# Robust MLLM Unlearning via Visual Knowledge Distillation

arXiv: https://arxiv.org/abs/2512.11325 · HTML: https://mlsanity.com/paper/2512.11325/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Machine unlearning, one of 12 directions in LLM safety and security.
- Builds on 27 of its 33 references that are in the map.
- First posted on arXiv 12 Dec 2025.

## Placed in

- LLM safety and security / [Machine unlearning](https://mlsanity.com/safety/machine-unlearning/)
- Knowledge distillation / [Multimodal distillation](https://mlsanity.com/distill/multimodal-distillation/)

## Builds on (most cited first, 27 of 27)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models](https://mlsanity.com/paper/2304.10592/) (2023)
- [InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks](https://mlsanity.com/paper/2312.14238/) (2023)
- [Transformer Feed-Forward Layers Are Key-Value Memories](https://mlsanity.com/paper/2012.14913/) (2020)
- [TOFU: A Task of Fictitious Unlearning for LLMs](https://mlsanity.com/paper/2401.06121/) (2024)
- [Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning](https://mlsanity.com/paper/2404.05868/) (2024)
- [Making AI Forget You: Data Deletion in Machine Learning](https://mlsanity.com/paper/1907.05012/) (2019)
- [Remember What You Want to Forget: Algorithms for Machine Unlearning](https://mlsanity.com/paper/2103.03279/) (2021)
- [Unrolling SGD: Understanding Factors Influencing Machine Unlearning](https://mlsanity.com/paper/2109.13398/) (2021)
- [Adaptive Machine Unlearning](https://mlsanity.com/paper/2106.04378/) (2021)
- [Eight Methods to Evaluate Robust Unlearning in LLMs](https://mlsanity.com/paper/2402.16835/) (2024)
- [Machine Unlearning for Random Forests](https://mlsanity.com/paper/2009.05567/) (2020)
- [Machine Unlearning via Algorithmic Stability](https://mlsanity.com/paper/2102.13179/) (2021)
- [Continual Learning and Private Unlearning](https://mlsanity.com/paper/2203.12817/) (2022)
- [MLLM-Protector: Ensuring MLLM's Safety without Hurting Performance](https://mlsanity.com/paper/2401.02906/) (2024)
- [Unlearning or Obfuscating? Jogging the Memory of Unlearned LLMs via Benign Relearning](https://mlsanity.com/paper/2406.13356/) (2024)
- [Formalizing Data Deletion in the Context of the Right to be Forgotten](https://mlsanity.com/paper/2002.10635/) (2020)
- [Single Image Unlearning: Efficient Machine Unlearning in Multimodal Large Language Models](https://mlsanity.com/paper/2405.12523/) (2024)
- [Unified Gradient-Based Machine Unlearning with Remain Geometry Enhancement](https://mlsanity.com/paper/2409.19732/) (2024)
- [Dissecting Language Models: Machine Unlearning via Selective Pruning](https://mlsanity.com/paper/2403.01267/) (2024)
- [MMUnlearner: Reformulating Multimodal Machine Unlearning in the Era of Multimodal Large Language Models](https://mlsanity.com/paper/2502.11051/) (2025)
- [Modality-Aware Neuron Pruning for Unlearning in Multimodal Large Language Models](https://mlsanity.com/paper/2502.15910/) (2025)
- [Digger: Detecting Copyright Content Mis-usage in Large Language Model Training](https://mlsanity.com/paper/2401.00676/) (2024)
- [Commonsense Knowledge Editing Based on Free-Text in LLMs](https://mlsanity.com/paper/2410.23844/) (2024)
- [Performance Gap in Entity Knowledge Extraction Across Modalities in Vision Language Models](https://mlsanity.com/paper/2412.14133/) (2024)
- [Understanding Multimodal LLMs: the Mechanistic Interpretability of Llava in Visual Question Answering](https://mlsanity.com/paper/2411.10950/) (2024)

## Built on by (most cited first, 1 of 1)

- [Open-MMUnlearning: Unifying Methods and Evaluation for MLLM Unlearning](https://mlsanity.com/paper/2610.10358/) (2026)

## Abstract (arXiv)

Recently, machine unlearning approaches have been proposed to remove sensitive information from well-trained large models. However, most existing methods are tailored for LLMs, while MLLM-oriented unlearning remains at its early stage. Inspired by recent studies exploring the internal mechanisms of MLLMs, we propose to disentangle the visual and textual knowledge embedded within MLLMs and introduce a dedicated approach to selectively erase target visual knowledge while preserving textual knowledge. Unlike previous unlearning methods that rely on output-level supervision, our approach introduces a Visual Knowledge Distillation (VKD) scheme, which leverages intermediate visual representations within the MLLM as supervision signals. This design substantially enhances both unlearning effectiveness and model utility. Moreover, since our method only fine-tunes the visual components of the MLLM, it offers significant efficiency advantages. Extensive experiments demonstrate that our approach outperforms state-of-the-art unlearning methods in terms of both effectiveness and efficiency. Moreover, we are the first to evaluate the robustness of MLLM unlearning against relearning attacks.