Explanation-augmented LoRA fine-tuning lets small LLMs detect phishing emails with accuracy and F1 around 0.94 to 0.96 on the SpamAssassin test set, while transferring to unseen datasets.
A bayesian approach to filtering junk e-mail
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Improving Phishing Email Detection Performance of Small Large Language Models
Explanation-augmented LoRA fine-tuning lets small LLMs detect phishing emails with accuracy and F1 around 0.94 to 0.96 on the SpamAssassin test set, while transferring to unseen datasets.