A Survey on Generative Model Unlearning: Fundamentals, Taxonomy, Evaluation, and Future Direction

TL;DR

This survey reviews the current state of generative model unlearning, proposing a unified framework to organize existing research, evaluate methods, and identify future challenges in addressing privacy concerns.

Contribution

It provides a comprehensive overview and a unified analytical framework for categorizing unlearning objectives, strategies, and evaluation metrics in generative model unlearning research.

Findings

Identifies key challenges in unlearning objectives and evaluation protocols.

Connects GenMU with related techniques like model editing and reinforcement learning.

Highlights practical applications and future research directions.

Abstract

With the rapid advancement of generative models, associated privacy concerns have attracted growing attention. To address this, researchers have begun adapting machine unlearning techniques from traditional classification models to generative settings. Although notable progress has been made in this area, a unified framework for systematically organizing and integrating existing work is still lacking. The substantial differences among current studies in terms of unlearning objectives and evaluation protocols hinder the objective and fair comparison of various approaches. While some studies focus on specific types of generative models, they often overlook the commonalities and systematic characteristics inherent in Generative Model Unlearning (GenMU). To bridge this gap, we provide a comprehensive review of current research on GenMU and propose a unified analytical framework for categorizing unlearning objectives, methodological strategies, and evaluation metrics. In addition, we explore the connections between GenMU and related techniques, including model editing, reinforcement learning from human feedback, and controllable generation. We further highlight the potential practical value of unlearning techniques in real-world applications. Finally, we identify key challenges and outline future research directions aimed at laying a solid foundation for further advancements in this field. We consistently maintain the related open-source materials at https://github.com/caxLee/Generative-model-unlearning-survey.