SecReranker

Cybersecurity-specialized cross-encoder reranker trained on SecEmbed contrastive pairs (ATT&CK, Sigma, CVE, CWE, SOC playbooks).

Base model

cross-encoder/ms-marco-MiniLM-L-6-v2

Training

• Dataset: alirezaaminzadeh/secembed-pairs
• Loss: MultipleNegativesRankingLoss (bi-encoder) / Binary cross-entropy (reranker)
• Hard negatives: sibling ATT&CK techniques, same-CWE CVEs, unrelated Sigma rules
• Hardware: Hugging Face ZeroGPU

Evaluation (CyberSec Retrieval Benchmark sample)

{
  "note": "pair classification trained; use demo Space for retrieval@k with bi-encoder cascade"
}

Usage

from sentence_transformers import CrossEncoder

model = CrossEncoder("alirezaaminzadeh/SecReranker")
print(model.predict([["detect powershell encoded command", "T1059.001 PowerShell..."]]))

Intended use

RAG over security knowledge bases, threat-intelligence search, SOC alert enrichment, CVE/CWE similarity, Sigma rule retrieval, and ATT&CK technique mapping.