Guardado en:
Detalles Bibliográficos
Autores principales: Yu, Zhongzhi, Wang, Zheng, Li, Yuhan, You, Haoran, Gao, Ruijie, Zhou, Xiaoya, Bommu, Sreenidhi Reedy, Zhao, Yang Katie, Lin, Yingyan Celine
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2406.15758
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916298616733696
author Yu, Zhongzhi
Wang, Zheng
Li, Yuhan
You, Haoran
Gao, Ruijie
Zhou, Xiaoya
Bommu, Sreenidhi Reedy
Zhao, Yang Katie
Lin, Yingyan Celine
author_facet Yu, Zhongzhi
Wang, Zheng
Li, Yuhan
You, Haoran
Gao, Ruijie
Zhou, Xiaoya
Bommu, Sreenidhi Reedy
Zhao, Yang Katie
Lin, Yingyan Celine
contents Efficient adaption of large language models (LLMs) on edge devices is essential for applications requiring continuous and privacy-preserving adaptation and inference. However, existing tuning techniques fall short because of the high computation and memory overheads. To this end, we introduce a computation- and memory-efficient LLM tuning framework, called Edge-LLM, to facilitate affordable and effective LLM adaptation on edge devices. Specifically, Edge-LLM features three core components: (1) a layer-wise unified compression (LUC) technique to reduce the computation overhead by generating layer-wise pruning sparsity and quantization bit-width policies, (2) an adaptive layer tuning and voting scheme to reduce the memory overhead by reducing the backpropagation depth, and (3) a complementary hardware scheduling strategy to handle the irregular computation patterns introduced by LUC and adaptive layer tuning, thereby achieving efficient computation and data movements. Extensive experiments demonstrate that Edge-LLM achieves a 2.92x speed up and a 4x memory overhead reduction as compared to vanilla tuning methods with comparable task accuracy. Our code is available at https://github.com/GATECH-EIC/Edge-LLM
format Preprint
id arxiv_https___arxiv_org_abs_2406_15758
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle EDGE-LLM: Enabling Efficient Large Language Model Adaptation on Edge Devices via Layerwise Unified Compression and Adaptive Layer Tuning and Voting
Yu, Zhongzhi
Wang, Zheng
Li, Yuhan
You, Haoran
Gao, Ruijie
Zhou, Xiaoya
Bommu, Sreenidhi Reedy
Zhao, Yang Katie
Lin, Yingyan Celine
Machine Learning
Distributed, Parallel, and Cluster Computing
Efficient adaption of large language models (LLMs) on edge devices is essential for applications requiring continuous and privacy-preserving adaptation and inference. However, existing tuning techniques fall short because of the high computation and memory overheads. To this end, we introduce a computation- and memory-efficient LLM tuning framework, called Edge-LLM, to facilitate affordable and effective LLM adaptation on edge devices. Specifically, Edge-LLM features three core components: (1) a layer-wise unified compression (LUC) technique to reduce the computation overhead by generating layer-wise pruning sparsity and quantization bit-width policies, (2) an adaptive layer tuning and voting scheme to reduce the memory overhead by reducing the backpropagation depth, and (3) a complementary hardware scheduling strategy to handle the irregular computation patterns introduced by LUC and adaptive layer tuning, thereby achieving efficient computation and data movements. Extensive experiments demonstrate that Edge-LLM achieves a 2.92x speed up and a 4x memory overhead reduction as compared to vanilla tuning methods with comparable task accuracy. Our code is available at https://github.com/GATECH-EIC/Edge-LLM
title EDGE-LLM: Enabling Efficient Large Language Model Adaptation on Edge Devices via Layerwise Unified Compression and Adaptive Layer Tuning and Voting
topic Machine Learning
Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2406.15758