@article{preferenceoptimizationviacontrastive,
  title = {Preference Optimization via Contrastive Divergence: Your Reward Model is Secretly an NLL Estimator},
  author = {Zhuotong Chen and Fang Liu and Xuan Zhu and Yanjun Qi and Mohammad Ghavamzadeh},
  year = {2025},
  eprint = {2502.04567},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2502.04567v1},
}