Course-Correction: Safety Alignment Using Synthetic Preferences EMNLP

Rongwu Xu | Yishuo Cai | Zhenhong Zhou | Renjie Gu | Haiqin Weng | Liu Yan | Tianwei Zhang | Wei Xu | Han Qiu |

How Alignment and Jailbreak Work: Explain LLM Safety through Intermediate Hidden States F i n d i n g s - E M N L P

Zhenhong Zhou | Haiyang Yu | Xinghua Zhang | Rongwu Xu | Fei Huang | Yongbin Li |

Alignment-Enhanced Decoding: Defending Jailbreaks via Token-Level Adaptive Refining of Probability Distributions EMNLP

Quan Liu | Zhenhong Zhou | Longzhu He | Yi Liu | Wei Zhang | Sen Su |