RKLD：用于大型语言模型遗忘个人信息的基于反向KL散度的知识蒸馏
RKLD: Reverse KL-Divergence-based Knowledge Distillation for Unlearning Personal Information in Large Language Models
AI总结：
针对大型语言模型遗忘个人信息时难以平衡遗忘质量与模型效用的问题，提出基于反向KL散度的知识蒸馏遗忘算法RKLD，有效实现高质量遗忘并维持模型效用。
摘要：
随着“被遗忘权”（RTBF）法规的通过以及语言模型训练数据集规模的扩大，大型语言模型（LLMs）中的模型遗忘研究变得更加关键。在LLMs时代之前，机器遗忘研究主要集中在具有小参数量模型的分类任务上。在这些任务中，需要遗忘或保留的内容清晰明了。然而，随着参数规模的增长和任务复杂性的增加，平衡遗忘质量和模型效用变得更加具有挑战性，特别是在涉及个人数据而非分类结果的场景中。现有的基于梯度上升及其变体的方法往往难以实现这种平衡，导致意外的信息丢失或部分遗忘。为了应对这一挑战，我们提出了RKLD，一种新颖的基于反向KL散度的知识蒸馏（RKLD）遗忘算法，针对LLMs中个人信息的遗忘。通过RKLD，我们在实验中实现了显著的遗忘质量，并有效维持了模型效用。
英文摘要：
With the passage of the Right to Be Forgotten (RTBF) regulations and the scaling up of language model training datasets, research on model unlearning in large language models (LLMs) has become more crucial. Before the era of LLMs, machine unlearning research focused mainly on classification tasks in models with small parameters. In these tasks, the content to be forgotten or retained is clear and straightforward. However, as parameter sizes have grown and tasks have become more complex, balancing forget quality and model utility has become more challenging, especially in scenarios involving personal data instead of classification results. Existing methods based on gradient ascent and its variants often struggle with this balance, leading to unintended information loss or partial forgetting. To address this challenge, we propose RKLD, a novel \textbf{R}everse \textbf{KL}-Divergence-based Knowledge \textbf{D}istillation unlearning algorithm for LLMs targeting the unlearning of personal information. Through RKLD, we achieve significant forget quality and effectively maintain the model utility in our experiments.
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。
