English(EN) CoFiBreak: A Coarse-to-Fine Diversity Reward for Effective Multi-Turn Jailbreak

新的CoFiBreak方法提高了LLM红队测试的多样性和有效性

作者 PulseAugur 编辑部 · [1 个来源] · 2026-11-02 05:00

研究人员开发了一种名为CoFiBreak的新方法，以提高大型语言模型自动化红队测试的多样性。该技术采用粗粒度到细粒度的多样性奖励系统，鼓励更广泛的攻击策略和提示变体，同时不牺牲有效性或增加推理成本。与现有基线相比，CoFiBreak已证明能够发现明显更多不同的成功攻击策略，并且这些攻击在对抗未见的尖端LLM（包括Claude Opus-4.6）方面显示出有效性。 AI

影响通过提高自动化红队测试技术的多样性和有效性，增强了LLM安全评估。

排序理由该集群包含一篇详细介绍LLM安全研究新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 1 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇详细介绍LLM安全研究新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, safety; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: 1 days old; Coverage has settled into its steady-state source set.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.CL TIER_1 English(EN) · Jialin Song, Xiaodong Liu, Weiwei Yang, Mingqian Feng, Chengquan Guo, Qiyao Ma, Carolyn Buractaon, Anson Ho, Wuyang Chen, Christopher White · 2026-11-02 05:00

CoFiBreak: A Coarse-to-Fine Diversity Reward for Effective Multi-Turn Jailbreak

arXiv:2610.40926v1 Announce Type: new Abstract: Automated red teaming trains an attacker large language model (LLM) to elicit harmful responses from a victim LLM, but reinforcement learning can collapse attackers onto a narrow set of prompts when attack success is the only reward…

本文如何被排名

报道来源 [1]

CoFiBreak: A Coarse-to-Fine Diversity Reward for Effective Multi-Turn Jailbreak

相关实体

相关话题