Wav2DF-TSL: Two-stage Learning with Efficient Pre-training and Hierarchical Experts Fusion for Robust Audio Deepfake Detection

@inproceedings{hao2025wavdftsl,
  title = {Wav2DF-TSL: Two-stage Learning with Efficient Pre-training and Hierarchical Experts Fusion for Robust Audio Deepfake Detection},
  author = {Yunqi Hao and Yihao Chen and Minqiang Xu and Jian-Bo Zhan and Liang He and Lei Fang and Sian Fang and Lin Liu},
  year = {2025},
  booktitle = {IEEE International Joint Conference on Neural Network},
  eprint = {2509.04161},
  archivePrefix = {arXiv},
  doi = {10.1109/IJCNN64981.2025.11228261},
  url = {https://arxiv.org/abs/2509.04161},
}