Skip to content
Search
paperMay 2026Unreviewed

Towards Demystifying and Repairing LLM-in-the-Loop Vulnerabilities

Yujie Ma, Jialin Rong, Chenxi Yang, Lili Quan, Xiaofei Xie, Yongqiang Lyu, Qiang Hu

View resourcePDF

Abstract

Large Language Models(LLMs) have been actively integrated into modern software systems as critical components. LLM-in-the-loop vulnerabilities, where vulnerabilities are introduced by LLMs and their dependent downstream components, such as frameworks, introduce new risks. Although some benchmark datasets have been constructed to study the impact of such vulnerabilities, most works still remain at the analysis from the conventional software level, ignoring the harm actually caused by LLMs. Unders

Categories

Benchmarks & Evaluation

Cite

@misc{ma2026demystifying,
  title = {{Towards Demystifying and Repairing LLM-in-the-Loop Vulnerabilities}},
  author = {Yujie Ma and Jialin Rong and Chenxi Yang and Lili Quan and Xiaofei Xie and Yongqiang Lyu and Qiang Hu},
  year = {2026},
  month = may,
  eprint = {2605.28893},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2605.28893}
}