English(EN) `From Prompt to Perturbation': An Adaptive Framework for Voice-Based Jailbreaks on Audio LLMs

新框架揭示音频大语言模型漏洞

作者 PulseAugur 编辑部 · [1 个来源] · 2026-08-20 04:00

研究人员开发了一个新的自适应框架，用于测试基于语音的大语言模型（LLMs）的安全性。该框架能够生成和优化文本提示和音频扰动，以利用级联流水线和端到端大型音频语言模型（LALMs）中的漏洞。实验表明，当前的音频大语言模型系统仍然容易受到这些越狱攻击，并且所提出的框架比现有方法取得了更高的成功率。 AI

影响凸显了基于语音的AI系统潜在的安全风险，需要进一步研究鲁棒的防御机制。

排序理由详细介绍评估音频大语言模型安全性的新框架的学术论文。[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: 0 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 详细介绍评估音频大语言模型安全性的新框架的学术论文。[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: 49 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.AI TIER_1 English(EN) · Linghan Huang, Bo Li, Huaming Chen, Kim-Kwang Raymond Choo · 2026-08-20 04:00

“从提示到扰动”：一种用于音频大语言模型语音越狱的自适应框架

arXiv:2502.00735v4 Announce Type: replace-cross Abstract: As large language models (LLMs) are increasingly integrated into audio-based applications, growing concerns have emerged regarding their vulnerability to audio-based adversarial attacks. These systems typically follow two …

本文如何被排名

报道来源 [1]

“从提示到扰动”：一种用于音频大语言模型语音越狱的自适应框架

相关实体

相关话题