6 papers
Identifying Good and Bad Neurons for Task-Level Controllable LLMs
Wenjie Li, Guansong Pang, Hezhe Qiao +2
Large Language Models have demonstrated remarkable capabilities on multiple-choice question answering benchmarks, but the complex mechanisms underlying their large-scale neurons re…
Automated TEE Adaptation with LLMs: Identifying, Transforming, and Porting Sensitive Functions in Programs
Ruidong Han, Zhou Yang, Chengyan Ma +5
Trusted Execution Environments (TEEs) isolate a special space within a device memory that is not accessible to the normal world (also known as the untrusted environment), even when…
Bamboo: LLM-Driven Discovery of API-Permission Mappings in the Android Framework
Han Hu, Wei Minn, Yonghui Liu +6
The permission mechanism in the Android Framework is integral to safeguarding the privacy of users by managing users' and processes' access to sensitive resources and operations. A…
DITING: A Static Analyzer for Identifying Bad Partitioning Issues in TEE Applications
Chengyan Ma, Ruidong Han, Jieke Shi +7
Trusted Execution Environment (TEE) enhances the security of mobile applications and cloud services by isolating sensitive code in the secure world from the non-secure normal world…
Towards Secure Program Partitioning for Smart Contracts with LLM's In-Context Learning
Ye Liu, Yuqing Niu, Chengyan Ma +5
Smart contracts are highly susceptible to manipulation attacks due to the leakage of sensitive information. Addressing manipulation vulnerabilities is particularly challenging beca…
Shelving it rather than Ditching it: Dynamically Debloating DEX and Native Methods of Android Applications without APK Modification
Zicheng Zhang, Jiakun Liu, Ferdian Thung +10
Today's Android developers tend to include numerous features to accommodate diverse user requirements, which inevitably leads to bloated apps. Yet more often than not, only a fract…