14 papers
zkComposer: Decomposing Proof Construction to Scale zkML
Pawan Kumar Sanjaya, Christina Giannoula, Valdy Oktavian +4
Zero-knowledge machine learning (zkML) enables a server to perform verifiable inference while keeping model parameters private from the client. However, existing zkML systems incur…
AutoPRAC: Automating Attack Discovery for PRAC-Based Rowhammer Defenses using Model Checkers
Joyce Qu, Gururaj Saileshwar
Per-Row Activation Counting (PRAC) in DDR5 is a specification to mitigate Rowhammer attacks by tracking activations per row and triggering mitigative refreshes when needed. However…
GPUBreach: Privilege Escalation Attacks on GPUs using Rowhammer
Chris S. Lin, Yuqin Yan, Guozhen Ding +4
NVIDIA GPUs with GDDR memories have been shown susceptible to Rowhammer-based bit-flips, similar to CPUs. However, Rowhammer exploits on GPUs have been limited to injecting untarge…
When Speculation Spills Secrets: Side Channels via Speculative Decoding In LLMs
Jiankun Wei, Abdulrahman Abdulrazzag, Tianchen Zhang +2
Deployed large language models (LLMs) often rely on speculative decoding, a technique that generates and verifies multiple candidate tokens in parallel, to improve throughput and l…
CLIP: Client-Side Invariant Pruning for Mitigating Stragglers in Secure Federated Learning
Anthony DiMaggio, Raghav Sharma, Gururaj Saileshwar
Secure federated learning (FL) preserves data privacy during distributed model training. However, deploying such frameworks across heterogeneous devices results in performance bott…
PrisonBreak: Jailbreaking Large Language Models with at Most Twenty-Five Targeted Bit-flips
Zachary Coalson, Jeonghyun Woo, Chris S. Lin +8
We study a new vulnerability in commercial-scale safety-aligned large language models (LLMs): their refusal to generate harmful responses can be broken by flipping only a few bits…