Cordyceps: Covert Control Attacks on LLMs via Data Poisoning

Abstract

Large language models (LLMs) are often fine-tuned on uncurated text datasets that adversaries can poison. Existing poisoning attacks primarily rely on fixed trigger phrases that defenses such as outlier detection, clean-data regularization, or online monitoring can neutralize. In this paper, we propose a data poisoning method that teaches an LLM an information hiding scheme reliably and stealthily through semantic associations between shared knowledge such as facts or concepts and attacker-chose

Categories

Framework mappings

Suggested from the entry's categories.

@misc{shao2026cordyceps,
  title = {{Cordyceps: Covert Control Attacks on LLMs via Data Poisoning}},
  author = {Zedian Shao and Charles Fleming and Teodora Baluta},
  year = {2026},
  month = may,
  eprint = {2605.26595},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2605.26595}
}