UnlearnShield: Shielding Forgotten Privacy against Unlearning Inversion
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914286077476864 |
|---|---|
| author | Xue, Lulu Hu, Shengshan Lu, Wei Zhou, Ziqi Song, Yufei Cheng, Jianhong Li, Minghui Zhang, Yanjun Zhang, Leo Yu |
| author_facet | Xue, Lulu Hu, Shengshan Lu, Wei Zhou, Ziqi Song, Yufei Cheng, Jianhong Li, Minghui Zhang, Yanjun Zhang, Leo Yu |
| contents | Machine unlearning is an emerging technique that aims to remove the influence of specific data from trained models, thereby enhancing privacy protection. However, recent research has uncovered critical privacy vulnerabilities, showing that adversaries can exploit unlearning inversion to reconstruct data that was intended to be erased. Despite the severity of this threat, dedicated defenses remain lacking. To address this gap, we propose UnlearnShield, the first defense specifically tailored to counter unlearning inversion. UnlearnShield introduces directional perturbations in the cosine representation space and regulates them through a constraint module to jointly preserve model accuracy and forgetting efficacy, thereby reducing inversion risk while maintaining utility. Experiments demonstrate that it achieves a good trade-off among privacy protection, accuracy, and forgetting. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_20325 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | UnlearnShield: Shielding Forgotten Privacy against Unlearning Inversion Xue, Lulu Hu, Shengshan Lu, Wei Zhou, Ziqi Song, Yufei Cheng, Jianhong Li, Minghui Zhang, Yanjun Zhang, Leo Yu Cryptography and Security Computer Vision and Pattern Recognition Machine unlearning is an emerging technique that aims to remove the influence of specific data from trained models, thereby enhancing privacy protection. However, recent research has uncovered critical privacy vulnerabilities, showing that adversaries can exploit unlearning inversion to reconstruct data that was intended to be erased. Despite the severity of this threat, dedicated defenses remain lacking. To address this gap, we propose UnlearnShield, the first defense specifically tailored to counter unlearning inversion. UnlearnShield introduces directional perturbations in the cosine representation space and regulates them through a constraint module to jointly preserve model accuracy and forgetting efficacy, thereby reducing inversion risk while maintaining utility. Experiments demonstrate that it achieves a good trade-off among privacy protection, accuracy, and forgetting. |
| title | UnlearnShield: Shielding Forgotten Privacy against Unlearning Inversion |
| topic | Cryptography and Security Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2601.20325 |