"Energon": Unveiling Transformers from GPU Power and Thermal Side-Channels
Arunava Chaudhuri, Shubhi Shukla, Sarani Bhattacharya, and Debdeep Mukhopadhyay

TL;DR
This paper uncovers GPU power and thermal side-channels as a new attack vector to extract architectural details from pre-trained transformer models, revealing significant security vulnerabilities in shared GPU environments.
Contribution
It is the first study to analyze GPU power and thermal fluctuations as side-channels for extracting transformer model architecture information.
Findings
Achieves over 89% accuracy in model family identification
100% accuracy in hyperparameter classification
Over 93% success rate in transfer adversarial attacks
Abstract
Transformers have become the backbone of many Machine Learning (ML) applications, including language translation, summarization, and computer vision. As these models are increasingly deployed in shared Graphics Processing Unit (GPU) environments via Machine Learning as a Service (MLaaS), concerns around their security grow. In particular, the risk of side-channel attacks that reveal architectural details without physical access remains underexplored, despite the high value of the proprietary models they target. This work to the best of our knowledge is the first to investigate GPU power and thermal fluctuations as side-channels and further exploit them to extract information from pre-trained transformer models. The proposed analysis shows how these side channels can be exploited at user-privilege to reveal critical architectural details such as encoder/decoder layer and attention head…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAdversarial Robustness in Machine Learning · Security and Verification in Computing · Physical Unclonable Functions (PUFs) and Hardware Security
