Adaptive and Resource-efficient Agentic AI Systems for Mobile and Embedded Devices: A Survey

2025-09-29

AI Summary

To address the fundamental tension between resource constraints (memory, energy, bandwidth, latency) and system requirements (long-term adaptability, real-time interaction) in deploying foundation model (FM)-based agents on mobile and embedded devices, this paper proposes the first unified co-optimization framework tailored for edge-side agents. Our method integrates elastic inference, test-time adaptation, dynamic multimodal fusion, edge-coordinated deployment, and distribution-shift-robust optimization—establishing interpretable mappings among FM architecture, cognitive capabilities, and hardware resources. Through algorithm-system co-design, we achieve Pareto-optimal trade-offs among accuracy, latency, and communication overhead. We systematically analyze the intelligent evolution of the perception-decision-execution loop under resource constraints, identifying key challenges and future research directions. This work provides both theoretical foundations and practical paradigms for scalable, adaptive edge AI agents.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Multiagent Systems: Other Foundations of Multi Agent Systems
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Foundation models have reshaped AI by unifying fragmented architectures into scalable backbones with multimodal reasoning and contextual adaptation. In parallel, the long-standing notion of AI agents, defined by the sensing-decision-action loop, is entering a new paradigm: with FMs as their cognitive core, agents transcend rule-based behaviors to achieve autonomy, generalization, and self-reflection. This dual shift is reinforced by real-world demands such as autonomous driving, robotics, virtual assistants, and GUI agents, as well as ecosystem advances in embedded hardware, edge computing, mobile deployment platforms, and communication protocols that together enable large-scale deployment. Yet this convergence collides with reality: while applications demand long-term adaptability and real-time interaction, mobile and edge deployments remain constrained by memory, energy, bandwidth, and latency. This creates a fundamental tension between the growing complexity of FMs and the limited resources of deployment environments. This survey provides the first systematic characterization of adaptive, resource-efficient agentic AI systems. We summarize enabling techniques into elastic inference, test-time adaptation, dynamic multimodal integration, and agentic AI applications, and identify open challenges in balancing accuracy-latency-communication trade-offs and sustaining robustness under distribution shifts. We further highlight future opportunities in algorithm-system co-design, cognitive adaptation, and collaborative edge deployment. By mapping FM structures, cognition, and hardware resources, this work establishes a unified perspective toward scalable, adaptive, and resource-efficient agentic AI. We believe this survey can help readers to understand the connections between enabling technologies while promoting further discussions on the fusion of agentic intelligence and intelligent agents.

Problem

• Adapting large AI models to resource-limited mobile devices
• Balancing computational complexity with energy and latency constraints
• Enabling autonomous AI agents on embedded systems efficiently

Innovation

• Elastic inference adapts to resource constraints
• Test-time adaptation enables real-time learning
• Dynamic multimodal integration enhances contextual reasoning