Take Package as Language: Anomaly Detection Using Transformer
Jie Huang · 2024
Read Take Package as Language: Anomaly Detection Using Transformer by Jie Huang (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-04473.txt.
Description
Guarding a computer network means spotting hostile packets in a flood of normal ones. Jie Huang treats those packets as a language in this 2024 paper and hands them to a GPT-style model. Each number in a packet becomes its own word, a finer view than the usual field-based split, and a special tokenizer and embedding layer suit the model to packet structure.
The system, NIDS-GPT, can pretrain without labels and shows its reasoning through attention maps. On two datasets it reached perfect accuracy under extreme class imbalance, far past classic methods, and topped ninety percent when learning from a single example per class. Huang presents it as a strong fit for imbalanced, resource-poor settings. The code is public.
Catalog record
| CREATOR | Jie Huang |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 38 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-04473