EDGE-LLM: Enabling Efficient Large Language Model Adaptation on Edge Devices via Layerwise Unified Compression and Adaptive Layer Tuning and Voting

TL;DR

Edge-LLM is a novel framework that enables efficient adaptation of large language models on edge devices by combining layer-wise compression, adaptive layer tuning, and hardware-aware scheduling, significantly reducing computation and memory overheads.

Contribution

The paper introduces Edge-LLM, a comprehensive framework that integrates layer-wise compression, adaptive tuning, and hardware scheduling for efficient LLM adaptation on edge devices, addressing limitations of existing methods.

Findings

Achieves 2.92x speedup over vanilla tuning methods.

Reduces memory overhead by 4x while maintaining accuracy.

Demonstrates effectiveness through extensive experiments.

Abstract

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