Current Strategies for Exosome Cargo Loading and Targeting Delivery.

Extracellular vesicles (EVs) such as ectosomes and exosomes have gained attention as promising natural carriers for drug delivery. Exosomes, which range from 30 to 100 nm in diameter, possess a lipid bilayer and are secreted by various cells. Due to their high biocompatibility, stability, and low immunogenicity, exosomes are favored as cargo carriers. The lipid bilayer membrane of exosomes also offers protection against cargo degradation, making them a desirable candidate for drug delivery. However, loading cargo into exosomes remains to be a challenge. Despite various strategies such as incubation, electroporation, sonication, extrusion, freeze-thaw cycling, and transfection that have been developed to facilitate cargo loading, inadequate efficiency still persists. This review offers an overview of current cargo delivery strategies using exosomes and summarizes recent approaches for loading small-molecule, nucleic acid, and protein drugs into exosomes. With insights from these studies, we provide ideas for more efficient and effective delivery of drug molecules by using exosomes.

Adversarial parameter defense by multi-step risk minimization.

Previous studies demonstrate DNNs' vulnerability to adversarial examples and adversarial training can establish a defense to adversarial examples. In addition, recent studies show that deep neural networks also exhibit vulnerability to parameter corruptions. The vulnerability of model parameters is of crucial value to the study of model robustness and generalization. In this work, we introduce the concept of parameter corruption and propose to leverage the loss change indicators for measuring the flatness of the loss basin and the parameter robustness of neural network parameters. On such basis, we analyze parameter corruptions and propose the multi-step adversarial corruption algorithm. To enhance neural networks, we propose the adversarial parameter defense algorithm that minimizes the average risk of multiple adversarial parameter corruptions. Experimental results show that the proposed algorithm can improve both the parameter robustness and accuracy of neural networks.