Re-understanding Graph Unlearning through Memorization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ding, Pengfei, Wang, Yan, Liu, Guanfeng
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908779373658112
author Ding, Pengfei
Wang, Yan
Liu, Guanfeng
author_facet Ding, Pengfei
Wang, Yan
Liu, Guanfeng
contents Graph unlearning (GU), which removes nodes, edges, or features from trained graph neural networks (GNNs), is crucial in Web applications where graph data may contain sensitive, mislabeled, or malicious information. However, existing GU methods lack a clear understanding of the key factors that determine unlearning effectiveness, leading to three fundamental limitations: (1) impractical and inaccurate GU difficulty assessment due to test-access requirements and invalid assumptions, (2) ineffectiveness on hard-to-unlearn tasks, and (3) misaligned evaluation protocols that overemphasize easy tasks and fail to capture true forgetting capability. To address these issues, we establish GNN memorization as a new perspective for understanding graph unlearning and propose MGU, a Memorization-guided Graph Unlearning framework. MGU achieves three key advances: it provides accurate and practical difficulty assessment across different GU tasks, develops an adaptive strategy that dynamically adjusts unlearning objectives based on difficulty levels, and establishes a comprehensive evaluation protocol that aligns with practical requirements. Extensive experiments on ten real-world graphs demonstrate that MGU consistently outperforms state-of-the-art baselines in forgetting quality, computational efficiency, and utility preservation.
format Preprint
id arxiv_https___arxiv_org_abs_2601_14694
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Re-understanding Graph Unlearning through Memorization
Ding, Pengfei
Wang, Yan
Liu, Guanfeng
Machine Learning
Artificial Intelligence
Graph unlearning (GU), which removes nodes, edges, or features from trained graph neural networks (GNNs), is crucial in Web applications where graph data may contain sensitive, mislabeled, or malicious information. However, existing GU methods lack a clear understanding of the key factors that determine unlearning effectiveness, leading to three fundamental limitations: (1) impractical and inaccurate GU difficulty assessment due to test-access requirements and invalid assumptions, (2) ineffectiveness on hard-to-unlearn tasks, and (3) misaligned evaluation protocols that overemphasize easy tasks and fail to capture true forgetting capability. To address these issues, we establish GNN memorization as a new perspective for understanding graph unlearning and propose MGU, a Memorization-guided Graph Unlearning framework. MGU achieves three key advances: it provides accurate and practical difficulty assessment across different GU tasks, develops an adaptive strategy that dynamically adjusts unlearning objectives based on difficulty levels, and establishes a comprehensive evaluation protocol that aligns with practical requirements. Extensive experiments on ten real-world graphs demonstrate that MGU consistently outperforms state-of-the-art baselines in forgetting quality, computational efficiency, and utility preservation.
title Re-understanding Graph Unlearning through Memorization
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2601.14694