# Stable Multimodal Graph Unlearning via Feature-Dimension Aware Quantile Selection

arXiv: https://arxiv.org/abs/2605.03303 · HTML: https://mlsanity.com/paper/2605.03303/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Graph unlearning, one of 12 directions in LLM safety and security.
- Builds on 29 of its 44 references that are in the map.
- First posted on arXiv 5 May 2026.

## Placed in

- LLM safety and security / [Graph unlearning](https://mlsanity.com/safety/graph-unlearning/)

## Builds on (most cited first, 29 of 29)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [ImageBind: One Embedding Space To Bind Them All](https://mlsanity.com/paper/2305.05665/) (2023)
- [Machine Unlearning](https://mlsanity.com/paper/1912.03817/) (2019)
- [Large Language Models on Graphs: A Comprehensive Survey](https://mlsanity.com/paper/2312.02783/) (2023)
- [Fast Machine Unlearning Without Retraining Through Selective Synaptic Dampening](https://mlsanity.com/paper/2308.07707/) (2023)
- [Graph Unlearning](https://mlsanity.com/paper/2103.14991/) (2021)
- [Membership Inference Attack on Graph Neural Networks](https://mlsanity.com/paper/2101.06570/) (2021)
- [Fine-Grained Spoiler Detection from Large-Scale Review Corpora](https://mlsanity.com/paper/1905.13416/) (2019)
- [UniGraph2: Learning a Unified Embedding Space to Bind Multimodal Graphs](https://mlsanity.com/paper/2502.00806/) (2025)
- [GITA: Graph to Visual and Textual Integration for Vision-Language Graph Reasoning](https://mlsanity.com/paper/2402.02130/) (2024)
- [Graph Transformers: A Survey](https://mlsanity.com/paper/2407.09777/) (2024)
- [Inductive Graph Unlearning](https://mlsanity.com/paper/2304.03093/) (2023)
- [Towards Effective and General Graph Unlearning via Mutual Evolution](https://mlsanity.com/paper/2401.11760/) (2024)
- [Mosaic of Modalities: A Comprehensive Benchmark for Multimodal Graph Learning](https://mlsanity.com/paper/2406.16321/) (2024)
- [Learning on Multimodal Graphs: A Survey](https://mlsanity.com/paper/2402.05322/) (2024)
- [IDEA: A Flexible Framework of Certified Unlearning for Graph Neural Networks](https://mlsanity.com/paper/2407.19398/) (2024)
- [MultiDelete for Multimodal Machine Unlearning](https://mlsanity.com/paper/2311.12047/) (2023)
- [GRAPHGPT-O: Synergistic Multimodal Comprehension and Generation on Graphs](https://mlsanity.com/paper/2502.11925/) (2025)
- [Graph Learning for Anomaly Analytics: Algorithms, Applications, and Challenges](https://mlsanity.com/paper/2212.05532/) (2022)
- [Scalable and Certifiable Graph Unlearning: Overcoming the Approximation Error Barrier](https://mlsanity.com/paper/2408.09212/) (2024)
- [Erase then Rectify: A Training-Free Parameter Editing Approach for Cost-Effective Graph Unlearning](https://mlsanity.com/paper/2409.16684/) (2024)
- [Multi-Modal Recommendation Unlearning for Legal, Licensing, and Modality Constraints](https://mlsanity.com/paper/2405.15328/) (2024)
- [OpenViewer: Openness-Aware Multi-View Learning](https://mlsanity.com/paper/2412.12596/) (2024)
- [Heterogeneous Graph Learning for Explainable Recommendation over Academic Networks](https://mlsanity.com/paper/2202.07832/) (2022)
- [Synthetic Forgetting without Access: A Few-shot Zero-glance Framework for Machine Unlearning](https://mlsanity.com/paper/2511.13116/) (2025)

## Abstract (arXiv)

Graph unlearning remains a critical technique for supporting privacy-preserving and sustainable multimodal graph learning. However, we observe that existing unlearning strategies tend to apply uniform parameter selection and editing across all graph neural network (GNN) layers, which is especially harmful for multimodal graphs where high-dimensional input projections encode dominant cross-modal knowledge. As a result, over-editing these sensitive layers often leads to catastrophic utility degradation after forgetting, undermining both stable learning and effective privacy protection. To address this gap, we propose FDQ, a Feature-Dimension Aware Quantile framework for multimodal graph unlearning. FDQ adaptively identifies high-dimensional input projection layers and applies more conservative, FDQ-guided quantile thresholds when constructing suppression sets, while keeping the underlying importance estimation mechanism unchanged. FDQ is seamlessly integrated with diagonal sensitivity-based parameter importance analysis to enable efficient node and edge unlearning under general forget requests. Through extensive experiments on Ele-Fashion and Goodreads-NC, we demonstrate that FDQ consistently achieves strong utility preservation while maintaining effective forgetting against membership inference attacks. Overall, FDQ offers a principled and robust solution for privacy-aware unlearning in high-dimensional multimodal graph systems.