Breaking the Gold Standard: Extracting Forgotten Data under Exact Unlearning in Large Language Models

• URL: http://arxiv.org/abs/2505.24379v1
• Date: Fri, 30 May 2025 09:09:33 GMT
• Title: Breaking the Gold Standard: Extracting Forgotten Data under Exact Unlearning in Large Language Models
• Authors: Xiaoyu Wu, Yifei Pang, Terrance Liu, Zhiwei Steven Wu,
• Abstract summary: We introduce a novel data extraction attack that compromises even exact unlearning.<n>We demonstrate our attack's effectiveness on a simulated medical diagnosis dataset.
• Score: 26.5039481643457
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models are typically trained on datasets collected from the web, which may inadvertently contain harmful or sensitive personal information. To address growing privacy concerns, unlearning methods have been proposed to remove the influence of specific data from trained models. Of these, exact unlearning -- which retrains the model from scratch without the target data -- is widely regarded the gold standard, believed to be robust against privacy-related attacks. In this paper, we challenge this assumption by introducing a novel data extraction attack that compromises even exact unlearning. Our method leverages both the pre- and post-unlearning models: by guiding the post-unlearning model using signals from the pre-unlearning model, we uncover patterns that reflect the removed data distribution. Combining model guidance with a token filtering strategy, our attack significantly improves extraction success rates -- doubling performance in some cases -- across common benchmarks such as MUSE, TOFU, and WMDP. Furthermore, we demonstrate our attack's effectiveness on a simulated medical diagnosis dataset to highlight real-world privacy risks associated with exact unlearning. In light of our findings, which suggest that unlearning may, in a contradictory way, increase the risk of privacy leakage, we advocate for evaluation of unlearning methods to consider broader threat models that account not only for post-unlearning models but also for adversarial access to prior checkpoints.

Related papers

• Forget to Flourish: Leveraging Machine-Unlearning on Pretrained Language Models for Privacy Leakage [12.892449128678516]
  Fine-tuning language models on private data for downstream applications poses significant privacy risks. Several popular community platforms now offer convenient distribution of a large variety of pre-trained models. We introduce a novel poisoning technique that uses model-unlearning as an attack tool.
  arXiv  Detail & Related papers  (2024-08-30T15:35:09Z)
• Learn while Unlearn: An Iterative Unlearning Framework for Generative Language Models [52.03511469562013]
  We introduce the Iterative Contrastive Unlearning (ICU) framework, which consists of three core components.<n>A Knowledge Unlearning Induction module targets specific knowledge for removal using an unlearning loss.<n>A Contrastive Learning Enhancement module preserves the model's expressive capabilities against the pure unlearning goal.<n>An Iterative Unlearning Refinement module dynamically adjusts the unlearning process through ongoing evaluation and updates.
  arXiv  Detail & Related papers  (2024-07-25T07:09:35Z)
• Privacy Backdoors: Enhancing Membership Inference through Poisoning Pre-trained Models [112.48136829374741]
  In this paper, we unveil a new vulnerability: the privacy backdoor attack. When a victim fine-tunes a backdoored model, their training data will be leaked at a significantly higher rate than if they had fine-tuned a typical model. Our findings highlight a critical privacy concern within the machine learning community and call for a reevaluation of safety protocols in the use of open-source pre-trained models.
  arXiv  Detail & Related papers  (2024-04-01T16:50:54Z)
• Assessing Privacy Risks in Language Models: A Case Study on Summarization Tasks [65.21536453075275]
  We focus on the summarization task and investigate the membership inference (MI) attack. We exploit text similarity and the model's resistance to document modifications as potential MI signals. We discuss several safeguards for training summarization models to protect against MI attacks and discuss the inherent trade-off between privacy and utility.
  arXiv  Detail & Related papers  (2023-10-20T05:44:39Z)
• Leveraging Adversarial Examples to Quantify Membership Information Leakage [30.55736840515317]
  We develop a novel approach to address the problem of membership inference in pattern recognition models. We argue that this quantity reflects the likelihood of belonging to the training data. Our method performs comparable or even outperforms state-of-the-art strategies.
  arXiv  Detail & Related papers  (2022-03-17T19:09:38Z)
• Learning to Learn Transferable Attack [77.67399621530052]
  Transfer adversarial attack is a non-trivial black-box adversarial attack that aims to craft adversarial perturbations on the surrogate model and then apply such perturbations to the victim model. We propose a Learning to Learn Transferable Attack (LLTA) method, which makes the adversarial perturbations more generalized via learning from both data and model augmentation. Empirical results on the widely-used dataset demonstrate the effectiveness of our attack method with a 12.85% higher success rate of transfer attack compared with the state-of-the-art methods.
  arXiv  Detail & Related papers  (2021-12-10T07:24:21Z)
• Enhanced Membership Inference Attacks against Machine Learning Models [9.26208227402571]
  Membership inference attacks are used to quantify the private information that a model leaks about the individual data points in its training set. We derive new attack algorithms that can achieve a high AUC score while also highlighting the different factors that affect their performance. Our algorithms capture a very precise approximation of privacy loss in models, and can be used as a tool to perform an accurate and informed estimation of privacy risk in machine learning models.
  arXiv  Detail & Related papers  (2021-11-18T13:31:22Z)
• Delving into Data: Effectively Substitute Training for Black-box Attack [84.85798059317963]
  We propose a novel perspective substitute training that focuses on designing the distribution of data used in the knowledge stealing process. The combination of these two modules can further boost the consistency of the substitute model and target model, which greatly improves the effectiveness of adversarial attack.
  arXiv  Detail & Related papers  (2021-04-26T07:26:29Z)
• Quantifying and Mitigating Privacy Risks of Contrastive Learning [4.909548818641602]
  We perform the first privacy analysis of contrastive learning through the lens of membership inference and attribute inference. Our results show that contrastive models are less vulnerable to membership inference attacks but more vulnerable to attribute inference attacks compared to supervised models. To remedy this situation, we propose the first privacy-preserving contrastive learning mechanism, namely Talos.
  arXiv  Detail & Related papers  (2021-02-08T11:38:11Z)
• Knowledge-Enriched Distributional Model Inversion Attacks [49.43828150561947]
  Model inversion (MI) attacks are aimed at reconstructing training data from model parameters. We present a novel inversion-specific GAN that can better distill knowledge useful for performing attacks on private models from public data. Our experiments show that the combination of these techniques can significantly boost the success rate of the state-of-the-art MI attacks by 150%.
  arXiv  Detail & Related papers  (2020-10-08T16:20:48Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.