GMA-SAWGAN-GP: A Novel Data Generative Framework to Enhance IDS Detection Performance
Ziyu Mu, Xiyu Shi, Safak Dogan · 2026
Read GMA-SAWGAN-GP: A Novel Data Generative Framework to Enhance IDS Detection Performance by Ziyu Mu, Xiyu Shi, Safak Dogan (2026) 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-2603-28838.txt.
Description
Mu, Shi, and Dogan present a data making framework to boost intrusion detection. Detection tools learn known attacks well but miss new ones. Their model builds on a Wasserstein GAN with gradient penalty and self-attention. A Gumbel-Softmax part handles discrete fields, and an autoencoder keeps the data on a sound manifold. A small gating net balances the loss terms and cuts mode collapse. Tests on three datasets raised detection on known attacks. Gains were larger for unseen attacks, with clear rises in key scores.
Catalog record
| CREATOR | Ziyu Mu, Xiyu Shi, Safak Dogan |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 69379 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.28838