MMUnlearner: Reformulating Multimodal Machine Unlearning in the Era of Multimodal Large Language Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909624399036416 |
|---|---|
| author | Huo, Jiahao Yan, Yibo Zheng, Xu Lyu, Yuanhuiyi Zou, Xin Wei, Zhihua Hu, Xuming |
| author_facet | Huo, Jiahao Yan, Yibo Zheng, Xu Lyu, Yuanhuiyi Zou, Xin Wei, Zhihua Hu, Xuming |
| contents | Recent progress in Machine Unlearning (MU) has introduced solutions for the selective removal of private or sensitive information encoded within deep neural networks. Nonetheless, MU for Multimodal Large Language Models (MLLMs) remains in its nascent phase. Therefore, we propose to reformulate the task of multimodal MU in the era of MLLMs, which aims to erase only the visual patterns associated with a given entity while preserving the corresponding textual knowledge encoded within the original parameters of the language model backbone. Furthermore, we develop a novel geometry-constrained gradient ascent method MMUnlearner. It updates the weights of MLLMs with a weight saliency map jointly restricted by the remaining concepts and textual knowledge during unlearning, thereby preserving parameters essential for non-target knowledge. Extensive experiments demonstrate that MMUnlearner surpasses baselines that finetuning MLLMs with VQA data directly through Gradient Ascent (GA) or Negative Preference Optimization (NPO), across all evaluation dimensions. Our code can be found in [this URL](https://github.com/Z1zs/MMUnlearner). |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_11051 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MMUnlearner: Reformulating Multimodal Machine Unlearning in the Era of Multimodal Large Language Models Huo, Jiahao Yan, Yibo Zheng, Xu Lyu, Yuanhuiyi Zou, Xin Wei, Zhihua Hu, Xuming Computation and Language Artificial Intelligence Recent progress in Machine Unlearning (MU) has introduced solutions for the selective removal of private or sensitive information encoded within deep neural networks. Nonetheless, MU for Multimodal Large Language Models (MLLMs) remains in its nascent phase. Therefore, we propose to reformulate the task of multimodal MU in the era of MLLMs, which aims to erase only the visual patterns associated with a given entity while preserving the corresponding textual knowledge encoded within the original parameters of the language model backbone. Furthermore, we develop a novel geometry-constrained gradient ascent method MMUnlearner. It updates the weights of MLLMs with a weight saliency map jointly restricted by the remaining concepts and textual knowledge during unlearning, thereby preserving parameters essential for non-target knowledge. Extensive experiments demonstrate that MMUnlearner surpasses baselines that finetuning MLLMs with VQA data directly through Gradient Ascent (GA) or Negative Preference Optimization (NPO), across all evaluation dimensions. Our code can be found in [this URL](https://github.com/Z1zs/MMUnlearner). |
| title | MMUnlearner: Reformulating Multimodal Machine Unlearning in the Era of Multimodal Large Language Models |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2502.11051 |