@article{saeexplainerinterpretingsaefeatureswitha,
  title = {SAEExplainer: Interpreting SAE Features with Activation-Guided Preference Optimization},
  author = {Jingyi He and Haiyan Zhao and Ruxue Shi and Yanguang Liu and Xin Wang and Fei Sun and Mengnan Du},
  year = {2026},
  eprint = {2606.08496},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2606.08496},
}