Does your LLM truly unlearn? An embarrassingly simple approach to recover unlearned knowledge

• URL: http://arxiv.org/abs/2410.16454v1
• Date: Mon, 21 Oct 2024 19:28:37 GMT
• Title: Does your LLM truly unlearn? An embarrassingly simple approach to recover unlearned knowledge
• Authors: Zhiwei Zhang, Fali Wang, Xiaomin Li, Zongyu Wu, Xianfeng Tang, Hui Liu, Qi He, Wenpeng Yin, Suhang Wang,
• Abstract summary: We show that applying quantization to models that have undergone unlearning can restore the "forgotten" information. We find that for unlearning methods with utility constraints, the unlearned model retains an average of 21% of the intended forgotten knowledge in full precision.
• Score: 36.524827594501495
• License:
• Abstract: Large language models (LLMs) have shown remarkable proficiency in generating text, benefiting from extensive training on vast textual corpora. However, LLMs may also acquire unwanted behaviors from the diverse and sensitive nature of their training data, which can include copyrighted and private content. Machine unlearning has been introduced as a viable solution to remove the influence of such problematic content without the need for costly and time-consuming retraining. This process aims to erase specific knowledge from LLMs while preserving as much model utility as possible. Despite the effectiveness of current unlearning methods, little attention has been given to whether existing unlearning methods for LLMs truly achieve forgetting or merely hide the knowledge, which current unlearning benchmarks fail to detect. This paper reveals that applying quantization to models that have undergone unlearning can restore the "forgotten" information. To thoroughly evaluate this phenomenon, we conduct comprehensive experiments using various quantization techniques across multiple precision levels. We find that for unlearning methods with utility constraints, the unlearned model retains an average of 21\% of the intended forgotten knowledge in full precision, which significantly increases to 83\% after 4-bit quantization. Based on our empirical findings, we provide a theoretical explanation for the observed phenomenon and propose a quantization-robust unlearning strategy to mitigate this intricate issue...

Related papers

• Does Unlearning Truly Unlearn? A Black Box Evaluation of LLM Unlearning Methods [1.9799527196428242]
  Large language model unlearning aims to remove harmful information that LLMs have learnt to prevent their use for malicious purposes. LMU and RMU have been proposed as two methods for LLM unlearning, achieving impressive results on unlearning benchmarks.
  arXiv  Detail & Related papers  (2024-11-18T22:31:17Z)
• 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)
• To Forget or Not? Towards Practical Knowledge Unlearning for Large Language Models [39.39428450239399]
  Large Language Models (LLMs) trained on extensive corpora inevitably retain sensitive data, such as personal privacy information and copyrighted material. Recent advancements in knowledge unlearning involve updating LLM parameters to erase specific knowledge. We introduce KnowUnDo to evaluate if the unlearning process inadvertently erases essential knowledge.
  arXiv  Detail & Related papers  (2024-07-02T03:34:16Z)
• 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)
• Unlearning with Control: Assessing Real-world Utility for Large Language Model Unlearning [97.2995389188179]
  Recent research has begun to approach large language models (LLMs) unlearning via gradient ascent (GA) Despite their simplicity and efficiency, we suggest that GA-based methods face the propensity towards excessive unlearning. We propose several controlling methods that can regulate the extent of excessive unlearning.
  arXiv  Detail & Related papers  (2024-06-13T14:41:00Z)
• Offset Unlearning for Large Language Models [49.851093293780615]
  Unlearning has emerged as a potential remedy for Large Language Models affected by problematic training data. We propose $delta$-unlearning, an offset unlearning framework for black-box LLMs. Experiments demonstrate that $delta$-unlearning can effectively unlearn target data while maintaining similar or even stronger performance on general out-of-forget-scope tasks.
  arXiv  Detail & Related papers  (2024-04-17T03:39:51Z)
• Rethinking Machine Unlearning for Large Language Models [85.92660644100582]
  We explore machine unlearning in the domain of large language models (LLMs) This initiative aims to eliminate undesirable data influence (e.g., sensitive or illegal information) and the associated model capabilities.
  arXiv  Detail & Related papers  (2024-02-13T20:51:58Z)
• Knowledge Unlearning for LLMs: Tasks, Methods, and Challenges [11.228131492745842]
  Large language models (LLMs) have spurred a new research paradigm in natural language processing. Despite their excellent capability in knowledge-based question answering and reasoning, their potential to retain faulty or even harmful knowledge poses risks of malicious application. Knowledge unlearning, derived from analogous studies on machine unlearning, presents a promising avenue to address this concern.
  arXiv  Detail & Related papers  (2023-11-27T12:37:51Z)
• Thrust: Adaptively Propels Large Language Models with External Knowledge [58.72867916604562]
  Large-scale pre-trained language models (PTLMs) are shown to encode rich knowledge in their model parameters. The inherent knowledge in PTLMs can be opaque or static, making external knowledge necessary. We propose the instance-level adaptive propulsion of external knowledge (IAPEK), where we only conduct the retrieval when necessary.
  arXiv  Detail & Related papers  (2023-07-19T20:16:46Z)
• Learn to Unlearn: A Survey on Machine Unlearning [29.077334665555316]
  This article presents a review of recent machine unlearning techniques, verification mechanisms, and potential attacks. We highlight emerging challenges and prospective research directions. We aim for this paper to provide valuable resources for integrating privacy, equity, andresilience into ML systems.
  arXiv  Detail & Related papers  (2023-05-12T14:28:02Z)

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.