Backdoor attacks have raised significant concerns in machine learning (ML) systems.
Mainstream ML backdoor attacks typically involve either poisoning the victim's training …

Efficient management of GPU memory is essential for high throughput LLM inference. Prior
systems used to reserve KV-cache memory ahead-of-time that resulted in wasted capacity …

While extensive research has been conducted on CPU cache side-channel attacks, the
landscape of similar studies on modern GPUs remains largely uncharted. In this paper, we …

As modern applications increasingly rely on GPUs to accelerate the computation, it has
become very critical to study and understand the security implications of GPUs. In this work …

A critical component of high-throughput processors such as GPUs is the network-on-chip
(NoC) that interconnects the large number of cores and the memory partitions together. In …

In recent years, the widespread informatization and rapid data explosion have increased the
demand for high-performance heterogeneous systems that integrate multiple computing …

AI applications are employed in diverse scenarios, including data centers, personal
computers, smart cars, and so on. Their privacy is threatened by the intricate software stacks …

NVIDIA's Multi-Instance GPU (MIG) technology enables partitioning GPU computing power
and memory into separate hardware instances, providing complete isolation including …

Modern GPU applications, such as machine learning (ML), can only partially utilize GPUs,
leading to GPU underutilization in cloud environments. Sharing GPUs across multiple …

With the emergence of GPUs as first-class compute engines, more concentrated focus has
been put into covert and side channel discovery in these architectures. However, most of the …