Agentic Artificial Intelligence (AI): Architectures, Taxonomies, and Evaluation of Large Language Model Agents

TL;DR

The paper addresses how to design and evaluate agentic AI built on LLMs, proposing a unified six-dimension taxonomy and a three-layer architecture that ties perception, memory, planning, tool use, and profiling to robust multi-agent collaboration. It formalizes the agentic control loop within a POMDP-like framework and analyzes single versus multi-agent configurations, graph-based orchestration, and flow engineering. The authors also discuss diverse environments and applications, from digital Web/OS to embodied robotics and healthcare, paired with evaluation and safety frameworks (CLASSic) and enterprise benchmarks. The work highlights open challenges—hallucination in action, infinite loops, latency, and security—and outlines open directions toward open-ended learning, open governance, and robust, verifiable autonomous systems with practical deployment considerations.

Abstract

Artificial Intelligence is moving from models that only generate text to Agentic AI, where systems behave as autonomous entities that can perceive, reason, plan, and act. Large Language Models (LLMs) are no longer used only as passive knowledge engines but as cognitive controllers that combine memory, tool use, and feedback from their environment to pursue extended goals. This shift already supports the automation of complex workflows in software engineering, scientific discovery, and web navigation, yet the variety of emerging designs, from simple single loop agents to hierarchical multi agent systems, makes the landscape hard to navigate. In this paper, we investigate architectures and propose a unified taxonomy that breaks agents into Perception, Brain, Planning, Action, Tool Use, and Collaboration. We use this lens to describe the move from linear reasoning procedures to native inference time reasoning models, and the transition from fixed API calls to open standards like the Model Context Protocol (MCP) and Native Computer Use. We also group the environments in which these agents operate, including digital operating systems, embodied robotics, and other specialized domains, and we review current evaluation practices. Finally, we highlight open challenges, such as hallucination in action, infinite loops, and prompt injection, and outline future research directions toward more robust and reliable autonomous systems.

Figures (4)

• Figure 1: Taxonomy of the Agentic AI ecosystem. The figure organizes the literature into six main dimensions: Core Components, Cognitive Architecture, Learning, Multi Agent Systems, Environments, and Evaluation. Together, these dimensions trace the field's progression from simple text based loops to complex hierarchical systems that can operate in open ended environments.
• Figure 2: The unified architecture of Agentic AI. The system is shown as a modified POMDP loop. The agent brain at the center transforms each observation ($O_t$) into a reasoning trace ($Z_t$) using hierarchical planning and self reflection. A dual stream memory module at the top supports context retrieval, while a tool library at the bottom executes code based actions ($A_t$) that change the external environment on the right.
• Figure 3: Communication Topologies in Multi-Agent Systems. We classify collaboration patterns into three dominant structures: (Left) Chain Topology, utilized by MetaGPT to enforce Standard Operating Procedures (SOPs) via sequential hand-offs; (Center) Star Topology, employed by AutoGen where a Controller agent dispatches tasks to specialized workers; and (Right) Mesh Topology, used in social simulations like Generative Agents to enable dynamic, unstructured interaction.
• Figure 4: Multidimensional Architectural Comparison. We compare architectures across the CLASSic dimensions. While Hierarchical Agents (Red) achieve superior Reasoning Depth and Tool Proficiency, they incur a significant Cost Penalty and Latency compared to standard LLMs (Grey).