arXiv cs.CL· Zhouyuan Ma, Yutao Wu, Hanxun Huang, Xiang Zheng, Yi Liu, Xingjun Ma·· 12 小时前AI 评分49
RedDarwin：以可执行攻击技能进化越狱能力，对抗前沿 LLM
RedDarwin: Evolving Jailbreak Skills against Frontier LLMs
AI 导读
RedDarwin 是一个将越狱进化单元从提示词改为可执行攻击技能的智能体框架，含 47 个可复用基础技能，进化出 65 个技能并改进其中 44 个基础攻击，在 AdvBench 上把平均攻击成功率从 18.8% 提升至 47.3%。
正文 · 原文
Abstract:Red teaming must continually adapt as frontier LLM safeguards evolve, yet existing approaches largely search for adversarial prompts or optimize within fixed attack procedures. We observe that even when an attack becomes ineffective, its underlying mechanisms can remain useful and can be revised or recombined to construct stronger attacks. Based on this insight, we introduce RedDarwin, an agentic framework that shifts the unit of jailbreak evolution from prompts and fixed attacks to executable attack skills. RedDarwin consists of 47 reusable base skills and autonomously revises, composes, executes, and selects them using target-model feedback, allowing successful discoveries to become building blocks for subsequent evolution. RedDarwin produces 65 evolved skills, improving 44 of 47 base attacks and increasing their mean attack success rate (ASR) on AdvBench from 18.8% to 47.3%. These evolved skills consistently outperform baselines on the held-out StrongREJECT and frontier LLMs unseen during evolution; with only 10 target-model queries, RedDarwin achieves 100% ASR on GPT-5.6-Sol, 92% on Claude-Sonnet-5, and 80% on Gemini-3.8-Flash. Analysis of the evolution trajectories further shows that strong attacks often emerge from novel combinations of reusable mechanisms rather than from individually dominant ones. We believe this paradigm can accelerate adaptive red teaming and inform stronger safeguards, creating complementary value for both safety evaluation and defense.
| Subjects: | Cryptography and Security (cs.CR); Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.23726 [cs.CR] |
| (or arXiv:2610.23726v1 [cs.CR] for this version) | |
| https://doi.org/10.48550/arXiv.2610.23726 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Zhouyuan Ma [view email]
[v1]
Wed, 14 Oct 2026 08:26:44 UTC (940 KB)
来源：arXiv cs.CL · arxiv.org