@article{fromdemonstrationstorewardsalignment,
  title = {From Demonstrations to Rewards: Alignment Without Explicit Human Preferences},
  author = {Siliang Zeng and Yao Liu and Huzefa Rangwala and George Karypis and Mingyi Hong and Rasool Fakoor},
  year = {2025},
  eprint = {2503.13538},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2503.13538v1},
}