扩大成员推理的规模:攻击在大型语言模型上何时以及如何成功
Scaling Up Membership Inference: When and How Attacks Succeed on Large Language Models
AI总结:
研究指出MIA在LLM上仍有效但需多文档测试,构建新基准并适配DI方法聚合段落特征,首次在预训练和微调LLM上实现成功攻击。
摘要:
成员推理攻击(MIA)试图验证给定数据样本是否包含在模型的训练集中。随着大型语言模型(LLM)的快速发展,MIA近年来变得重要。许多人担心训练中使用受版权保护的材料,并呼吁检测此类使用的方法。然而,最近的研究普遍认为当前的MIA方法在LLM上不起作用。即使看似有效,通常也是由于设计不当的实验设置,其中其他捷径特征导致了“作弊”。在这项工作中,我们论证MIA在LLM上仍然有效,但前提是提供多个文档进行测试。我们构建了新的基准,在从句子(n-grams)到文档集合(多个token块)的连续数据样本尺度上衡量MIA性能。为了验证当前MIA方法在更大规模上的有效性,我们将最近关于数据集推理(DI)的工作适配到二元成员检测任务中,通过聚合段落级MIA特征,实现文档和文档集合级别的MIA。该基线首次在预训练和微调的LLM上实现了成功的MIA。
英文摘要:
Membership inference attacks (MIA) attempt to verify the membership of a given data sample in the training set for a model. MIA has become relevant in recent years, following the rapid development of large language models (LLM). Many are concerned about the usage of copyrighted materials for training them and call for methods for detecting such usage. However, recent research has largely concluded that current MIA methods do not work on LLMs. Even when they seem to work, it is usually because of the ill-designed experimental setup where other shortcut features enable "cheating." In this work, we argue that MIA still works on LLMs, but only when multiple documents are presented for testing. We construct new benchmarks that measure the MIA performances at a continuous scale of data samples, from sentences (n-grams) to a collection of documents (multiple chunks of tokens). To validate the efficacy of current MIA approaches at greater scales, we adapt a recent work on Dataset Inference (DI) for the task of binary membership detection that aggregates paragraph-level MIA features to enable MIA at document and collection of documents level. This baseline achieves the first successful MIA on pre-trained and fine-tuned LLMs.
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。