JavIndra/attack-cves-data

该数据集结合了MITRE ATT&CK（对抗性战术、技术和常见知识）和CVE（常见漏洞和暴露）的信息，为训练语言模型提供了网络安全概念和漏洞的全面资源。每个数据条目通常包括MITRE ATT&CK的策略和技术、CVE标识符和描述、ATT&CK技术与CVE之间的关系，以及与网络安全分析相关的额外元数据。数据集主要用于llama模型的微调，所有数据均来自MITRE的公共资源。

The dataset combines information from MITRE ATT&CK (Adversarial Tactics, Techniques, and Common Knowledge) and CVE (Common Vulnerabilities and Exposures) to provide a comprehensive resource for training language models on cybersecurity concepts and vulnerabilities. Each entry typically includes MITRE ATT&CK tactics and techniques, CVE identifiers and descriptions, relationships between ATT&CK techniques and CVEs, and additional metadata relevant to cybersecurity analysis. The dataset sources include MITRE ATT&CK, CVE-Attack mappings, and CVEs. The purpose of the dataset is to facilitate llama finetuning with MITRE information.