Yichen Gong (Tsinghua University), Delong Ran (Tsinghua University), Xinlei He (Hong Kong University of Science and Technology (Guangzhou)), Tianshuo Cong (Tsinghua University), Anyu Wang (Tsinghua University), Xiaoyun Wang (Tsinghua University)

The safety alignment of Large Language Models (LLMs) is crucial to prevent unsafe content that violates human values.
To ensure this, it is essential to evaluate the robustness of their alignment against diverse malicious attacks.
However, the lack of a large-scale, unified measurement framework hinders a comprehensive understanding of potential vulnerabilities.
To fill this gap, this paper presents the first comprehensive evaluation of existing and newly proposed safety misalignment methods for LLMs. Specifically, we investigate four research questions: (1) evaluating the robustness of LLMs with different alignment strategies, (2) identifying the most effective misalignment method, (3) determining key factors that influence misalignment effectiveness, and (4) exploring various defenses.
The safety misalignment attacks in our paper include system-prompt modification, model fine-tuning, and model editing.
Our findings show that Supervised Fine-Tuning is the most potent attack but requires harmful model responses.
In contrast, our novel Self-Supervised Representation Attack (SSRA) achieves significant misalignment without harmful responses.
We also examine defensive mechanisms such as safety data filter, model detoxification, and our proposed Self-Supervised Representation Defense (SSRD), demonstrating that SSRD can effectively re-align the model.
In conclusion, our unified safety alignment evaluation framework empirically highlights the fragility of the safety alignment of LLMs.

View More Papers

Try to Poison My Deep Learning Data? Nowhere to...

Yansong Gao (The University of Western Australia), Huaibing Peng (Nanjing University of Science and Technology), Hua Ma (CSIRO's Data61), Zhi Zhang (The University of Western Australia), Shuo Wang (Shanghai Jiao Tong University), Rayne Holland (CSIRO's Data61), Anmin Fu (Nanjing University of Science and Technology), Minhui Xue (CSIRO's Data61), Derek Abbott (The University of Adelaide, Australia)

Read More

I Know What You Asked: Prompt Leakage via KV-Cache...

Guanlong Wu (Southern University of Science and Technology), Zheng Zhang (ByteDance Inc.), Yao Zhang (ByteDance Inc.), Weili Wang (Southern University of Science and Technolog), Jianyu Niu (Southern University of Science and Technolog), Ye Wu (ByteDance Inc.), Yinqian Zhang (Southern University of Science and Technology (SUSTech))

Read More

Mixnets on a Tightrope: Quantifying the Leakage of Mix...

Sebastian Meiser, Debajyoti Das, Moritz Kirschte, Esfandiar Mohammadi, Aniket Kate

Read More

Space Cybersecurity Testbed: Fidelity Framework, Example Implementation, and Characterization

Jose Luis Castanon Remy, Caleb Chang, Ekzhin Ear, Shouhuai Xu (University of Colorado Colorado Springs (UCCS))

Read More