CVE-Fix-RAG: Multi-Language Security Patches for RAG

2,000+ real-world security patches enriched with BGE embeddings for Retrieval-Augmented Generation. The dataset appears to be sourced from Kaggle and focuses on multi-language vulnerability fixes. Its author, organization, and last update date are unknown.

Use Cases

• Training patch generation models based on the collection of real-world security fixes.
• Benchmarking RAG systems for cybersecurity applications using the provided BGE embeddings.
• Analyzing multi-language vulnerability patterns from the described patch data.
• Developing tools for automated vulnerability detection and remediation based on historical patch examples.

Strengths

• Contains over 2,000 real-world security patches.
• Includes BGE embeddings pre-computed for RAG applications.
• Focuses on multi-language vulnerability fixes, which suggests a diverse source.

Limitations

• Row count and file size are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

Kaggle

Collection Method

Likely scraped or aggregated from public vulnerability databases and code repositories.