测试大模型遗忘能力时，数据格式变化会暴露记忆漏洞。
Towards Robust Evaluation of Unlearning in LLMs via Data Transformations
- 用数据变换检测大模型是否真正遗忘敏感信息。
- 在TOFU数据集上发现，格式改变后模型仍可复现遗忘内容。
- 适合关注隐私安全与模型可控性的研究者参考。
大型语言模型（LLMs）在众多应用中表现出色，但训练数据中可能包含个人身份信息（PII）等不期望内容。近期机器遗忘（MUL）研究旨在让模型遗忘特定信息（如PII），同时保持常规任务性能。本文评估现有MUL技术的鲁棒性，重点考察输入数据格式变化对遗忘效果的影响：当输入格式改变时，已遗忘的模型是否仍能回忆敏感信息？在TOFU数据集上的实验表明，仅使用单一数据格式会低估遗忘不彻底的风险，因此需采用多样化数据格式才能更可靠地量化遗忘效果。
原文摘要 · Abstract (English)
Large Language Models (LLMs) have shown to be a great success in a wide range of applications ranging from regular NLP-based use cases to AI agents. LLMs have been trained on a vast corpus of texts from various sources; despite the best efforts during the data pre-processing stage while training the LLMs, they may pick some undesirable information such as personally identifiable information (PII). Consequently, in recent times research in the area of Machine Unlearning (MUL) has become active, the main idea is to force LLMs to forget (unlearn) certain information (e.g., PII) without suffering from performance loss on regular tasks. In this work, we examine the robustness of the existing MUL techniques for their ability to enable leakage-proof forgetting in LLMs. In particular, we examine the effect of data transformation on forgetting, i.e., is an unlearned LLM able to recall forgotten information if there is a change in the format of the input? Our findings on the TOFU dataset highlight the necessity of using diverse data formats to quantify unlearning in LLMs more reliably.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。