Pruning as a Defense: Reducing Memorization in Large Language Models
Large language models have been shown to memorize significant portions of their training data, which they can reproduce when appropriately prompted. This work investigates the impact of simple pruning techniques on this behavior. Our findings reveal that pruning effectively reduces the extent of memorization in LLMs, demonstrating its potential as a foundational approach for mitigating membership inference attacks.
Code (0)
등록된 구현이 없습니다.
Tasks
MemorizationMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Private Memorization Editing: Turning Memorization into a Defense to Strengthen Data Privacy in Large Language Models
Large Language Models (LLMs) memorize, and thus, among huge amounts of uncontrolled data, may memorize Personally Identifiable Information (PII), which should not be stored and, consequently, not leaked. In this paper, w…
Language ModelingLanguage ModellingMemorizationData-centric NLP Backdoor Defense from the Lens of Memorization
Backdoor attack is a severe threat to the trustworthiness of DNN-based language models. In this paper, we first extend the definition of memorization of language models from sample-wise to more fine-grained sentence elem…
Backdoor Attackbackdoor defenseLanguage ModelingLanguage Modelling+2Preventing Verbatim Memorization in Language Models Gives a False Sense of Privacy
Studying data memorization in neural language models helps us understand the risks (e.g., to privacy or copyright) associated with models regurgitating training data and aids in the development of countermeasures. Many p…
MemorizationOpen-Ended Question AnsweringStyle TransferMembership Inference Attacks and Defenses in Neural Network Pruning
Neural network pruning has been an essential technique to reduce the computation and memory requirements for using deep neural networks for resource-constrained devices. Most existing research focuses primarily on balanc…
Inference AttackMembership Inference AttackMemorizationNetwork PruningOn the Effect of Pruning on Adversarial Robustness
Pruning is a well-known mechanism for reducing the computational cost of deep convolutional networks. However, studies have shown the potential of pruning as a form of regularization, which reduces overfitting and improv…
Adversarial Robustness