On protecting the data privacy of large language models (llms): A survey

B Yan, K Li, M Xu, Y Dong, Y Zhang, Z Ren… - arXiv preprint arXiv …, 2024 - arxiv.org
Large language models (LLMs) are complex artificial intelligence systems capable of
understanding, generating and translating human language. They learn language patterns …

{VeriSMo}: A verified security module for confidential {VMs}

Z Zhou, W Chen, S Gong, C Hawblitzel… - 18th USENIX Symposium …, 2024 - usenix.org
Hardware vendors have introduced confidential VM architectures (eg, AMD SEV-SNP, Intel
TDX and Arm CCA) in recent years. They eliminate the trust in the hypervisor and lead to the …

{UBA-Inf}: Unlearning Activated Backdoor Attack with {Influence-Driven} Camouflage

Z Huang, Y Mao, S Zhong - 33rd USENIX Security Symposium (USENIX …, 2024 - usenix.org
Machine-Learning-as-a-Service (MLaaS) is an emerging product to meet the market
demand. However, end users are required to upload data to the remote server when using …

AGORA: Open More and Trust Less in Binary Verification Service

H Chen, Q Zhou, S Yang, X Han, F Zhang… - arXiv preprint arXiv …, 2024 - arxiv.org
Binary verification plays a pivotal role in software security, yet building a verification service
that is both open and trustworthy poses a formidable challenge. In this paper, we introduce a …

The Design and Optimization of Memory Ballooning in SEV Confidential Virtual Machines

C Deng, Z Shen, D Li, Z Mi, Y Xia - 2024 IEEE International …, 2024 - ieeexplore.ieee.org
With the popularity of confidential computing, confidential virtual machines (CVMs) have
been widely adopted and they guarantee strong security by hardware. However, there still …

[PDF][PDF] VeriSMo: A Verified Security Module for Confidential VMs

W Chen, S Gong, C Hawblitzel, W Cui - sishuaigong.github.io
Hardware vendors have introduced confidential VM architectures (eg, AMD SEV-SNP, Intel
TDX and Arm CCA) in recent years. They eliminate the trust in the hypervisor and lead to the …