当心你的相册！多模态大语言模型中不经意的隐私记忆问题研究
Watch Out Your Album! On the Inadvertent Privacy Memorization in Multi-Modal Large Language Models
AI总结：
该研究针对多模态大语言模型是否会不经意记忆训练中任务无关隐私内容的问题，通过在微调图像中加入随机水印并搭建逐层探测框架，证实模型会因部分小批量训练产生虚假关联并记忆无关内容，为MLLMs隐私风险防范提供了新依据。
摘要：
多模态大语言模型（MLLMs）已在视觉问答（VQA）等各类视觉-语言任务上展现出卓越性能。尽管已有越来越多证据表明与任务相关的内容存在隐私风险，但MLLMs是否会不经意间记忆与训练任务完全无关的隐私内容，目前仍不明确。本文研究了随机生成的任务无关隐私内容如何因部分小批量训练动态与下游目标产生虚假关联，进而引发不经意记忆。具体而言，我们以不同概率在VQA微调图像中随机生成任务无关水印，并提出一种新型探测框架，用于判定MLLMs是否已不经意地编码了此类内容。实验表明，在嵌入任务无关水印的部分小批量设置下，MLLMs表现出显著不同的训练行为。此外，通过逐层探测，我们证实MLLMs在遇到此前见过的任务无关知识时会触发独特的表征模式，即便这类知识在提示推理过程中不会影响模型输出。我们的代码可在https://github.com/illusionhi/ProbingPrivacy获取。
英文摘要：
Multi-Modal Large Language Models (MLLMs) have exhibited remarkable performance on various vision-language tasks such as Visual Question Answering (VQA). Despite accumulating evidence of privacy concerns associated with task-relevant content, it remains unclear whether MLLMs inadvertently memorize private content that is entirely irrelevant to the training tasks. In this paper, we investigate how randomly generated task-irrelevant private content can become spuriously correlated with downstream objectives due to partial mini-batch training dynamics, thus causing inadvertent memorization. Concretely, we randomly generate task-irrelevant watermarks into VQA fine-tuning images at varying probabilities and propose a novel probing framework to determine whether MLLMs have inadvertently encoded such content. Our experiments reveal that MLLMs exhibit notably different training behaviors in partial mini-batch settings with task-irrelevant watermarks embedded. Furthermore, through layer-wise probing, we demonstrate that MLLMs trigger distinct representational patterns when encountering previously seen task-irrelevant knowledge, even if this knowledge does not influence their output during prompting. Our code is available at https://github.com/illusionhi/ProbingPrivacy.
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。
