Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

UserCentrix: An Agentic Memory-augmented AI Framework for Smart Spaces

Alaa Saleh, Sasu Tarkoma, Praveen Kumar Donta, Naser Hossein Motlagh, Schahram Dustdar, Susanna Pirttikangas, Lauri Lovén

TL;DR

UserCentrix presents a memory-augmented agentic AI framework for smart spaces that unifies personalized user-side LLM agents with a memory-enabled, hierarchical building-side control system. It employs VoI-driven decision-making, meta-reasoning, cooperative agent negotiation, and environment-tracking to optimize real-time responsiveness and resource use across edge-cloud continuum. The framework is validated on the University of Oulu Smart Campus dataset with multiple LLMs, demonstrating improvements in response accuracy, efficiency, and adaptability, while revealing trade-offs among latency, memory, and computational overhead. The work advances practical AI-driven smart environments by integrating memory augmentation and cooperative reasoning, with future work aimed at broader domains and richer user feedback loops.

Abstract

Agentic AI, with its autonomous and proactive decision-making, has transformed smart environments. By integrating Generative AI (GenAI) and multi-agent systems, modern AI frameworks can dynamically adapt to user preferences, optimize data management, and improve resource allocation. This paper introduces UserCentrix, an agentic memory-augmented AI framework designed to enhance smart spaces through dynamic, context-aware decision-making. This framework integrates personalized Large Language Model (LLM) agents that leverage user preferences and LLM memory management to deliver proactive and adaptive assistance. Furthermore, it incorporates a hybrid hierarchical control system, balancing centralized and distributed processing to optimize real-time responsiveness while maintaining global situational awareness. UserCentrix achieves resource-efficient AI interactions by embedding memory-augmented reasoning, cooperative agent negotiation, and adaptive orchestration strategies. Our key contributions include (i) a self-organizing framework with proactive scaling based on task urgency, (ii) a Value of Information (VoI)-driven decision-making process, (iii) a meta-reasoning personal LLM agent, and (iv) an intelligent multi-agent coordination system for seamless environment adaptation. Experimental results across various models confirm the effectiveness of our approach in enhancing response accuracy, system efficiency, and computational resource management in real-world application.

UserCentrix: An Agentic Memory-augmented AI Framework for Smart Spaces

TL;DR

UserCentrix presents a memory-augmented agentic AI framework for smart spaces that unifies personalized user-side LLM agents with a memory-enabled, hierarchical building-side control system. It employs VoI-driven decision-making, meta-reasoning, cooperative agent negotiation, and environment-tracking to optimize real-time responsiveness and resource use across edge-cloud continuum. The framework is validated on the University of Oulu Smart Campus dataset with multiple LLMs, demonstrating improvements in response accuracy, efficiency, and adaptability, while revealing trade-offs among latency, memory, and computational overhead. The work advances practical AI-driven smart environments by integrating memory augmentation and cooperative reasoning, with future work aimed at broader domains and richer user feedback loops.

Abstract

Agentic AI, with its autonomous and proactive decision-making, has transformed smart environments. By integrating Generative AI (GenAI) and multi-agent systems, modern AI frameworks can dynamically adapt to user preferences, optimize data management, and improve resource allocation. This paper introduces UserCentrix, an agentic memory-augmented AI framework designed to enhance smart spaces through dynamic, context-aware decision-making. This framework integrates personalized Large Language Model (LLM) agents that leverage user preferences and LLM memory management to deliver proactive and adaptive assistance. Furthermore, it incorporates a hybrid hierarchical control system, balancing centralized and distributed processing to optimize real-time responsiveness while maintaining global situational awareness. UserCentrix achieves resource-efficient AI interactions by embedding memory-augmented reasoning, cooperative agent negotiation, and adaptive orchestration strategies. Our key contributions include (i) a self-organizing framework with proactive scaling based on task urgency, (ii) a Value of Information (VoI)-driven decision-making process, (iii) a meta-reasoning personal LLM agent, and (iv) an intelligent multi-agent coordination system for seamless environment adaptation. Experimental results across various models confirm the effectiveness of our approach in enhancing response accuracy, system efficiency, and computational resource management in real-world application.
Paper Structure (24 sections, 4 equations, 13 figures, 7 tables, 2 algorithms)

This paper contains 24 sections, 4 equations, 13 figures, 7 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Related Works
  3. LLMs for edge-cloud continuum
  4. Structures design of multi-agent systems
  5. Reasoning LLM agent
  6. UserCentrix Framework
  7. User Side
  8. Smart Building Side
  9. Decision-making Module:
  10. Sub-tasks Execution Module
  11. Management and Analysis Module
  12. Use Case
  13. User Task Processing Scenario:
  14. High-urgency Scenario:
  15. Low-urgency Scenario:
  16. ...and 9 more sections

Figures (13)

  • Figure 1: UserCentrix Framework.
  • Figure 2: User Task Processing within UserCentrix Framework.
  • Figure 3: High-urgency Workflow within UserCentrix Framework.
  • Figure 4: Low-urgency Workflow within UserCentrix Framework.
  • Figure 5: Personal Agent Performance Evaluation.
  • ...and 8 more figures