Shengyao Zhang — accepted AI-conference papers

2023

SlowBERT: Slow-down Attacks on Input-adaptive Multi-exit BERT

Shengyao Zhang, Xudong Pan, Mi Zhang, Min Yang

ACL 2023findings

For pretrained language models such as Google’s BERT, recent research designs several input-adaptive inference mechanisms to improve the efficiency on cloud and edge devices. In this paper, we reveal a new attack surface on input-adaptive multi-exit BERT, where the adversary imperceptibly modifies t…

Cited by 2 Source ↗PDF ↗Scholar ↗

2022

House of Cans: Covert Transmission of Internal Datasets via Capacity-Aware Neuron Steganography

Xudong Pan, Shengyao Zhang, Mi Zhang, Yifan Yan, Min Yang

NeurIPS 2022accept

In this paper, we present a capacity-aware neuron steganography scheme (i.e., Cans) to covertly transmit multiple private machine learning (ML) datasets via a scheduled-to-publish deep neural network (DNN) as the carrier model. Unlike existing steganography schemes which treat the DNN parameters as…

Cited by 2 Source ↗PDF ↗Scholar ↗