arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2502.07495cs.NIcs.LG
收藏

LLM-Sketch:用LLM增强网络草图

LLM-Sketch: Enhancing Network Sketches with LLM

Peking University(北京大学);Zhejiang University(浙江大学);Tsinghua University(清华大学);Zhongguancun Laboratory(中关村实验室)

Yuanpeng Li, Zhen Xu, Zongwei Lv, Yannan Hu, Yong Cui, Tong Yang

AI总结:

针对现有机器学习优化网络草图缺乏动态适应性和训练成本高的问题,提出LLM-Sketch,利用数据包头部流ID外字段推断流大小,通过双层结构分离记录大小流并结合微调LLM,在最小化内存的同时实现7.5倍精度提升。

摘要:

网络流挖掘是许多网络操作的基础。草图作为紧凑的数据结构,具有低内存开销和有界精度,已成为网络流挖掘的一种有前景的解决方案。最近的研究尝试使用机器学习优化草图;然而,这些方法面临着缺乏对动态网络的适应性和训练成本高的挑战。在本文中,我们提出LLM-Sketch,基于的洞见是数据包头部中流ID之外的字段也有助于推断流大小。通过使用双层数据结构并分别记录大流量和小流量,LLM-Sketch在最小化内存使用的同时提高了准确性。此外,它利用微调的大型语言模型(LLM)可靠地估计流大小。我们在三个代表性任务上评估了LLM-Sketch,结果表明LLM-Sketch优于最先进的方法,实现了7.5倍的准确性提升。

英文摘要:

Network stream mining is fundamental to many network operations. Sketches, as compact data structures that offer low memory overhead with bounded accuracy, have emerged as a promising solution for network stream mining. Recent studies attempt to optimize sketches using machine learning; however, these approaches face the challenges of lacking adaptivity to dynamic networks and incurring high training costs. In this paper, we propose LLM-Sketch, based on the insight that fields beyond the flow IDs in packet headers can also help infer flow sizes. By using a two-tier data structure and separately recording large and small flows, LLM-Sketch improves accuracy while minimizing memory usage. Furthermore, it leverages fine-tuned large language models (LLMs) to reliably estimate flow sizes. We evaluate LLM-Sketch on three representative tasks, and the results demonstrate that LLM-Sketch outperforms state-of-the-art methods by achieving a $7.5\times$ accuracy improvement.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑