English(EN) Forecasting Trajectory-Level Safety Risks in Black-Box Multi-Turn Interactions

新框架可预测LLM安全风险的发生

作者 PulseAugur 编辑部 · [1 个来源] · 2026-07-30 04:00

研究人员开发了Recast，一个旨在预测大型语言模型（LLM）在多轮交互中安全风险的新框架。与现有对违规行为做出反应的方法不同，Recast通过分析风险在对话轨迹中的演变来预测潜在的安全故障。该系统使用对话历史的双尺度视图和因果时间编码来预测未来风险的出现，在预测安全故障方面取得了88.3%的成功率，平均提前量为2.41轮。 AI

影响该框架可以为LLM代理启用更主动的安全措施，在故障发生前进行预防。

排序理由该集群包含一篇详细介绍LLM安全新框架的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 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: 0 / 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安全新框架的研究论文。[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: 70 days old; Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.LG TIER_1 English(EN) · Shi Lin, Peng Qian, Dinghao Liu, Renjie Sun, Sifan Wu, Dezhang Kong, Chenpei Wang, Xun Wang · 2026-07-30 04:00

预测黑盒多轮交互中的轨迹级安全风险

arXiv:2607.26820v1 Announce Type: new Abstract: As large language models (LLMs) evolve from standalone assistants into autonomous agents, ensuring their safety requires shifting beyond pointwise risk assessment to understand how risks emerge and unfold over long-horizon trajector…

本文如何被排名

报道来源 [1]

预测黑盒多轮交互中的轨迹级安全风险

相关实体

相关话题