@misc{indiciaeb30fa9ed70da,
  title = {How to Enhance Downstream Adversarial Robustness (almost) without Touching the Pre-Trained Foundation Model?},
  author = {Meiqi Liu and Zhuoqun Huang and Yue Xing},
  year = {2025},
  url = {https://arxiv.org/abs/2504.10850},
  note = {Source identifier: 2504.10850}
}