40 citations · 65 across the 7 of their papers we have counts for
5 papers · 1 filter
Trojaning Language Models for Fun and Profit
Xinyang Zhang, Zheng Zhang, Shouling Ji +1
Recent years have witnessed the emergence of a new paradigm of building natural language processing (NLP) systems: general-purpose, pre-trained language models (LMs) are composed w…
Interpretable Deep Learning under Fire
Xinyang Zhang, Ningfei Wang, Hua Shen +3
Providing explanations for deep neural network (DNN) models is crucial for their use in security-sensitive domains. A plethora of interpretation models have been proposed to help u…
EagleEye: Attack-Agnostic Defense against Adversarial Inputs (Technical Report)
Yujie Ji, Xinyang Zhang, Ting Wang
Deep neural networks (DNNs) are inherently vulnerable to adversarial inputs: such maliciously crafted samples trigger DNNs to misbehave, leading to detrimental consequences for DNN…
Differentially Private Releasing via Deep Generative Model (Technical Report)
Xinyang Zhang, Shouling Ji, Ting Wang
Privacy-preserving releasing of complex data (e.g., image, text, audio) represents a long-standing challenge for the data mining research community. Due to rich semantics of the da…
Modular Learning Component Attacks: Today's Reality, Tomorrow's Challenge
Xinyang Zhang, Yujie Ji, Ting Wang
Many of today's machine learning (ML) systems are not built from scratch, but are compositions of an array of {\em modular learning components} (MLCs). The increasing use of MLCs s…