Saved in:
Bibliographic Details
Main Authors: La O, Reynier Leyva, Catania, Carlos A., Parlanti, Tatiana
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2411.03307
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915006379982848
author La O, Reynier Leyva
Catania, Carlos A.
Parlanti, Tatiana
author_facet La O, Reynier Leyva
Catania, Carlos A.
Parlanti, Tatiana
contents This work analyzes the use of large language models (LLMs) for detecting domain generation algorithms (DGAs). We perform a detailed evaluation of two important techniques: In-Context Learning (ICL) and Supervised Fine-Tuning (SFT), showing how they can improve detection. SFT increases performance by using domain-specific data, whereas ICL helps the detection model to quickly adapt to new threats without requiring much retraining. We use Meta's Llama3 8B model, on a custom dataset with 68 malware families and normal domains, covering several hard-to-detect schemes, including recent word-based DGAs. Results proved that LLM-based methods can achieve competitive results in DGA detection. In particular, the SFT-based LLM DGA detector outperforms state-of-the-art models using attention layers, achieving 94% accuracy with a 4% false positive rate (FPR) and excelling at detecting word-based DGA domains.
format Preprint
id arxiv_https___arxiv_org_abs_2411_03307
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle LLMs for Domain Generation Algorithm Detection
La O, Reynier Leyva
Catania, Carlos A.
Parlanti, Tatiana
Computation and Language
Cryptography and Security
This work analyzes the use of large language models (LLMs) for detecting domain generation algorithms (DGAs). We perform a detailed evaluation of two important techniques: In-Context Learning (ICL) and Supervised Fine-Tuning (SFT), showing how they can improve detection. SFT increases performance by using domain-specific data, whereas ICL helps the detection model to quickly adapt to new threats without requiring much retraining. We use Meta's Llama3 8B model, on a custom dataset with 68 malware families and normal domains, covering several hard-to-detect schemes, including recent word-based DGAs. Results proved that LLM-based methods can achieve competitive results in DGA detection. In particular, the SFT-based LLM DGA detector outperforms state-of-the-art models using attention layers, achieving 94% accuracy with a 4% false positive rate (FPR) and excelling at detecting word-based DGA domains.
title LLMs for Domain Generation Algorithm Detection
topic Computation and Language
Cryptography and Security
url https://arxiv.org/abs/2411.03307