@article{honestytosubterfugeincontext,
  title = {Honesty to Subterfuge: In-Context Reinforcement Learning Can Make Honest Models Reward Hack},
  author = {Leo McKee-Reid and Christoph Sträter and Maria Angelica Martinez and Joe Needham and Mikita Balesni},
  year = {2024},
  eprint = {2410.06491},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2410.06491v1},
}