AdapterSwap: Continuous Training of LLMs with Data Removal and Access-Control Guarantees

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Fleshman, William, Khan, Aleem, Marone, Marc, Van Durme, Benjamin
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866929704381972480
author Fleshman, William
Khan, Aleem
Marone, Marc
Van Durme, Benjamin
author_facet Fleshman, William
Khan, Aleem
Marone, Marc
Van Durme, Benjamin
contents Large language models (LLMs) are increasingly capable of completing knowledge intensive tasks by recalling information from a static pretraining corpus. Here we are concerned with LLMs in the context of evolving data requirements. For instance: batches of new data that are introduced periodically; subsets of data with user-based access controls; or requirements on dynamic removal of documents with guarantees that associated knowledge cannot be recalled. We wish to satisfy these requirements while at the same time ensuring a model does not forget old information when new data becomes available. To address these issues, we introduce AdapterSwap, a training and inference scheme that organizes knowledge from a data collection into a set of low-rank adapters, which are dynamically composed during inference. Our experiments demonstrate AdapterSwap's ability to support efficient continual learning, while also enabling organizations to have fine-grained control over data access and deletion.
format Preprint
id arxiv_https___arxiv_org_abs_2404_08417
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle AdapterSwap: Continuous Training of LLMs with Data Removal and Access-Control Guarantees
Fleshman, William
Khan, Aleem
Marone, Marc
Van Durme, Benjamin
Machine Learning
Artificial Intelligence
Computation and Language
Large language models (LLMs) are increasingly capable of completing knowledge intensive tasks by recalling information from a static pretraining corpus. Here we are concerned with LLMs in the context of evolving data requirements. For instance: batches of new data that are introduced periodically; subsets of data with user-based access controls; or requirements on dynamic removal of documents with guarantees that associated knowledge cannot be recalled. We wish to satisfy these requirements while at the same time ensuring a model does not forget old information when new data becomes available. To address these issues, we introduce AdapterSwap, a training and inference scheme that organizes knowledge from a data collection into a set of low-rank adapters, which are dynamically composed during inference. Our experiments demonstrate AdapterSwap's ability to support efficient continual learning, while also enabling organizations to have fine-grained control over data access and deletion.
title AdapterSwap: Continuous Training of LLMs with Data Removal and Access-Control Guarantees
topic Machine Learning
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2404.08417