Beyond Surface-Level Patterns: An Essence-Driven Defense Framework Against Jailbreak Attacks in LLMs
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913862952943616 |
|---|---|
| author | Xiang, Shiyu Zhang, Ansen Cao, Yanfei Fan, Yang Chen, Ronghao |
| author_facet | Xiang, Shiyu Zhang, Ansen Cao, Yanfei Fan, Yang Chen, Ronghao |
| contents | Although Aligned Large Language Models (LLMs) are trained to refuse harmful requests, they remain vulnerable to jailbreak attacks. Unfortunately, existing methods often focus on surface-level patterns, overlooking the deeper attack essences. As a result, defenses fail when attack prompts change, even though the underlying "attack essence" remains the same. To address this issue, we introduce EDDF, an \textbf{E}ssence-\textbf{D}riven \textbf{D}efense \textbf{F}ramework Against Jailbreak Attacks in LLMs. EDDF is a plug-and-play input-filtering method and operates in two stages: 1) offline essence database construction, and 2) online adversarial query detection. The key idea behind EDDF is to extract the "attack essence" from a diverse set of known attack instances and store it in an offline vector database. Experimental results demonstrate that EDDF significantly outperforms existing methods by reducing the Attack Success Rate by at least 20\%, underscoring its superior robustness against jailbreak attacks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_19041 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Beyond Surface-Level Patterns: An Essence-Driven Defense Framework Against Jailbreak Attacks in LLMs Xiang, Shiyu Zhang, Ansen Cao, Yanfei Fan, Yang Chen, Ronghao Cryptography and Security Although Aligned Large Language Models (LLMs) are trained to refuse harmful requests, they remain vulnerable to jailbreak attacks. Unfortunately, existing methods often focus on surface-level patterns, overlooking the deeper attack essences. As a result, defenses fail when attack prompts change, even though the underlying "attack essence" remains the same. To address this issue, we introduce EDDF, an \textbf{E}ssence-\textbf{D}riven \textbf{D}efense \textbf{F}ramework Against Jailbreak Attacks in LLMs. EDDF is a plug-and-play input-filtering method and operates in two stages: 1) offline essence database construction, and 2) online adversarial query detection. The key idea behind EDDF is to extract the "attack essence" from a diverse set of known attack instances and store it in an offline vector database. Experimental results demonstrate that EDDF significantly outperforms existing methods by reducing the Attack Success Rate by at least 20\%, underscoring its superior robustness against jailbreak attacks. |
| title | Beyond Surface-Level Patterns: An Essence-Driven Defense Framework Against Jailbreak Attacks in LLMs |
| topic | Cryptography and Security |
| url | https://arxiv.org/abs/2502.19041 |