Single-label classification technology has difficulty meeting the needs of text classification, and multi-label text classification has become an important research issue in natural language processing (NLP). Extracting semantic features from different levels and granularities of text is a basic and key task in multi-label text classification research. A topic model is an effective method for the automatic organization and induction of text information. It can reveal the latent semantics of documents and analyze the topics contained in massive information. Therefore, this paper proposes a multi-label text classification method based on tALBERT-CNN: an LDA topic model and ALBERT model are used to obtain the topic vector and semantic context vector of each word (document), a certain fusion mechanism is adopted to obtain in-depth topic and semantic representations of the document, and the multi-label features of the text are extracted through the TextCNN model to train a multi-label classifier. The experimental results obtained on standard datasets show that the proposed method can extract multi-label features from documents, and its performance is better than that of the existing state-of-the-art multi-label text classification algorithms.

Binary code similarity detection is used to calculate the code similarity of a pair of binary functions or files, through a certain calculation method and judgment method. It is a fundamental task in the field of computer binary security. Traditional methods of similarity detection usually use graph matching algorithms, but these methods have poor performance and unsatisfactory effects. Recently, graph neural networks have become an effective method for analyzing graph embeddings in natural language processing. Although these methods are effective, the existing methods still do not sufficiently learn the information of the binary code. To solve this problem, we propose Codeformer, an iterative model of a graph neural network (GNN)-nested Transformer. The model uses a Transformer to obtain an embedding vector of the basic block and uses the GNN to update the embedding vector of each basic block of the control flow graph (CFG). Codeformer iteratively executes basic block embedding to learn abundant global information and finally uses the GNN to aggregate all the basic blocks of a function. We conducted experiments on the OpenSSL, Clamav and Curl datasets. The evaluation results show that our method outperforms the state-of-the-art models.

In the field of network security, although there has been related work on software vulnerability detection based on classic machine learning, detection ability is directly proportional to the scale of training data. A quantum neural network has been proven to solve the memory bottleneck problem of classical machine learning, so it has far-reaching prospects in the field of vulnerability detection. To fill the gap in this field, we propose a quantum neural network structure named QDENN for software vulnerability detection. This work is the first attempt to implement word embedding of vulnerability codes based on a quantum neural network, which proves the feasibility of a quantum neural network in the field of vulnerability detection. Experiments demonstrate that our proposed QDENN can effectively solve the inconsistent input length problem of quantum neural networks and the problem of batch processing with long sentences. Furthermore, it can give full play to the advantages of quantum computing and realize a vulnerability detection model at the cost of a small amount of measurement. Compared to other quantum neural networks, our proposed QDENN can achieve higher vulnerability detection accuracy. On the sub dataset with a small-scale interval, the model accuracy rate reaches 99%. On each subinterval data, the best average vulnerability detection accuracy of the model reaches 86.3%.

In recent years, the number of smart contracts running in the blockchain has increased rapidly, accompanied by many security problems, such as vulnerability propagation caused by code reuse or vicious transaction caused by malicious contract deployment, for example. Most smart contracts do not publish the source code, but only the bytecode. Based on the research of bytecode similarity of smart contract, smart contract upgrade, vulnerability search and malicious contract analysis can be carried out. The difficulty of bytecode similarity research is that different compilation versions and optimization options lead to the diversification of bytecode of the same source code. This paper presents a solution, including a series of methods to measure the similarity of smart contract bytecode. Starting from the opcode of smart contract, a method of pre-training the basic block sequence of smart contract is proposed, which can embed the basic block vector. Positive samples were obtained by basic block marking, and the negative sampling method is improved. After these works, we put the obtained positive samples, negative samples and basic blocks themselves into the triplet network composed of transformers. Our solution can obtain evaluation results with an accuracy of 97.8%, so that the basic block sequence of optimized and unoptimized options can be transformed into each other. At the same time, the instructions are normalized, and the order of compiled version instructions is normalized. Experiments show that our solution can effectively reduce the bytecode difference caused by optimization options and compiler version, and improve the accuracy by 1.4% compared with the existing work. We provide a data set covering 64 currently used Solidity compilers, including one million basic block pairs extracted from them.

Short text representation is one of the basic and key tasks of NLP. The traditional method is to simply merge the bag-of-words model and the topic model, which may lead to the problem of ambiguity in semantic information, and leave topic information sparse. We propose an unsupervised text representation method that involves fusing word embeddings and extended topic information. Following this, two fusion strategies of weighted word embeddings and extended topic information are designed: static linear fusion and dynamic fusion. This method can highlight important semantic information, flexibly fuse topic information, and improve the capabilities of short text representation. We use classification and prediction tasks to verify the effectiveness of the method. The testing results show that the method is valid.