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

• URL: http://arxiv.org/abs/2603.07379v1
• Date: Sat, 07 Mar 2026 23:38:57 GMT
• Title: SoK: Agentic Retrieval-Augmented Generation (RAG): Taxonomy, Architectures, Evaluation, and Research Directions
• Abstract summary: This paper provides the first unified framework for understanding these autonomous systems.<n>We formalize agentic retrieval-generation loops as finite-horizon partially observable Markov decision processes.<n>We identify severe systemic risks inherent to autonomous loops, including compounding hallucination propagation, memory poisoning, retrieval misalignment, and cascading tool-execution vulnerabilities.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• 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.

Related papers

• From Holistic Evaluation to Structured Criteria: Rubrics Across the Evolving LLM Landscape [79.30826980815927]
  rubrics are explicit criteria sets that transform complex quality judgments into structured and actionable standards.<n>We systematically organize existing rubric designs, examine their construction and optimization, and analyze their role across evaluation and training.
  arXiv  Detail & Related papers  (2026-06-07T13:34:55Z)
• COMPASS: The explainable agentic framework for Sovereignty, Sustainability, Compliance, and Ethics [0.0]
  This paper introduces a novel multi-agent orchestration system designed to enforce value-aligned AI.<n>The framework comprises an Orchestrator and four specialised sub-agents addressing sovereignty, carbon-aware computing, compliance, and ethics.<n>By employing an LLM-as-a-judge methodology, the system assigns quantitative scores and generates explainable justifications for each assessment dimension.
  arXiv  Detail & Related papers  (2026-03-11T20:14:35Z)
• A Survey of Reasoning in Autonomous Driving Systems: Open Challenges and Emerging Paradigms [49.66022971508878]
  We argue that reasoning should be elevated from a modular component to the system's cognitive core.<n>We derive and systematize seven core reasoning challenges, such as the responsiveness-reasoning trade-off and social-game reasoning.<n>We identify a fundamental and unresolved tension between the high-latency, deliberative nature of LLM-based reasoning and the millisecond-scale, safety-critical demands of vehicle control.
  arXiv  Detail & Related papers  (2026-03-11T07:40:53Z)
• Agentic AI for Cybersecurity: A Meta-Cognitive Architecture for Governable Autonomy [0.0]
  This paper argues that cybersecurity orchestration should be reconceptualized as an agentic, multi-agent cognitive system.<n>We introduce a conceptual framework in which heterogeneous AI agents responsible for detection, hypothesis formation, contextual interpretation, explanation, and governance are coordinated through an explicit meta-cognitive judgement function.<n>Our contribution is to make this cognitive structure architecturally explicit and governable by embedding meta-cognitive judgement as a first-class system function.
  arXiv  Detail & Related papers  (2026-02-12T12:52:49Z)
• Multi-hop Reasoning via Early Knowledge Alignment [68.28168992785896]
  Early Knowledge Alignment (EKA) aims to align Large Language Models with contextually relevant retrieved knowledge.<n>EKA significantly improves retrieval precision, reduces cascading errors, and enhances both performance and efficiency.<n>EKA proves effective as a versatile, training-free inference strategy that scales seamlessly to large models.
  arXiv  Detail & Related papers  (2025-12-23T08:14:44Z)
• From Educational Analytics to AI Governance: Transferable Lessons from Complex Systems Interventions [0.0]
  We argue that five core principles developed within CAPIRE transfer directly to the challenge of governing AI systems.<n>The isomorphism is not merely analogical: both domains exhibit non-linearity, emergence, feedback loops, strategic adaptation, and path dependence.<n>We propose Complex Systems AI Governance (CSAIG) as an integrated framework that operationalises these principles for regulatory design.
  arXiv  Detail & Related papers  (2025-12-15T12:16:57Z)
• Agentic AI: A Comprehensive Survey of Architectures, Applications, and Future Directions [10.453339156813852]
  Agentic AI represents a transformative shift in artificial intelligence.<n>Its rapid advancement has led to a fragmented understanding, often conflating modern neural systems with outdated symbolic models.<n>This survey introduces a novel dual-paradigm framework that categorizes agentic systems into two distinct lineages.
  arXiv  Detail & Related papers  (2025-10-29T12:11:34Z)
• PISA: A Pragmatic Psych-Inspired Unified Memory System for Enhanced AI Agency [50.712873697511206]
  Existing work often lacks adaptability to diverse tasks and overlooks the constructive and task-oriented role of AI agent memory.<n>We propose PISA, a pragmatic, psych-inspired unified memory system that treats memory as a constructive and adaptive process.<n>Our empirical evaluation, conducted on the existing LOCOMO benchmark and our newly proposed AggQA benchmark for data analysis tasks, confirms that PISA sets a new state-of-the-art by significantly enhancing adaptability and long-term knowledge retention.
  arXiv  Detail & Related papers  (2025-10-12T10:34:35Z)
• STARec: An Efficient Agent Framework for Recommender Systems via Autonomous Deliberate Reasoning [54.28691219536054]
  We introduce STARec, a slow-thinking augmented agent framework that endows recommender systems with autonomous deliberative reasoning capabilities.<n>We develop anchored reinforcement training - a two-stage paradigm combining structured knowledge distillation from advanced reasoning models with preference-aligned reward shaping.<n>Experiments on MovieLens 1M and Amazon CDs benchmarks demonstrate that STARec achieves substantial performance gains compared with state-of-the-art baselines.
  arXiv  Detail & Related papers  (2025-08-26T08:47:58Z)
• A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems [53.37728204835912]
  Most existing AI systems rely on manually crafted configurations that remain static after deployment.<n>Recent research has explored agent evolution techniques that aim to automatically enhance agent systems based on interaction data and environmental feedback.<n>This survey aims to provide researchers and practitioners with a systematic understanding of self-evolving AI agents.
  arXiv  Detail & Related papers  (2025-08-10T16:07:32Z)
• Deep Research Agents: A Systematic Examination And Roadmap [109.53237992384872]
  Deep Research (DR) agents are designed to tackle complex, multi-turn informational research tasks.<n>In this paper, we conduct a detailed analysis of the foundational technologies and architectural components that constitute DR agents.
  arXiv  Detail & Related papers  (2025-06-22T16:52:48Z)
• Beyond Accidents and Misuse: Decoding the Structural Risk Dynamics of Artificial Intelligence [0.0]
  This paper advances the concept of structural risk by introducing a framework grounded in complex systems research.<n>We classify structural risks into three categories: antecedent structural causes, antecedent AI system causes, and deleterious feedback loops.<n>To anticipate and govern these dynamics, the paper proposes a methodological agenda incorporating scenario mapping, simulation, and exploratory foresight.
  arXiv  Detail & Related papers  (2024-06-21T05:44:50Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.