Tri-LLM Cooperative Federated Zero-Shot Intrusion Detection with Semantic Disagreement and Trust-Aware Aggregation
Saeid Jamshidi, Omar Abdul-Wahab, Foutse Khomh, Kawser Wazed Nafi · 2026
Read Tri-LLM Cooperative Federated Zero-Shot Intrusion Detection with Semantic Disagreement and Trust-Aware Aggregation by Saeid Jamshidi, Omar Abdul-Wahab, Foutse Khomh, Kawser Wazed Nafi (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-2602-00219.txt.
Description
Saeid Jamshidi and coauthors build a federated intrusion detection system for zero shot, open set attacks in cyber physical and Internet of Things networks. Their semantics driven frame adds language derived supervision to federated tuning, enabling detection of unseen attack behaviors. Attack prototypes come from a Tri-LLM ensemble of GPT-4o, DeepSeek-V3, and LLaMA-3-8B, aligning clients around shared meaning rather than shared raw data. The design thus aims at new attacks without pooling private traffic in one place.
Catalog record
| CREATOR | Saeid Jamshidi, Omar Abdul-Wahab, Foutse Khomh, Kawser Wazed Nafi |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 85808 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602.00219