Assessing the Effectiveness of Vulnerability Detection via Prompt Tuning: An Empirical Study

• Published in: Proceedings / Asia Pacific Software Engineering Conference pp. 415 - 424
• Main Authors: Lu, Guilong, Ju, Xiaolin, Chen, Xiang, Yang, Shaoyu, Chen, Liang, Shen, Hao
• Format: Conference Proceeding
• Language: English
• Published: IEEE 04.12.2023
• Subjects: Cross-domain vulnerability detection; Deep learning; Prompt tuning; Software engineering; Task analysis; Tuning; Vulner-ability type detection; Vulnerability detection
• ISSN: 2640-0715
• DOI: 10.1109/APSEC60848.2023.00052

In vulnerability detection approaches based on deep learning, fine-tuning with Pre-trained Language Models (PLMs) is a prevalent technique. Unfortunately, a natural gap exists between model pre-training tasks and vulnerability detection tasks due to different input formats, and the performance of fi...