare, presenting a versatile solution for data augmentation in numerous fields struggling with longitudinal data constraints, thus enhancing decision-making processes in precision medicine.…”
Get full text

Identifying the electron diffusion region in a realistic simulation of Earth's magnetotail by Maha Ashour‐Abdalla, Giovanni Lapenta, Raymond Walker, Mostafa El‐Alaoui, Haoming Liang, Meng Zhou, Jean Berchem, Melvyn L. Goldstein

“…Key to unravel this fundamental process is in determining the location and extent of the electron diffusion region (EDR) where the energy conversion is initiated. …”
Get full text

Diffusion Behavior and Kinetics of the Iron–Nickel Interface During Annealing Treatment by Wenlin Wu, Fei Zhong, Pu Zhou, Jue Lu, Feng Wang

“…The results reveal annealing induces Fe-Ni alloy layer formation at the plating–substrate interface, with thicknesses ranging from 2.2 μm to 4.8 μm as the temperature increases from 650 °C to 800 °C and the duration extends from 1 min to 5 min. Additionally, a diffusion kinetics model based on Fick’s second law was established to describe the diffusion process of nickel atoms in the iron matrix. …”
Get full text

Image instance segmentation based on diffusion model improved by step noisy by Hui Ma, Wanchun Sun, Shujia Li, Jinjun Zhang

“…During the inverse denoising process at different stages of the same instance segmentation task, a controllable relationship exists between different step sizes corresponding to different Gaussian distributions. …”
Get full text

Research progress on characteristics of leakage and diffusion of buried CO2 pipeline by Junpeng ZHANG, Qing MIAO, Shaodong JING, Yongsheng WANG, Xin OUYANG, Zhenning FAN, Haining LIANG, Jian ZHANG

“…The influences of soil geological conditions, soil temperatures, ambient pressures, and wind velocities on the CO2 seepage and diffusion processes in the soil were also examined. …”
Get full text

Development of neutron diffusion and transport algorithms based on finite volume method by LI Wei, LIU Xiaojing, CHAI Xiang, ZHAO Pengcheng

“…Simultaneous solutions to the conservation equations of multiple physics fields need to be pursued, providing a unified approach to modeling, discretization, and iterative computation processes.PurposeThis study aims to achieve discrete and iterative solutions for multigroup neutron diffusion equations and neutron transport equations, considering the strong coupling between neutronics and thermal-hydraulics.MethodsFirstly, based on the open-source computational fluid dynamics (CFD) platform OpenFOAM, the finite volume method (FVM) was employed to discretize the control equations for neutron diffusion and neutron transport using the Gauss theorem. …”
Get full text

Underwater Image Enhancement Using a Diffusion Model with Adversarial Learning by Xueyan Ding, Xiyu Chen, Yixin Sui, Yafei Wang, Jianxin Zhang

“…Moreover, we propose a multi-scale dynamic-windowed attention mechanism to effectively fuse global and local features, optimizing the process of capturing and integrating information. …”
Get full text

Discrete Element Analysis of Grouting Reinforcement and Slurry Diffusion in Overburden Strata by Pengfei Guo, Weiquan Zhao, Yahui Ma, Huiling Gen

“…Research on the grouting reinforcement mechanism of overburden is constrained by the concealed and heterogeneous nature of geotechnical media, posing dual challenges in theoretical analysis and process visualization. Based on discrete element numerical simulations and laboratory tests, an analytical model for grouting reinforcement in overburden layers is developed, revealing the influence of grouting pressure on slurry diffusion shape and distance. …”
Get full text

Prognostic significance of AIF expression in diffuse large B-cell lymphoma by Vitaly A. Rosin, Elena V. Vaneeva, Dmitriy A. Diakonov, Natalya V. Glubokovskikh, Svetlana V. Samarina

“…Dysregulation of apoptosis in diffuse B-large cell lymphoma (DLBCL) contributes to increased tumor proliferation, as well as the formation of resistance to treatment. …”
Get full text

Real-world defocus deblurring via score-based diffusion models by Yuhao Li, Haoran Fang, Xiang Lei, Qi Wang, Gang Hu, Jiaqing Dong, Zilong Li, Jiabin Lin, Qiegen Liu, Xianlin Song

“…This process allows the distribution of generated images to increasingly approximate that of sharply focused images. …”
Get full text

Knowledge-guided diffusion model for 3D ligand-pharmacophore mapping by Jun-Lin Yu, Cong Zhou, Xiang-Li Ning, Jun Mou, Fan-Bo Meng, Jing-Wei Wu, Yi-Ting Chen, Biao-Dan Tang, Xiang-Gen Liu, Guo-Bo Li

“…We herein propose a knowledge-guided diffusion framework for ‘on-the-fly’ 3D ligand-pharmacophore mapping, named DiffPhore. …”
Get full text

Leakage and diffusion simulation of submarine hydrogen-blended natural gas pipeline by Yitao FANG, Haichuan TAN, Liang TANG, Mengjie WANG, Jingfa LI, Yue SU

“…In the aperture leakage scenarios, the diffusion process of leaking hydrogen-blended natural gas exhibited a deceleration compared to anchor break leakage, primarily attributed to the resistance posed by seabed sediment. …”
Get full text

Classification of Microbial Activity and Inhibition Zones Using Neural Network Analysis of Laser Speckle Images by Ilya Balmages, Dmitrijs Bļizņuks, Inese Polaka, Alexey Lihachev, Ilze Lihacova

“…The novelty of the study lies in combining a speckle imaging subpixel correlation algorithm with ML classification and with pre- and post-processing. This approach enables early automated assessment of antimicrobial effects with potential applications in rapid drug susceptibility testing and microbiological research.…”
Get full text

REDUCTION OF TEMPERATURE OF MELTING OF THE DIFFUSIVE HIGH-HARDENED ALLOYS FOR INDUCTION SURFACING by V. G. Scherbakov

Get full text

Some problems of mass-transfer in binary mixtures in terms of diffusion model by L. A. Serafimov, A. V. Anisimov, K. Yu. Tarkhov

“…The accurate equation of mass-transfer based on diffusion model using linearization is proposed for qualitative investigations of binary mixtures rectification process…”
Get full text

Wavelet-based diffusion with spatial-frequency attention for hyperspectral anomaly detection by Sitian Liu, Chunli Zhu, Lintao Peng, Xinyue Su, Lianjie Li, Guanghui Wen

“…To address those challenges, we propose a novel frequency decomposition framework wavelet-enhanced diffusion framework for HAD, termed as WDHAD. Following a 2D discrete wavelet transformation, the LFC and HFC are processed in parallel: 1) The LFC is handled via a Low-Frequency Diffusion Model (LFDM), which employs a Low-Frequency Denoising Autoencoder (LFDAE) with spatial-frequency attention to recover key features and remove background noise. 2) The HFC is processed through a High-Frequency Enhancement Module (HFEM) that preserves edges and textures to improve anomaly detection. …”
Get full text

The Normal Cross Numerical Method Used to Solve the Anisotropic Diffusion Problem by LIU Xiaogang, LIAO Jipeng, CHEN Juhui, LU Yiping, SUN Haifeng

“…The anisotropic and anisotropic diffusion problems with analytic solutions are simulated by using a variety of rules and twisted meshes.The normal cross method decomposes the normal part into two parts: the current unit and the adjacent unit, and the diffusion cross part is displayed and processed according to the source term. …”
Get full text