Textual Unlearning Gives a False Sense of Unlearning

• URL: http://arxiv.org/abs/2406.13348v1
• Date: Wed, 19 Jun 2024 08:51:54 GMT
• Title: Textual Unlearning Gives a False Sense of Unlearning
• Authors: Jiacheng Du, Zhibo Wang, Kui Ren,
• Abstract summary: Language models (LMs) are susceptible to "memorizing" training data, including a large amount of private or copyright-protected content. We propose the Textual Unlearning Leakage Attack (TULA), where an adversary can infer information about unlearned data only by accessing the models before and after unlearning. Our work is the first to reveal that machine unlearning in LMs can inversely create greater knowledge risks and inspire the development of more secure unlearning mechanisms.
• Score: 12.792770622915906
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Language models (LMs) are susceptible to "memorizing" training data, including a large amount of private or copyright-protected content. To safeguard the right to be forgotten (RTBF), machine unlearning has emerged as a promising method for LMs to efficiently "forget" sensitive training content and mitigate knowledge leakage risks. However, despite its good intentions, could the unlearning mechanism be counterproductive? In this paper, we propose the Textual Unlearning Leakage Attack (TULA), where an adversary can infer information about the unlearned data only by accessing the models before and after unlearning. Furthermore, we present variants of TULA in both black-box and white-box scenarios. Through various experimental results, we critically demonstrate that machine unlearning amplifies the risk of knowledge leakage from LMs. Specifically, TULA can increase an adversary's ability to infer membership information about the unlearned data by more than 20% in black-box scenario. Moreover, TULA can even reconstruct the unlearned data directly with more than 60% accuracy with white-box access. Our work is the first to reveal that machine unlearning in LMs can inversely create greater knowledge risks and inspire the development of more secure unlearning mechanisms.

Related papers

• A Closer Look at Machine Unlearning for Large Language Models [46.245404272612795]
  Large language models (LLMs) may memorize sensitive or copyrighted content, raising privacy and legal concerns. We discuss several issues in machine unlearning for LLMs and provide our insights on possible approaches.
  arXiv  Detail & Related papers  (2024-10-10T16:56:05Z)
• Verification of Machine Unlearning is Fragile [48.71651033308842]
  We introduce two novel adversarial unlearning processes capable of circumventing both types of verification strategies. This study highlights the vulnerabilities and limitations in machine unlearning verification, paving the way for further research into the safety of machine unlearning.
  arXiv  Detail & Related papers  (2024-08-01T21:37:10Z)
• UnUnlearning: Unlearning is not sufficient for content regulation in advanced generative AI [50.61495097098296]
  We revisit the paradigm in which unlearning is used for Large Language Models (LLMs) We introduce a concept of ununlearning, where unlearned knowledge gets reintroduced in-context. We argue that content filtering for impermissible knowledge will be required and even exact unlearning schemes are not enough for effective content regulation.
  arXiv  Detail & Related papers  (2024-06-27T10:24:35Z)
• Machine Unlearning: A Comprehensive Survey [14.235752586133158]
  This survey aims to systematically classify a wide range of machine unlearning. We categorize current unlearning methods into four scenarios: centralized unlearning, distributed and irregular data unlearning, unlearning verification, and privacy and security issues in unlearning.
  arXiv  Detail & Related papers  (2024-05-13T00:58:34Z)
• Learn What You Want to Unlearn: Unlearning Inversion Attacks against Machine Unlearning [16.809644622465086]
  We conduct the first investigation to understand the extent to which machine unlearning can leak the confidential content of unlearned data. Under the Machine Learning as a Service setting, we propose unlearning inversion attacks that can reveal the feature and label information of an unlearned sample. The experimental results indicate that the proposed attack can reveal the sensitive information of the unlearned data.
  arXiv  Detail & Related papers  (2024-04-04T06:37:46Z)
• The Frontier of Data Erasure: Machine Unlearning for Large Language Models [56.26002631481726]
  Large Language Models (LLMs) are foundational to AI advancements. LLMs pose risks by potentially memorizing and disseminating sensitive, biased, or copyrighted information. Machine unlearning emerges as a cutting-edge solution to mitigate these concerns.
  arXiv  Detail & Related papers  (2024-03-23T09:26:15Z)
• Harnessing the Power of Text-image Contrastive Models for Automatic Detection of Online Misinformation [50.46219766161111]
  We develop a self-learning model to explore the constrastive learning in the domain of misinformation identification. Our model shows the superior performance of non-matched image-text pair detection when the training data is insufficient.
  arXiv  Detail & Related papers  (2023-04-19T02:53:59Z)
• Context-faithful Prompting for Large Language Models [51.194410884263135]
  Large language models (LLMs) encode parametric knowledge about world facts. Their reliance on parametric knowledge may cause them to overlook contextual cues, leading to incorrect predictions in context-sensitive NLP tasks. We assess and enhance LLMs' contextual faithfulness in two aspects: knowledge conflict and prediction with abstention.
  arXiv  Detail & Related papers  (2023-03-20T17:54:58Z)

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.