@article{prefgrpopairwisepreferencerewardbasedgrp,
  title = {Pref-GRPO: Pairwise Preference Reward-based GRPO for Stable Text-to-Image Reinforcement Learning},
  author = {Yibin Wang and Zhimin Li and Yuhang Zang and Yujie Zhou and Jiazi Bu and Chunyu Wang and Qinglin Lu and Cheng Jin and Jiaqi Wang},
  year = {2025},
  eprint = {2508.20751},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2508.20751},
}