A Survey of Foundation Models for IoT: Taxonomy and Criteria-Based Analysis
2025/06/13 by Hui Wei, Dong Yoon Lee, Wei, Hui +11 · 2 citations
Computer Science · Engineering · #Advanced Data Processing Techniques #Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #FOS: Electrical engineering #IoT and Edge/Fog Computing #Machine Learning (cs.LG) #Systems and Control (eess.SY) #Traffic Prediction and Management Techniques #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2506.12263
openalex publication_date 2025/06/13 · openalex created_date 2025/10/11 · openalex updated_date 2026/07/28
Abstract
Foundation models have gained growing interest in the IoT domain due to their reduced reliance on labeled data and strong generalizability across tasks, which address key limitations of traditional machine learning approaches. However, most existing foundation model based methods are developed for specific IoT tasks, making it difficult to compare approaches across IoT domains and limiting guidance for applying them to new tasks. This survey aims to bridge this gap by providing a comprehensive overview of current methodologies and organizing them around four shared performance objectives by different domains: efficiency, context-awareness, safety, and security & privacy. For each objective, we review representative works, summarize commonly-used techniques and evaluation metrics. This objective-centric organization enables meaningful cross-domain comparisons and offers practical insights for selecting and designing foundation model based solutions for new IoT tasks. We conclude with key directions for future research to guide both practitioners and researchers in advancing the use of foundation models in IoT applications.
Citations
- DailyLLM: Context-Aware Activity Log Generation Using Multi-Modal Sensors and LLMs
- Safety at Scale: A Comprehensive Survey of Large Model and Agent Safety
- Large Language Models in the IoT Ecosystem -- A Survey on Security Challenges and Applications
- LLM-Powered AI Agent Systems and Their Applications in Industry
- How Memory Management Impacts LLM Agents: An Empirical Study of Experience-Following Behavior
- ContextAgent: Context-Aware Proactive LLM Agents with Open-World Sensory Perceptions
- Graph-Based Physics-Guided Urban PM2.5 Air Quality Imputation with Constrained Monitoring Data
- UserCentrix: An Agentic Memory-augmented AI Framework for Smart Spaces
- From LLM Reasoning to Autonomous AI Agents: A Comprehensive Review
- Medical Hallucinations in Foundation Models and Their Impact on Healthcare
- PlanGenLLMs: A Modern Survey of LLM Planning Capabilities
- LLM4WM: Adapting LLM for Wireless Multi-Tasking
- Large Language Model Safety: A Holistic Survey
- On the Structural Memory of LLM Agents
- ChainStream: An LLM-based Framework for Unified Synthetic Sensing
- WiFo: Wireless Foundation Model for Channel Prediction
- BARTPredict: Empowering IoT Security with LLM-Driven Cyber Threat Prediction
- SocialMind: LLM-based Proactive AR Social Assistive System with Human-like Perception for In-situ Live Interactions
- From Generation to Judgment: Opportunities and Challenges of LLM-as-a-judge
- A Survey on LLM-as-a-Judge
- MMBind: Unleashing the Potential of Distributed and Heterogeneous Data for Multimodal Learning in IoT
- Large Wireless Model (LWM): A Foundation Model for Wireless Channels
- Integrating Large Language Models with Internet of Things Applications
- Agents4PLC: Automating Closed-loop PLC Code Generation and Verification in Industrial Control Systems using LLM-based Agents
- Joint Verification and Refinement of Language Models for Safety-Constrained Planning
- Agent-as-a-Judge: Evaluate Agents with Agents
- SensorLLM: Aligning Large Language Models with Motion Sensors for Human Activity Recognition
- Large Model for Small Data: Foundation Model for Cross-Modal RF Human Activity Recognition
- Justice or Prejudice? Quantifying Biases in LLM-as-a-Judge
- LLMCount: Enhancing Stationary mmWave Detection with Multimodal-LLM
- LLMs Can Check Their Own Results to Mitigate Hallucinations in Traffic Understanding Tasks
- MindScape Study: Integrating LLM and Behavioral Sensing for Personalized AI-Driven Journaling Experiences
- LASP: Surveying the State-of-the-Art in Large Language Model-Assisted AI Planning
- HiAgent: Hierarchical Working Memory Management for Solving Long-Horizon Agent Tasks with Large Language Model
- Csi-LLM: A Novel Downlink Channel Prediction Method Aligned with LLM Pre-Training
- Leveraging Foundation Models for Zero-Shot IoT Sensing
- Multi-Step Reasoning with Large Language Models, a Survey
- IoT-LM: Large Multisensory Language Models for the Internet of Things
- Temporally Multi-Scale Sparse Self-Attention for Physical Activity Data Imputation
- LLM4CP: Adapting Large Language Models for Channel Prediction
- DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model
- From Persona to Personalization: A Survey on Role-Playing Language Agents
- A Survey on the Memory Mechanism of Large Language Model based Agents
- HGRN2: Gated Linear RNNs with State Expansion
- On the Efficiency and Robustness of Vibration-based Foundation Models for IoT Sensing: A Case Study
- LLMSense: Harnessing LLMs for High-level Reasoning Over Spatiotemporal Sensor Traces
- Securing Large Language Models: Threats, Vulnerabilities and Responsible Practices
- LLM-based Conversational AI Therapist for Daily Functioning Screening and Psychotherapeutic Intervention via Everyday Smart Devices
- Using Large Language Models to Compare Explainable Models for Smart Home Human Activity Recognition
- MEIT: Multimodal Electrocardiogram Instruction Tuning on Large Language Models for Report Generation
- HARGPT: Are LLMs Zero-Shot Human Activity Recognizers?
- Verifiably Following Complex Robot Instructions with Foundation Models
- Understanding the planning of LLM agents: A survey
- A Systematic Survey of Prompt Engineering in Large Language Models: Techniques and Applications
- Large Language Models for Time Series: A Survey
- Large Multi-Modal Models (LMMs) as Universal Foundation Models for AI-Native Wireless Systems
- EHRAgent: Code Empowers Large Language Models for Few-shot Complex Tabular Reasoning on Electronic Health Records
- An Edge-Cloud Collaboration Framework for Generative AI Service Provision with Synergetic Big Cloud Model and Small Edge Models
- LLMind: Orchestrating AI and IoT with LLM for Complex Task Execution
- Mamba: Linear-Time Sequence Modeling with Selective State Spaces
- A Language Agent for Autonomous Driving
- EdgeFM: Leveraging Foundation Model for Open-set Learning on the Edge
- A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions
- ADaPT: As-Needed Decomposition and Planning with Language Models
- Penetrative AI: Making LLMs Comprehend the Physical World
- Creating Trustworthy LLMs: Dealing with Hallucinations in Healthcare AI
- Drive as You Speak: Enabling Human-Like Interaction with Large Language Models in Autonomous Vehicles
- Plug in the Safety Chip: Enforcing Constraints for LLM-driven Robot Agents
- Prompt to Transfer: Sim-to-Real Transfer for Traffic Signal Control with Prompt Learning
- Pre-Trained Large Language Models for Industrial Control
- Llama 2: Open Foundation and Fine-Tuned Chat Models
- FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning
- Direct Preference Optimization: Your Language Model is Secretly a Reward Model
- GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints
- MemoryBank: Enhancing Large Language Models with Long-Term Memory
- Distilling Script Knowledge from Large Language Models for Constrained Language Planning
- Long-term Forecasting with TiDE: Time-series Dense Encoder
- Program of Thoughts Prompting: Disentangling Computation from Reasoning for Numerical Reasoning Tasks
- Diffusion Models: A Comprehensive Survey of Methods and Applications
- Inner Monologue: Embodied Reasoning through Planning with Language Models
- Significance of machine learning in healthcare: Features, pillars and applications
- FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness
- Are Transformers Effective for Time Series Forecasting?
- A Comprehensive Survey of Few-shot Learning: Evolution, Applications, Challenges, and Opportunities
- SecureBERT: A Domain-Specific Language Model for Cybersecurity
- On the Opportunities and Risks of Foundation Models
- Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing
- Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing
- Machine Learning in Agriculture: A Comprehensive Updated Review
- Empowering Things with Intelligence: A Survey of the Progress, Challenges, and Opportunities in Artificial Intelligence of Things
- Efficient Transformers: A Survey
- Language Models are Few-Shot Learners
- Fast Transformer Decoding: One Write-Head is All You Need
- DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter
- Axial Attention in Multidimensional Transformers
- Gaussian YOLOv3: An Accurate and Fast Object Detector Using Localization\n Uncertainty for Autonomous Driving
- Parameter-Efficient Transfer Learning for NLP
- Fast Scene Understanding for Autonomous Driving
- Attention Is All You Need
- Neural Machine Translation of Rare Words with Subword Units
- OmniDrive: A Holistic Vision-Language Dataset for Autonomous Driving with Counterfactual Reasoning
- Machine Learning Applications for Precision Agriculture: A Comprehensive Review
- BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Cited by
Related