Minkyung Park (University of Texas at Dallas), Zelun Kong (University of Texas at Dallas), Dave (Jing) Tian (Purdue University), Z. Berkay Celik (Purdue University), Chung Hwan Kim (University of Texas at Dallas)

Deep neural networks (DNNs) are integral to modern computing, powering applications such as image recognition, natural language processing, and audio analysis. The architectures of these models (e.g., the number and types of layers) are considered valuable intellectual property due to the significant expertise and computational effort required for their design. Although trusted execution environments (TEEs) like Intel SGX have been adopted to safeguard these models, recent studies on model extraction attacks have shown that side-channel attacks (SCAs) can still be leveraged to extract the architectures of DNN models. However, many existing model extraction attacks either do not account for TEE protections or are limited to specific model types, reducing their real-world applicability.

In this paper, we introduce DNN Latency Sequencing (DLS), a novel model extraction attack framework that targets DNN architectures running within Intel SGX enclaves. DLS employs SGX-Step to single-step model execution and collect fine-grained latency traces, which are then analyzed at both the function and basic block levels to reconstruct the model architecture. Our key insight is that DNN architectures inherently influence execution behavior, enabling accurate reconstruction from latency patterns. We evaluate DLS on models built with three widely used deep learning libraries, Darknet, TensorFlow Lite, and ONNX Runtime, and show that it achieves architecture recovery accuracies of 97.3%, 96.4%, and 93.6%, respectively. We further demonstrate that DLS enables advanced attacks, highlighting its practicality and effectiveness.

View More Papers

From Noise to Signal: Precisely Identify Affected Packages of...

Yingyuan Pu (QI-ANXIN Technology Research Institute), Lingyun Ying (QI-ANXIN Technology Research Institute), Yacong Gu (Tsinghua University, Tsinghua University-QI-ANXIN Group JCNS)

Read More

Chasing Shadows: Pitfalls in LLM Security Research

Jonathan Evertz (CISPA Helmholtz Center for Information Security), Niklas Risse (Max Planck Institute for Security and Privacy), Nicolai Neuer (Karlsruhe Institute of Technology), Andreas Müller (Ruhr University Bochum), Philipp Normann (TU Wien), Gaetano Sapia (Max Planck Institute for Security and Privacy), Srishti Gupta (Sapienza University of Rome), David Pape (CISPA Helmholtz Center for Information Security),…

Read More

Efficiently Detecting DBMS Bugs through Bottom-up Syntax-based SQL Generation

Yu Liang (The Pennsylvania State University), Peng Liu (The Pennsylvania State University)

Read More