一作ACL 2026Poster· Natural Language Processing
OASIS : Mitigating Harmful Fine-tuning Attacks on LLM s via Orthogonal and Adaptive Safety Alignment Strategy
Jiayu Tang · Guowei Peng · Qiuhao Xie · Yuning Yang · Xiurui Xie · Guisong Liu
摘要
The “Fine-Tuning-as-a-Service” paradigm exposes large language models to catastrophic safety degradation from less harmful samples. Alignment-stage defenses address this by proactively injecting adversarial perturbations to bolster the model’s inherent robustness against harmful drift. However, e...