GRAML: Graph-Grounded Reasoning and Multi-Task Learning for LLM-Based Software Vulnerability Detection
GRAML, a framework that combines graph evidence, vulnerability description generation, and multi-task training, is proposed, and practical guidance for building more reliable and secure software engineering systems with large language models is provided.
Ruo-Han Li, Miao-Qing Tian, Hai-Peng Qu et al.
· 0 citations