English(EN) Same Request, Different Boundary: Evaluating Cybersecurity Assistance across Conversational Contexts

新的3R-Bench基准评估LLM在对话中的网络安全辅助能力

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-02 04:00

研究人员开发了一个名为3R-Bench的新基准，用于评估大型语言模型（LLM）在对话情境下区分合法网络安全辅助请求和潜在有害请求的能力。该基准包含150个真实世界的网络安全请求和两个对抗性对话场景。对八个LLM的初步评估显示，模型对网络安全请求的遵从度显著受到先前对话历史的影响，在拒绝历史后的遵从度为62.0%，而在接受历史后则上升到85.1%。 AI

影响该基准可能促使为网络安全应用开发更强大的LLM安全机制。

排序理由该集群包含一篇详细介绍评估LLM能力新基准的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 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: 31 / 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评估LLM能力新基准的研究论文。[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: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Rui Yang, Yang Hong, Yichao Xu, Zhengyu Liu, Ziyang Li, Yinzhi Cao · 2026-09-02 04:00

相同请求，不同边界：跨对话上下文评估网络安全辅助

arXiv:2609.00578v1 Announce Type: new Abstract: Large Language Models (LLMs) can solve complex problems, but their misuse in high-risk domains can lead to severe consequences. Model providers therefore restrict assistance for potentially harmful requests. Refusing all cybersecuri…

本文如何被排名

报道来源 [1]

相同请求，不同边界：跨对话上下文评估网络安全辅助

相关实体

相关话题