Scholarlink

As software systems grow more complex and supply chain attacks become more advanced, ensuring security and transparency is a global priority. This paper introduces VEXine, an AI-powered framework that automates the generation of Software Bill of Materials (SBOM) and Vulnerability Exploitability Exchange (VEX) reports using transformer-based large language models. By combining advanced AI with automated security assessment tools, VEXine enhances the accuracy, efficiency, and scalability of supply chain vulnerability management. Evaluations against traditional software composition analysis (SCA) tools show that VEXine delivers superior performance, providing actionable insights and addressing critical security challenges in modern software ecosystems.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

VEXine: Automating SBOM and VEX Generation Using Transformer LLM Models with LangGraph

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Su-Yeon Yoon,
  • Min-Ju Kwon,
  • Sang-Beom Kim,
  • Youn-Gee Kim,
  • Jong-Won Park,
  • Seung-Ho Son,
  • Young-Ok Kim,
  • Si-Hyeong Jeong

摘要

As software systems grow more complex and supply chain attacks become more advanced, ensuring security and transparency is a global priority. This paper introduces VEXine, an AI-powered framework that automates the generation of Software Bill of Materials (SBOM) and Vulnerability Exploitability Exchange (VEX) reports using transformer-based large language models. By combining advanced AI with automated security assessment tools, VEXine enhances the accuracy, efficiency, and scalability of supply chain vulnerability management. Evaluations against traditional software composition analysis (SCA) tools show that VEXine delivers superior performance, providing actionable insights and addressing critical security challenges in modern software ecosystems.