@article{anchoringrefusaldirectionmitigatingsafet, title = {Anchoring Refusal Direction: Mitigating Safety Risks in Tuning via Projection Constraint}, author = {Yanrui Du and Fenglei Fan and Sendong Zhao and Jiawei Cao and Qika Lin and Kai He and Ting Liu and Bing Qin and Mengling Feng}, year = {2025}, eprint = {2509.06795}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2509.06795}, }