Skip to content
Search
paperSeptember 2026Unreviewed

PrivEscalate: Measuring and Augmenting the Threat of LLM-Automated Linux Privilege Escalation

Yixuan Liu, Zilong Zhen, Yin Wu, Yi Li

View resourcePDF

Abstract

As Large Language Model (LLM) agents increasingly automate offensive operations across the cyber kill chain, their efficacy in complex local post-exploitation tasks remains inadequately quantified. Among these, Linux privilege escalation is a key step between initial access and full system compromise. However, existing evaluations for this task are limited by small sample sizes (fewer than 15 scenarios), lacking the scale to compare model capabilities under executable verification. To address th

Categories

Threat Modeling

Cite

@misc{liu2026privescalate,
  title = {{PrivEscalate: Measuring and Augmenting the Threat of LLM-Automated Linux Privilege Escalation}},
  author = {Yixuan Liu and Zilong Zhen and Yin Wu and Yi Li},
  year = {2026},
  month = sep,
  eprint = {2609.09087},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.09087}
}