arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2605.28211cs.CL
收藏

当有用上下文泄露:领域自适应ASR中的隐私风险

When Helpful Context Leaks: Privacy Risks in Domain-Adapted ASR

Karlsruhe Institute of Technology(卡尔斯鲁厄理工学院)

Maike Züfle, Jan Niehues

AI总结:

本文识别并系统研究了领域自适应ASR中因上下文提示或微调导致模型泄露隐私的风险,通过构建控制数据集测量泄露率,并评估了提示级缓解策略及精度-泄露权衡。

摘要:

语音大语言模型越来越多地部署在专业环境中,领域定制是标准做法:用户在提示中提供包含敏感信息的上下文,在专有录音上进行微调,或两者兼有。我们识别并系统研究了这种定制的一个被忽视的隐私风险:适应于识别领域特定术语的模型可以被诱导转录其上下文或训练数据中一个语音相似的词,即使说的是不同的词,从而泄露私人信息。为了评估这一风险,我们构建了一个控制数据集,并测量了两种定制机制(提示和微调)下的泄露率。两种机制都会导致可测量的泄露,且组合时加剧。我们评估了一种提示级缓解策略,并分析了不同定制方法下的精度-泄露权衡,发现无上下文提示的微调提供了最佳平衡。我们公开了代码和数据集。

英文摘要:

SpeechLLMs are increasingly deployed in professional settings where domain customisation is standard practice: users supply context in prompts with sensitive information, fine-tune on proprietary recordings, or both. We identify and systematically investigate an overlooked privacy risk of such customisation: a model adapted to recognise domain-specific terminology can be nudged into transcribing a phonetically similar word from its context or training data, even when a different word is spoken, thereby leaking private information. To evaluate this risk, we propose a technique to automatically construct benchmarks of such attacks and apply it to measure leakage rates across two customisation mechanisms, prompting and fine-tuning. Both mechanisms cause measurable leakage, compounding when combined. We evaluate a prompt-level mitigation strategy and analyse the accuracy-leakage trade-off across customisation approaches, finding that fine-tuning without context prompts offers the best balance.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑