Unified Parameter-Efficient Unlearning for LLMs

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Ding, Chenlu, Wu, Jiancan, Yuan, Yancheng, Lu, Jinda, Zhang, Kai, Su, Alex, Wang, Xiang, He, Xiangnan
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866917989262032896
author Ding, Chenlu
Wu, Jiancan
Yuan, Yancheng
Lu, Jinda
Zhang, Kai
Su, Alex
Wang, Xiang
He, Xiangnan
author_facet Ding, Chenlu
Wu, Jiancan
Yuan, Yancheng
Lu, Jinda
Zhang, Kai
Su, Alex
Wang, Xiang
He, Xiangnan
contents The advent of Large Language Models (LLMs) has revolutionized natural language processing, enabling advanced understanding and reasoning capabilities across a variety of tasks. Fine-tuning these models for specific domains, particularly through Parameter-Efficient Fine-Tuning (PEFT) strategies like LoRA, has become a prevalent practice due to its efficiency. However, this raises significant privacy and security concerns, as models may inadvertently retain and disseminate sensitive or undesirable information. To address these issues, we introduce a novel instance-wise unlearning framework, LLMEraser, which systematically categorizes unlearning tasks and applies precise parameter adjustments using influence functions. Unlike traditional unlearning techniques that are often limited in scope and require extensive retraining, LLMEraser is designed to handle a broad spectrum of unlearning tasks without compromising model performance. Extensive experiments on benchmark datasets demonstrate that LLMEraser excels in efficiently managing various unlearning scenarios while maintaining the overall integrity and efficacy of the models.
format Preprint
id arxiv_https___arxiv_org_abs_2412_00383
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Unified Parameter-Efficient Unlearning for LLMs
Ding, Chenlu
Wu, Jiancan
Yuan, Yancheng
Lu, Jinda
Zhang, Kai
Su, Alex
Wang, Xiang
He, Xiangnan
Artificial Intelligence
Machine Learning
The advent of Large Language Models (LLMs) has revolutionized natural language processing, enabling advanced understanding and reasoning capabilities across a variety of tasks. Fine-tuning these models for specific domains, particularly through Parameter-Efficient Fine-Tuning (PEFT) strategies like LoRA, has become a prevalent practice due to its efficiency. However, this raises significant privacy and security concerns, as models may inadvertently retain and disseminate sensitive or undesirable information. To address these issues, we introduce a novel instance-wise unlearning framework, LLMEraser, which systematically categorizes unlearning tasks and applies precise parameter adjustments using influence functions. Unlike traditional unlearning techniques that are often limited in scope and require extensive retraining, LLMEraser is designed to handle a broad spectrum of unlearning tasks without compromising model performance. Extensive experiments on benchmark datasets demonstrate that LLMEraser excels in efficiently managing various unlearning scenarios while maintaining the overall integrity and efficacy of the models.
title Unified Parameter-Efficient Unlearning for LLMs
topic Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2412.00383