7 papers
OptiLeak: Efficient Prompt Reconstruction via Reinforcement Learning in Multi-tenant LLM Services
Longxiang Wang, Xiang Zheng, Xuhao Zhang +3
Multi-tenant LLM serving frameworks widely adopt shared Key-Value caches to enhance efficiency. However, this creates side-channel vulnerabilities enabling prompt leakage attacks.…
Rethinking Latency Denial-of-Service: Attacking the LLM Serving Framework, Not the Model
Tianyi Wang, Huawei Fan, Yuanchao Shu +2
LLM inference is inherently expensive, even a modest slowdown can translate into substantial operating costs and severe availability risks. Recently, a growing body of research kno…
PubSub-VFL: Towards Efficient Two-Party Split Learning in Heterogeneous Environments via Publisher/Subscriber Architecture
Yi Liu, Yang Liu, Leqian Zheng +5
With the rapid advancement of the digital economy, data collaboration between organizations has become a well-established business model, driving the growth of various industries.…
Exploring code portability solutions for HEP with a particle tracking test code
Hammad Ather, Sophie Berkman, Giuseppe Cerati +10
Traditionally, high energy physics (HEP) experiments have relied on x86 CPUs for the majority of their significant computing needs. As the field looks ahead to the next generation…
HORAM: A High-Performance Hierarchical Doubly Oblivious RAM
Leqian Zheng, Zheng Zhang, Wentao Dong +3
The combination of Oblivious RAM (ORAM) with Trusted Execution Environments (TEE) has found numerous real-world applications due to their complementary nature. TEEs alleviate the p…
Caching Aided Multi-Tenant Serverless Computing
Chu Qiao, Cong Wang, Zhenkai Zhang +2
One key to enabling high-performance serverless computing is to mitigate cold-starts. Current solutions utilize a warm pool to keep function alive: a warm-start can be analogous to…
Application of performance portability solutions for GPUs and many-core CPUs to track reconstruction kernels
Ka Hei Martin Kwok, Matti Kortelainen, Giuseppe Cerati +11
Next generation High-Energy Physics (HEP) experiments are presented with significant computational challenges, both in terms of data volume and processing power. Using compute acce…