Ke Zhou

TL;DR: This survey considered 15 security vulnerabilities in smart contracts and introduced the vulnerable areas and the causes of vulnerabilities, and found that a new attack cannot be detected by existing detection tools if the vulnerability without pre-defined is found.

TL;DR: Wang et al. as discussed by the authors proposed a lightweight smart contract vulnerability detection model based on Convolutional Neural Networks (CNN), which can automatically detect the vulnerabilities in the smart contract on a lightweight computer without expert knowledge.

TL;DR: Li et al. as discussed by the authors proposed a transfer learning method based on ResNet model to solve the problem of efficient classification of small-scale garbage image data sets, which can effectively improve the training speed and accuracy, and reduce the impact of over-fitting.

TL;DR: Wang et al. as discussed by the authors explored the application of attention-based convolutional neural networks for winter wheat identification on GF-2 high-resolution images and proposed improvements to enhance recognition accuracy.

TL;DR: In this article, a multi-level feature contrastive network (MLFCNet) is proposed to align the two domains in the label space by iteratively using clustering algorithm to obtain the pseudo-labels and then minimizing Multi-Level Contrastive Discrepancy (MLCD) loss.