The field of edge computing and large language models is rapidly evolving, with a focus on improving efficiency, reducing latency, and enhancing decision-making. Recent developments have led to the creation of innovative frameworks, such as adaptive model partitioning and hybrid edge-cloud resource allocation, which optimize inference and reduce costs. The application of large language models in edge computing has also shown great promise, with potential uses in urban computing, autonomous drone navigation, and real-time data analysis. Noteworthy papers in this area include SimDC, which proposes a high-fidelity device simulation platform for device-cloud collaborative computing, and Fragile Mastery, which investigates the trade-offs between domain-specific optimization and cross-domain robustness in on-device language models. Other notable papers, such as LAURA and AMP4EC, demonstrate the effectiveness of LLM-assisted UAV routing and adaptive model partitioning in edge computing environments.
Advances in Edge Computing and Large Language Models
Sources
SimDC: A High-Fidelity Device Simulation Platform for Device-Cloud Collaborative Computing
Fragile Mastery: Are Domain-Specific Trade-Offs Undermining On-Device Language Models?
PartialLoading: User Scheduling and Bandwidth Allocation for Parameter-sharing Edge Inference
LAURA: LLM-Assisted UAV Routing for AoI Minimization
Deep Learning Model Deployment in Multiple Cloud Providers: an Exploratory Study Using Low Computing Power Environments
Moving Edge for On-Demand Edge Computing: An Uncertainty-aware Approach
Are We There Yet? A Measurement Study of Efficiency for LLM Applications on Mobile Devices
Text Chunking for Document Classification for Urban System Management using Large Language Models
AMP4EC: Adaptive Model Partitioning Framework for Efficient Deep Learning Inference in Edge Computing Environments
HERA: Hybrid Edge-cloud Resource Allocation for Cost-Efficient AI Agents
Contextualized Autonomous Drone Navigation using LLMs Deployed in Edge-Cloud Computing
Estimating Hourly Neighborhood Population Using Mobile Phone Data in the United States
Urban Computing in the Era of Large Language Models
LLMPi: Optimizing LLMs for High-Throughput on Raspberry Pi