A clustering-guided LLM data augmentation pipeline raises macro-F1 for MITRE technique classification, e.g., ALBERT from 0.35 to 0.52 and SecureBERT to 0.66, across two CTI datasets.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
SynthCTI: LLM-Driven Synthetic CTI Generation to enhance MITRE Technique Mapping
A clustering-guided LLM data augmentation pipeline raises macro-F1 for MITRE technique classification, e.g., ALBERT from 0.35 to 0.52 and SecureBERT to 0.66, across two CTI datasets.