Topologies, Checkpoints, and Configurations for the paper "GVI-RL: Graph-Invariant RL for Attack Paths Discovery using Vulnerabilities Embedded with Large Language Models"

This repository consists of the files related to the paper "GVI-RL: Graph-Invariant RL for Attack Paths Discovery using Vulnerabilities Embedded with Large Language Models". In particular, this repository contains tensorboard logs, topologies, checkpoints, seeds, and results to ensure reproducibility of the results of the paper. The results included are related to the training/validation and hyper-parameters optimization of the outcome multi-label classifier, the GVI-RL agent, and the world model.The data folder contains also the topologies used in the study, the vulnerabilities data used to generate them and the dataset for multi-label classification. The README.md file describes the folders' structure.本仓库收录与论文《GVI-RL：基于嵌入大语言模型（Large Language Model）漏洞的攻击路径发现图不变强化学习》相关的全部文件。具体而言，本仓库包含TensorBoard日志、拓扑结构文件、模型检查点、随机种子文件与实验结果，以确保该论文实验结果的可复现性。 本次收录的实验结果涵盖所构建的多标签分类器、GVI-RL智能体（AI Agent）以及世界模型的训练、验证流程与超参数优化相关内容。数据文件夹中还包含本研究使用的拓扑结构、用于生成这些拓扑的漏洞数据，以及多标签分类任务专属数据集。 README.md文件对本仓库内各文件夹的组织结构进行了详细说明。