SoK: Agentic Retrieval-Augmented Generation (RAG): Taxonomy, Architectures, Evaluation, and Research Directions

TL;DR

This paper systematically analyzes agentic Retrieval-Augmented Generation systems, formalizing their decision processes, categorizing architectures, evaluating current practices, and outlining future research directions for reliable and scalable autonomous language models.

Contribution

It provides the first unified formal framework, taxonomy, and architectural decomposition for agentic RAG systems, addressing evaluation challenges and systemic risks.

Findings

Formalization of agentic RAG as finite-horizon POMDPs

Development of a comprehensive taxonomy and modular architecture

Identification of systemic risks like hallucinations and memory poisoning

Abstract

Retrieval-Augmented Generation (RAG) systems are increasingly evolving into agentic architectures where large language models autonomously coordinate multi-step reasoning, dynamic memory management, and iterative retrieval strategies. Despite rapid industrial adoption, current research lacks a systematic understanding of Agentic RAG as a sequential decision-making system, leading to highly fragmented architectures, inconsistent evaluation methodologies, and unresolved reliability risks. This Systematization of Knowledge (SoK) paper provides the first unified framework for understanding these autonomous systems. We formalize agentic retrieval-generation loops as finite-horizon partially observable Markov decision processes, explicitly modeling their control policies and state transitions. Building upon this formalization, we develop a comprehensive taxonomy and modular architectural decomposition that categorizes systems by their planning mechanisms, retrieval orchestration, memory paradigms, and tool-invocation behaviors. We further analyze the critical limitations of traditional static evaluation practices and identify severe systemic risks inherent to autonomous loops, including compounding hallucination propagation, memory poisoning, retrieval misalignment, and cascading tool-execution vulnerabilities. Finally, we outline key doctoral-scale research directions spanning stable adaptive retrieval, cost-aware orchestration, formal trajectory evaluation, and oversight mechanisms, providing a definitive roadmap for building reliable, controllable, and scalable agentic retrieval systems.