RobustKV: Defending Large Language Models against Jailbreak Attacks via KV Eviction
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917818345193472 |
|---|---|
| author | Jiang, Tanqiu Wang, Zian Liang, Jiacheng Li, Changjiang Wang, Yuhui Wang, Ting |
| author_facet | Jiang, Tanqiu Wang, Zian Liang, Jiacheng Li, Changjiang Wang, Yuhui Wang, Ting |
| contents | Jailbreak attacks circumvent LLMs' built-in safeguards by concealing harmful queries within jailbreak prompts. While existing defenses primarily focus on mitigating the effects of jailbreak prompts, they often prove inadequate as jailbreak prompts can take arbitrary, adaptive forms. This paper presents RobustKV, a novel defense that adopts a fundamentally different approach by selectively removing critical tokens of harmful queries from key-value (KV) caches. Intuitively, for a jailbreak prompt to be effective, its tokens must achieve sufficient `importance' (as measured by attention scores), which inevitably lowers the importance of tokens in the concealed harmful query. Thus, by strategically evicting the KVs of the lowest-ranked tokens, RobustKV diminishes the presence of the harmful query in the KV cache, thus preventing the LLM from generating malicious responses. Extensive evaluation using benchmark datasets and models demonstrates that RobustKV effectively counters state-of-the-art jailbreak attacks while maintaining the LLM's general performance on benign queries. Moreover, RobustKV creates an intriguing evasiveness dilemma for adversaries, forcing them to balance between evading RobustKV and bypassing the LLM's built-in safeguards. This trade-off contributes to RobustKV's robustness against adaptive attacks. (warning: this paper contains potentially harmful content generated by LLMs.) |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_19937 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | RobustKV: Defending Large Language Models against Jailbreak Attacks via KV Eviction Jiang, Tanqiu Wang, Zian Liang, Jiacheng Li, Changjiang Wang, Yuhui Wang, Ting Cryptography and Security Artificial Intelligence Computation and Language Jailbreak attacks circumvent LLMs' built-in safeguards by concealing harmful queries within jailbreak prompts. While existing defenses primarily focus on mitigating the effects of jailbreak prompts, they often prove inadequate as jailbreak prompts can take arbitrary, adaptive forms. This paper presents RobustKV, a novel defense that adopts a fundamentally different approach by selectively removing critical tokens of harmful queries from key-value (KV) caches. Intuitively, for a jailbreak prompt to be effective, its tokens must achieve sufficient `importance' (as measured by attention scores), which inevitably lowers the importance of tokens in the concealed harmful query. Thus, by strategically evicting the KVs of the lowest-ranked tokens, RobustKV diminishes the presence of the harmful query in the KV cache, thus preventing the LLM from generating malicious responses. Extensive evaluation using benchmark datasets and models demonstrates that RobustKV effectively counters state-of-the-art jailbreak attacks while maintaining the LLM's general performance on benign queries. Moreover, RobustKV creates an intriguing evasiveness dilemma for adversaries, forcing them to balance between evading RobustKV and bypassing the LLM's built-in safeguards. This trade-off contributes to RobustKV's robustness against adaptive attacks. (warning: this paper contains potentially harmful content generated by LLMs.) |
| title | RobustKV: Defending Large Language Models against Jailbreak Attacks via KV Eviction |
| topic | Cryptography and Security Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2410.19937 |