Evolving Skill-Structured Attack Memory Enhances LLM Jailbreaking

Computer Science > Cryptography and Security [Submitted on 28 May 2026] Evolving Skill-Structured Attack Memory Enhances LLM Jailbreaking Junke Zhang, Jianwei Wang, Sishuo Chen, Yizhang He, Qingshuai Feng, Zhengyi Yang Jailbreak attacks on large language models (LLMs) aim to induce LLMs to produce content that they are expected to refuse. Automated black-box jailbreak generation is especially important for safety evaluation, where the attacker observes only model outputs and needs to automatically search for effective adversarial prompts. Existing black-box jailbreak methods either depend on sample-wise heuristic search or leverage attack experience through accumulating strategy pools or method libraries, lacking a systematic organization and management of attack experience. To mitigate these drawbacks, we propose MemoAttack, a memory-driven black-box jailbreak framework with comprehensive attack memory modeling, evolution, and selection. Specifically, MemoAttack comprises three key designs: (1) Skill-Structured Memory Modeling, which abstracts accumulated attack experience into reusable skill-structured attack memory whose units pair attack skills with templates, evidence, and lifecycle state; (2) Lifecycle-Driven Memory Evolution, which evolves the memory through evidence-based probation, promotion, retirement, reactivation, elimination, and storage cleanup; and (3) Explore-Exploit Balanced Memory Selection, which balances reliable memory reuse with uncertainty-driven exploration via contextual Thompson Sampling. Experiments on AdvBench demonstrate that MemoAttack achieves an average attack success rate of 98.00%, outperforming the strongest baseline by 16.67 percentage points, while reducing request count by 45.9%. Moreover, MemoAttack continuously improves as memory accumulates over more samples. Comments: Under review Subjects: Cryptography and Security (cs.CR) Cite as: arXiv:2605.29237 [cs.CR]   (or arXiv:2605.29237v1 [cs.CR] for this version)   https://doi.org/10.48550/arXiv.2605.29237 Focus to learn more Submission history From: Jianwei Wang [view email] [v1] Thu, 28 May 2026 01:53:14 UTC (709 KB) Access Paper: HTML (experimental) view license Current browse context: cs.CR < prev   |   next > new | recent | 2026-05 Change to browse by: cs References & Citations NASA ADS Google Scholar Semantic Scholar Export BibTeX Citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Demos Related Papers About arXivLabs Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)