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

BinEnhance: An Enhancement Framework Based on External Environment Semantics...

Yongpan Wang (Institute of Information Engineering Chinese Academy of Sciences & University of Chinese Academy of Sciences, China), Hong Li (Institute of Information Engineering Chinese Academy of Sciences & University of Chinese Academy of Sciences, China), Xiaojie Zhu (King Abdullah University of Science and Technology, Thuwal, Saudi Arabia), Siyuan Li (Institute of Information Engineering Chinese…

Read More

Recurrent Private Set Intersection for Unbalanced Databases with Cuckoo...

Eduardo Chielle (New York University Abu Dhabi), Michail Maniatakos (New York University Abu Dhabi)

Read More

All your (data)base are belong to us: Characterizing Database...

Kevin van Liebergen (IMDEA Software Institute), Gibran Gomez (IMDEA Software Institute), Srdjan Matic (IMDEA Software Institute), Juan Caballero (IMDEA Software Institute)

Read More

EvoCrawl: Exploring Web Application Code and State using Evolutionary...

Xiangyu Guo (University of Toronto), Akshay Kawlay (University of Toronto), Eric Liu (University of Toronto), David Lie (University of Toronto)

Read More