Backdoor Attack against Log Anomaly Detection Models
Blog framework illustrationAbstract
We propose Blog, a backdoor attack framework targeting log anomaly detection models. Blog embeds carefully crafted triggers into training logs, enabling adversaries to manipulate predictions at inference while preserving benign performance. Experiments show that state-of-the-art log anomaly detection systems are highly vulnerable, underscoring the need for stronger defenses.
Type
Publication
In Companion Proceedings of the ACM Web Conference 2025
Citation
@inproceedings{cheng2025backdoor,
title={Backdoor Attack against Log Anomaly Detection Models},
author={Cheng, He and Xu, Depeng and Yuan, Shuhan},
booktitle={Companion Proceedings of the ACM on Web Conference 2025},
pages={915--918},
year={2025}
}
