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

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; s...

Full description

Saved in:
Bibliographic Details
Main Authors Fleshman, William, Khan, Aleem, Marone, Marc, Van Durme, Benjamin
Format Journal Article
LanguageEnglish
Published 12.04.2024
Subjects
Computer Science - Artificial Intelligence
Computer Science - Computation and Language
Computer Science - Learning
Online AccessGet full text

Cover

More Information
Summary: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.
DOI:10.48550/arxiv.2404.08417