@article{detoxifyingllmsviarepresentationerasureb,
  title = {Detoxifying LLMs via Representation Erasure-Based Preference Optimization},
  author = {Nazanin Mohammadi Sepahvand and Eleni Triantafillou and Hugo Larochelle and Doina Precup and Daniel M. Roy and Gintare Karolina Dziugaite},
  year = {2026},
  eprint = {2602.23391},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2602.23391},
}