@article{deceptiveautomatedinterpretabilitylangua,
  title = {Deceptive Automated Interpretability: Language Models Coordinating to Fool Oversight Systems},
  author = {Simon Lermen and Mateusz Dziemian and Natalia Pérez-Campanero Antolín},
  year = {2025},
  eprint = {2504.07831},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2504.07831v1},
}