Crayon: Customized On-Device LLM via Instant Adapter Blending and Edge-Server Hybrid Inference

Jihwan Bang, Juntae Lee, Kyuhong Shim, Seunghan Yang, Simyung Chang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

The customization of large language models (LLMs) for user-specified tasks gets important. However, maintaining all the customized LLMs on cloud servers incurs substantial memory and computational overheads, and uploading user data can also lead to privacy concerns. On-device LLMs can offer a promising solution by mitigating these issues. Yet, the performance of on-device LLMs is inherently constrained by the limitations of small-scaled models. To overcome these restrictions, we first propose Crayon, a novel approach for on-device LLM customization. Crayon begins by constructing a pool of diverse base adapters, and then we instantly blend them into a customized adapter without extra training. In addition, we develop a device-server hybrid inference strategy, which deftly allocates more demanding queries or non-customized tasks to a larger, more capable LLM on a server. This ensures optimal performance without sacrificing the benefits of on-device customization. We carefully craft a novel benchmark from multiple question-answer datasets, and show the efficacy of our method in the LLM customization.

Original languageEnglish
Title of host publicationLong Papers
EditorsLun-Wei Ku, Andre F. T. Martins, Vivek Srikumar
PublisherAssociation for Computational Linguistics (ACL)
Pages3720-3731
Number of pages12
ISBN (Electronic)9798891760943
DOIs
StatePublished - 2024
Externally publishedYes
Event62nd Annual Meeting of the Association for Computational Linguistics, ACL 2024 - Bangkok, Thailand
Duration: 11 Aug 202416 Aug 2024

Publication series

NameProceedings of the Annual Meeting of the Association for Computational Linguistics
Volume1
ISSN (Print)0736-587X

Conference

Conference62nd Annual Meeting of the Association for Computational Linguistics, ACL 2024
Country/TerritoryThailand
CityBangkok
Period11/08/2416/08/24

Fingerprint

Dive into the research topics of 'Crayon: Customized On-Device LLM via Instant Adapter Blending and Edge-Server Hybrid Inference'. Together they form a unique fingerprint.

Cite this