A Contrastive Dual-Task Framework for Few-Shot Traffic Classification in IoT Networks
Zikui Lu, Mo Chen, Sailong Cui, Bingbing Zhao, Yaoyuan Zheng · 2026
Read A Contrastive Dual-Task Framework for Few-Shot Traffic Classification in IoT Networks by Zikui Lu, Mo Chen, Sailong Cui, Bingbing Zhao, Yaoyuan Zheng (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/pmc-13258878.txt.
Description
Zikui Lu and coauthors teach network traffic classifiers to learn new classes from few examples in a 2026 computing paper. Encrypted sensor traffic hides its content, so models must judge from behavior patterns alone. Shared code libraries can blur those patterns and mislead the models.
Their framework pairs labeled distance training with a self taught masking task. Joint training builds features that split true behavior from library noise. Tests across nine datasets beat the strongest prior methods. Few example learning gained several points of precision with firm statistical support. Security teams watching device networks gain a classifier that adapts without huge new labeled sets.
Catalog record
| CREATOR | Zikui Lu, Mo Chen, Sailong Cui, Bingbing Zhao, Yaoyuan Zheng |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 47 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |