Výsledky vyhledávání

Akademický článek

Identification method for malicious traffic in industrial Internet under new unknown attack scenarios

Autor: ZENG Fanyi, MAN Dapeng, XU Chen, HAN Shuai, WANG Huanran, ZHOU Xue, LI Xinchun, YANG Wu

Publikováno v: Tongxin xuebao, Vol 45, Pp 75-86 (2024)

Aiming at the problem of traffic data distribution shift caused by new unknown attacks in the industrial Internet, a malicious traffic identification method based on neighborhood filtering and stable learning was proposed to enhance the effectiveness

Externí odkaz: https://doaj.org/article/0dea3ab755ad4bc7bec791da8896a45c

Zobrazit plný text záznamu

Report

Act in Collusion: A Persistent Distributed Multi-Target Backdoor in Federated Learning

Autor: Liu, Tao, Yang, Wu, Xu, Chen, Lv, Jiguang, Wang, Huanran, Zhang, Yuhang, Xu, Shuchun, Man, Dapeng

Federated learning, a novel paradigm designed to protect data privacy, is vulnerable to backdoor attacks due to its distributed nature. Current research often designs attacks based on a single attacker with a single backdoor, overlooking more realist

Externí odkaz: http://arxiv.org/abs/2411.03926

Zobrazit plný text záznamu

Report

Revisiting Interpolation Augmentation for Speech-to-Text Generation

Autor: Xu, Chen, Wang, Jie, Liu, Xiaoqian, Dong, Qianqian, Zhang, Chunliang, Xiao, Tong, Zhu, Jingbo, Man, Dapeng, Yang, Wu

Speech-to-text (S2T) generation systems frequently face challenges in low-resource scenarios, primarily due to the lack of extensive labeled datasets. One emerging solution is constructing virtual training samples by interpolating inputs and labels,

Externí odkaz: http://arxiv.org/abs/2406.15846

Zobrazit plný text záznamu

Report

Beyond Traditional Threats: A Persistent Backdoor Attack on Federated Learning

Autor: Liu, Tao, Zhang, Yuhang, Feng, Zhu, Yang, Zhiqin, Xu, Chen, Man, Dapeng, Yang, Wu

Publikováno v: Proceedings of the AAAI Conference on Artificial Intelligence. 2024, 38(19): 21359-21367

Backdoors on federated learning will be diluted by subsequent benign updates. This is reflected in the significant reduction of attack success rate as iterations increase, ultimately failing. We use a new metric to quantify the degree of this weakene

Externí odkaz: http://arxiv.org/abs/2404.17617

Zobrazit plný text záznamu

Report

Bridging the Gaps of Both Modality and Language: Synchronous Bilingual CTC for Speech Translation and Speech Recognition

Autor: Xu, Chen, Liu, Xiaoqian, He, Erfeng, Zhang, Yuhao, Dong, Qianqian, Xiao, Tong, Zhu, Jingbo, Man, Dapeng, Yang, Wu

In this study, we present synchronous bilingual Connectionist Temporal Classification (CTC), an innovative framework that leverages dual CTC to bridge the gaps of both modality and language in the speech translation (ST) task. Utilizing transcript an

Externí odkaz: http://arxiv.org/abs/2309.12234

Zobrazit plný text záznamu