Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2412.12192 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916526683062272 |
|---|---|
| author | Xue, Zhiyu Liu, Guangliang Chen, Bocheng Johnson, Kristen Marie Pedarsani, Ramtin |
| author_facet | Xue, Zhiyu Liu, Guangliang Chen, Bocheng Johnson, Kristen Marie Pedarsani, Ramtin |
| contents | The security of Large Language Models (LLMs) has become an important research topic since the emergence of ChatGPT. Though there have been various effective methods to defend against jailbreak attacks, prefilling attacks remain an unsolved and popular threat against open-sourced LLMs. In-Context Learning (ICL) offers a computationally efficient defense against various jailbreak attacks, yet no effective ICL methods have been developed to counter prefilling attacks. In this paper, we: (1) show that ICL can effectively defend against prefilling jailbreak attacks by employing adversative sentence structures within demonstrations; (2) characterize the effectiveness of this defense through the lens of model size, number of demonstrations, over-defense, integration with other jailbreak attacks, and the presence of safety alignment. Given the experimental results and our analysis, we conclude that there is no free lunch for defending against prefilling jailbreak attacks with ICL. On the one hand, current safety alignment methods fail to mitigate prefilling jailbreak attacks, but adversative structures within ICL demonstrations provide robust defense across various model sizes and complex jailbreak attacks. On the other hand, LLMs exhibit similar over-defensiveness when utilizing ICL demonstrations with adversative structures, and this behavior appears to be independent of model size. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_12192 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | No Free Lunch for Defending Against Prefilling Attack by In-Context Learning Xue, Zhiyu Liu, Guangliang Chen, Bocheng Johnson, Kristen Marie Pedarsani, Ramtin Cryptography and Security Artificial Intelligence The security of Large Language Models (LLMs) has become an important research topic since the emergence of ChatGPT. Though there have been various effective methods to defend against jailbreak attacks, prefilling attacks remain an unsolved and popular threat against open-sourced LLMs. In-Context Learning (ICL) offers a computationally efficient defense against various jailbreak attacks, yet no effective ICL methods have been developed to counter prefilling attacks. In this paper, we: (1) show that ICL can effectively defend against prefilling jailbreak attacks by employing adversative sentence structures within demonstrations; (2) characterize the effectiveness of this defense through the lens of model size, number of demonstrations, over-defense, integration with other jailbreak attacks, and the presence of safety alignment. Given the experimental results and our analysis, we conclude that there is no free lunch for defending against prefilling jailbreak attacks with ICL. On the one hand, current safety alignment methods fail to mitigate prefilling jailbreak attacks, but adversative structures within ICL demonstrations provide robust defense across various model sizes and complex jailbreak attacks. On the other hand, LLMs exhibit similar over-defensiveness when utilizing ICL demonstrations with adversative structures, and this behavior appears to be independent of model size. |
| title | No Free Lunch for Defending Against Prefilling Attack by In-Context Learning |
| topic | Cryptography and Security Artificial Intelligence |
| url | https://arxiv.org/abs/2412.12192 |