Related papers: Improvement of Printing Quality for Laser-induced …
Computational fluid dynamics (CFD) can be used for evaluation of hemodynamics. However, its routine use is limited by labor-intensive manual segmentation, CFD mesh creation, and time-consuming simulation. This study aims to train a deep…
Vehicle aerodynamics optimization has become critical for automotive electrification, where drag reduction directly determines electric vehicle range and energy efficiency. Traditional approaches face an intractable trade-off:…
We present a comprehensive comparative study of three generative modeling paradigms: Denoising Diffusion Probabilistic Models (DDPM), Conditional Flow Matching (CFM), and MeanFlow. While DDPM and CFM require iterative sampling, MeanFlow…
Flow-based Transformer models have achieved state-of-the-art image generation performance, but often suffer from high inference latency and computational cost due to their large parameter sizes. To improve inference efficiency without…
This work presents, to the best of the authors' knowledge, the first generalizable and fully data-driven adaptive framework designed to stabilize deep learning (DL) autoregressive forecasting models over long time horizons, with the goal of…
Battery thermal management system is an indispensable part of the electric vehicles working with Lithium-ion batteries. Accordingly, lithium-ion batteries modeling, battery heat generation, and thermal management are the main focus of…
Federated learning (FL) is a promising paradigm to enable privacy-preserving deep learning from distributed data. Most previous works are based on federated average (FedAvg), which, however, faces several critical issues, including a high…
We introduce an innovative lumped-parameter model of the aortic valve, designed to efficiently simulate the impact of valve dynamics on blood flow. Our reduced model includes the elastic effects associated with the leaflets' curvature and…
Communication constraints are one of the major challenges preventing the wide-spread adoption of Federated Learning systems. Recently, Federated Distillation (FD), a new algorithmic paradigm for Federated Learning with fundamentally…
Despite rapid improvements in the performance of central processing unit (CPU), the calculation cost of simulating chemically reacting flow using CFD remains infeasible in many cases. The application of the convolutional neural networks…
Existing methods for restoring degraded human-centric images often struggle with insufficient fidelity, particularly in human body restoration (HBR). Recent diffusion-based restoration methods commonly adapt pre-trained text-to-image…
In this work, we propose a novel framework to enable diffusion models to adapt their generation quality based on real-time network bandwidth constraints. Traditional diffusion models produce high-fidelity images by performing a fixed number…
A new design of plasma source that merges the main features of capacitive dielectric barrier discharge (DBD) and cold atmospheric plasma jet (CAPJ) is discussed. The DBD system consists of a flexible, porous matrix consisting of silica…
Despite the plethora of deep learning frameworks for flow field prediction, most of them deal with flow fields on regular domains, and although the best ones can cope with irregular domains, they mostly rely on graph networks, so that real…
Using deep learning models pre-trained on Imagenet is the traditional solution for medical image classification to deal with data scarcity. Nevertheless, relevant literature supports that this strategy may offer limited gains due to the…
Federated Learning (FL) is a suitable solution for making use of sensitive data belonging to patients, people, companies, or industries that are obligatory to work under rigid privacy constraints. FL mainly or partially supports data…
Laser-induced acoustic desorption (LIAD) enables loading nanoparticles into optical traps under vacuum for levitated optomechanics experiments. Current LIAD systems rely on empirical optimization using available laboratory lasers rather…
Conventional 3D printings are generally time-consuming and printable metal inks are rather limited. From an alternative way, we proposed a liquid phase 3D printing for quickly making metal objects. Through introducing metal alloys whose…
A machine learning method to predict steady external fluid flows using elliptic input features is introduced. Using data from as few as one high-fidelity simulation, the proposed method produces models generalizable under changes to…
Besides their huge technological importance, fluidized beds have attracted a large amount of research because they are perfect playgrounds to investigate highly dynamic particulate flows. Their over-all behavior is determined by…