English(EN) Breaking Adversarial Transferability in Fine-Tuned Speech Recognition

新框架TransferBreaker应对语音识别模型的对抗性攻击

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-08 04:00

研究人员开发了一个名为TransferBreaker的新框架，以增强微调自动语音识别（ASR）模型的安全性。这些模型通常在黑盒环境中部署，容易受到对抗性攻击，即为基础模型设计的扰动会显著降低微调版本的性能。TransferBreaker集成了多种技术，包括基础对抗性微调和潜在雅可比正则化，以抑制这种对抗性可迁移性。在多种语言和ASR模型上的评估表明，在对抗性条件下，词错误率得到了显著降低。 AI

影响增强了已部署语音识别系统在对抗性操纵下的安全性和可靠性。

排序理由该集群包含一篇详细介绍改进模型鲁棒性新方法的论文。[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: 8 / 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改进模型鲁棒性新方法的论文。[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, model release; 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: Same-day; Cluster formed today. Ranking reflects the current source set at time of score.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.LG TIER_1 English(EN) · Mojtaba Nafez, Aref Mousavi, Mohammad Ebrahim Mahdavi, Mobina Poulaei, Kiarash Kiani Feriz, Mohammad Hossein Rohban · 2026-10-08 04:00

微调语音识别中对抗可迁移性的突破

arXiv:2610.09109v1 Announce Type: new Abstract: Many organizations fine-tune publicly available pretrained Automatic Speech Recognition (ASR) models and deploy them in black-box settings, assuming limited access provides protection. We show this assumption is fragile: adversarial…

本文如何被排名

报道来源 [1]

微调语音识别中对抗可迁移性的突破

相关实体

相关话题