# UnlearnShield: Shielding Forgotten Privacy against Unlearning Inversion

arXiv: https://arxiv.org/abs/2601.20325 · HTML: https://mlsanity.com/paper/2601.20325/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Machine unlearning, one of 12 directions in LLM safety and security.
- Builds on 14 of its 19 references that are in the map.
- First posted on arXiv 28 Jan 2026.

## Placed in

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

## Builds on (most cited first, 14 of 14)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Understanding Black-box Predictions via Influence Functions](https://mlsanity.com/paper/1703.04730/) (2017)
- [Inverting Gradients -- How easy is it to break privacy in federated learning?](https://mlsanity.com/paper/2003.14053/) (2020)
- [Eternal Sunshine of the Spotless Net: Selective Forgetting in Deep Networks](https://mlsanity.com/paper/1911.04933/) (2019)
- [SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation](https://mlsanity.com/paper/2310.12508/) (2023)
- [Amnesiac Machine Learning](https://mlsanity.com/paper/2010.10981/) (2020)
- [Model Sparsity Can Simplify Machine Unlearning](https://mlsanity.com/paper/2304.04934/) (2023)
- [Evaluating Gradient Inversion Attacks and Defenses in Federated Learning](https://mlsanity.com/paper/2112.00059/) (2021)
- [Forgetting Outside the Box: Scrubbing Deep Networks of Information Accessible from Input-Output Observations](https://mlsanity.com/paper/2003.02960/) (2020)
- [Provable Defense against Privacy Leakage in Federated Learning from Representation Perspective](https://mlsanity.com/paper/2012.06043/) (2020)
- [Learn What You Want to Unlearn: Unlearning Inversion Attacks against Machine Unlearning](https://mlsanity.com/paper/2404.03233/) (2024)
- [Concealing Sensitive Samples against Gradient Leakage in Federated Learning](https://mlsanity.com/paper/2209.05724/) (2022)
- [Revisiting Gradient Pruning: A Dual Realization for Defending against Gradient Attacks](https://mlsanity.com/paper/2401.16687/) (2024)
- [Dual-View Inference Attack: Machine Unlearning Amplifies Privacy Exposure](https://mlsanity.com/paper/2512.16126/) (2025)

## Abstract (arXiv)

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.