Author:
Song Zhihui,Zhou Xin,Xu Jinchen,Ding Xiaodong,Shan Zheng
Abstract
AbstractIn recent years, deep learning has been widely used in vulnerability detection with remarkable results. These studies often apply natural language processing (NLP) technologies due to the natural similarity between code and language. Since NLP usually consumes a lot of computing resources, its combination with quantum computing is becoming a valuable research direction. In this paper, we present a Recurrent Quantum Embedding Neural Network (RQENN) for vulnerability detection. It aims to reduce the memory consumption of classical models for vulnerability detection tasks and improve the performance of quantum natural language processing (QNLP) methods. We show that the performance of RQENN achieves the above goals. Compared with the classic model, the space complexity of each stage of its execution is exponentially reduced, and the number of parameters used and the number of bits consumed are significantly reduced. Compared with other QNLP methods, RQENN uses fewer qubit resources and achieves a 15.7% higher accuracy in vulnerability detection.
Funder
Major Science and Technology Projects in Henan Province,China
Publisher
Springer Science and Business Media LLC
Reference51 articles.
1. Feng, Z. et al. CodeBERT: A pre-trained model for programming and natural languages. In Findings of the Association for Computational Linguistics: EMNLP 2020 (eds. Cohn, T., He, Y. & Liu, Y.). 1536–1547 https://doi.org/10.18653/v1/2020.findings-emnlp.139 (Association for Computational Linguistics, 2020).
2. Jiang, X., Zheng, Z., Lyu, C., Li, L. & Lyu, L. TreeBERT: A tree-based pre-trained model for programming language. In Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence. 54–63 (PMLR, 2021).
3. Wang, Y., Wang, W., Joty, S. & Hoi, S. C. H. CodeT5: Identifier-aware unified pre-trained encoder-decoder models for code understanding and generation. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (eds. Moens, M.-F., Huang, X., Specia, L. & Yih, S. W.). 8696–8708 https://doi.org/10.18653/v1/2021.emnlp-main.685 (Association for Computational Linguistics, 2021).
4. Aghaei, E., Niu, X., Shadid, W. & Al-Shaer, E. SecureBERT: A domain-specific language model for cybersecurity. In Security and Privacy in Communication Networks (eds. Li, F., Liang, K., Lin, Z. & Katsikas, S. K.). 39–56 https://doi.org/10.1007/978-3-031-25538-0_3 (Springer, 2023).
5. Xiang, G., Shi, C. & Zhang, Y. An APT event extraction method based on BERT-BiGRU-CRF for APT attack detection. Electronics 12, 3349 (2023).