Edge Computing and TinyML
Deployment of LLMs in resource-constrained edge devices and IoT applications with optimization for efficiency
Edge Computing and TinyML
Research on Large Language Models in Edge Computing and TinyML
LLMs as TinyML Lifecycle Enablers
Leveraging Large Language Models to streamline embedded AI development
Accelerating AI at the Edge
Optimizing Deep Learning for Resource-Constrained Environments
Next-Gen AI for Connected Systems
Reimagining CPS-IoT with Foundation Models
Efficient LLM Fine-Tuning on Edge Devices
A novel federated sketching approach for resource-constrained environments
Intelligent IoT Systems Through LLMs
Enabling Dynamic IoT Environments via Mixed-Initiative Interaction
Smarter Pruning for Smaller LMs
Adaptive structural pruning outperforms traditional approaches
Smart Routing for On-Device AI
Optimizing LLM Performance Through Uncertainty-Based Decision Making
Bringing LLMs to the Edge
Innovative pruning-aware pretraining for efficient language models
Accelerating Graph Neural Networks at the Edge
Hardware-optimized GNN execution for resource-constrained environments
Breaking Memory Barriers in NLP
Making Large Language Models Work on Tiny Devices
Efficient LLM Fine-Tuning at the Edge
Optimizing language models for resource-constrained devices
Optimizing MLLMs for Edge-Cloud Federated Learning
A Swarm Intelligence Approach to Deploy Advanced AI Models at the Edge
Breaking Memory Limits for Edge-Based LLMs
Maximizing FPGA potential for efficient LLM deployment at the edge
Edge-Powered Mobile AIGC Services
Enhancing AI Content Generation through Interactive Prompts and Dynamic Service Allocation
Smaller, Smarter AI Models
Building Efficient Models with Strong Reasoning Capabilities
Efficient Edge Inference for Ternary LLMs
Accelerating Large Language Models on Resource-Constrained Devices
Supercharging LLMs on Standard CPUs
How SparAMX makes AI more accessible through CPU optimization
Enabling LLMs on Edge Devices
A breakthrough approach for distributed LLM inference across multiple devices
Revolutionizing On-Device LLM Fine-Tuning
Fully Quantized Training with Integer-Only Operations
LLMs at the Edge: XR Device Performance Analysis
Benchmarking 17 language models across leading XR platforms
Bringing GenAI to the Edge
Optimizing AI models for secure, low-latency deployment on edge devices
Mobile-Friendly LLM Fine-Tuning
Enabling personalized AI on resource-constrained devices
Bringing AI Power to Your Device
CoSMoEs: Making Advanced AI Models Run Efficiently on Mobile Devices
Bringing LLM Power to IoT Edge Devices
Fine-tuned small language models for specialized IoT development
Optimizing LLMs for Edge Computing
Balancing performance and resource constraints at the network edge
FlexInfer: Breaking Device Memory Barriers
Enabling efficient LLM inference on resource-constrained devices
Automating IoT Development with LLMs
Natural language programming for secure AIoT applications
GenieBlue: Mobile-Optimized Multimodal AI
Enhancing On-Device Language Models Without Sacrificing Capabilities
Making LLM Fine-Tuning Private & Efficient
Using Layer Dropout to Enhance Federated Learning for LLMs
Optimizing LLMs for Edge Devices
Novel memory-efficient techniques for low-power environments
Edge-Optimized Language Models
Hardware-co-designed PLMs for resource-constrained devices
Intelligent Search for IoT Ecosystems
Unifying fragmented IoT data through agentic search capabilities
Intelligent IoT: The LLM Revolution for 6G Networks
A new architecture for autonomous, intelligent IoT systems
Collaborative LLM Inference at the Edge
Distributing LLM workloads across multiple devices for efficient processing
SplitFrozen: Making LLMs Work on Edge Devices
Efficient fine-tuning for resource-constrained environments
HAT: Rethinking LLM Deployment
A Device-Cloud Collaborative Framework for Faster, More Private LLMs
Breaking Memory Barriers for AI on Edge Devices
A solution for infinite context windows on resource-constrained hardware
Deploying LLMs on Mobile: Efficiency Tradeoffs
Measuring performance across mobile, edge, and cloud deployments
Optimizing LLMs for Edge Devices
Enabling high-throughput language models on resource-constrained hardware
Optimizing LLMs for Edge Devices
Quantization strategies for efficient AI at the edge
Accelerating LLMs on Consumer Devices
Pipelined Offloading for Efficient Inference with Limited GPU Memory
TaskEdge: Bringing LLMs to Resource-Constrained Devices
Smart Fine-Tuning for Edge Computing
Intent-Driven Computing Across Devices
Leveraging LLMs for Intelligent Resource Management in Distributed Systems
SmolVLM: Efficient Vision-Language Models
Optimizing multimodal AI for resource-constrained environments
Hybrid LLM Systems for Faster, Smarter Inference
Optimizing AI Decision-Making with Threshold-Based Control
Smart Token Routing for Edge AI
Optimizing LLM inference on resource-constrained devices
Jupiter: Fast LLM Collaboration at the Edge
Enabling efficient LLM inference across resource-constrained devices
Breaking Memory Barriers for Mobile LLMs
Enabling larger language models on memory-constrained devices
Revolutionizing TinyML with LLMs
A novel framework for efficient, explainable edge AI
