Toward Human-Aligned, Modular, and Emotion-Aware Multi-Agent Systems: A Systematic Literature Review
Abstract
Abstract Recent advances in agentic AI have accelerated a shift from monolithic language models toward modular, multi-agent systems enriched with memory, orchestration, and affect-aware interaction. Large Language Models now serve as reasoning and communication backbones, supported by architectures that integrate retrieval-augmented generation, neuro-symbolic reasoning, and alignment strategies. This paper presents a systematic literature review of multi-agent, LLM-based systems with a specific focus on long-term memory, emotion-aware dialogue, and human--AI alignment. Following a PRISMA-inspired methodology over a curated set of studies published between 2022 and 2025, the review synthesises four strands of work: multi-agent architectures, memory and retrieval mechanisms, emotion-aware conversational agents, and alignment frameworks. Building on and complementing recent surveys of agentic AI, we propose a taxonomy that classifies systems along four dimensions: orchestration structure, memory design, affect modelling, and alignment strategy. The analysis reveals that current frameworks strongly emphasise modularity and task performance, while long-term affective adaptation, relational alignment, and rigorous evaluation remain underexplored. Finally, we outline design principles and a research agenda for ``digital humans'': future multi-agent systems that integrate persistent memory, empathy, and alignment as first-class orchestration concerns rather than optional add-ons.
Read the paper
Related studies
- Knowledge Management Perspective of Generative Artificial Intelligence
- Navigating Generative Artificial Intelligence Promises and Perils for Knowledge and Creative Work
- Characterizing the evolving research on enterprise content management
- Large language models and multimodal foundation models for precision oncology
- From Generative AI to Extended Reality: Multidisciplinary Perspectives on the Challenges, Opportunities, and Future of Educational Computing
- Generative Artificial Intelligence: A New Engine for Advancing Environmental Science and Engineering
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.