From the 1 of 4 linked papers with an AI index.
4 papers
MARS: Multi-stage Accelerated Read Stack for Large-buffer Buffered Reads
Yang Shen, Kai Lu, Min Xie +3
MARS is a multi‑stage read stack for Linux that treats each large buffered read as a single unit, stages page‑cache operations, and parallelizes data copies and I/O to dramatically…
Rethinking Burst Buffer Optimization: Enabling Layout Heterogeneity via Hybrid Analysis and LLM Guidance
Yuhan Cai, Huijun Wu, Zhuo Tang +6
Burst buffers (BBs) are essential for mitigating I/O bottlenecks in modern HPC systems. However, existing BB file systems often suffer from structural performance degradation due t…
Perseus: Leveraging Common Data Patterns with Curriculum Learning for More Robust Graph Neural Networks
Kaiwen Xia, Huijun Wu, Duanyu Li +3
Graph Neural Networks (GNNs) excel at handling graph data but remain vulnerable to adversarial attacks. Existing defense methods typically rely on assumptions like graph sparsity a…
Talos: A More Effective and Efficient Adversarial Defense for GNN Models Based on the Global Homophily of Graphs
Duanyu Li, Huijun Wu, Min Xie +3
Graph neural network (GNN) models play a pivotal role in numerous tasks involving graph-related data analysis. Despite their efficacy, similar to other deep learning models, GNNs a…