Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG
2025/01/15 by Aditi Singh, Abul Ehtesham, Singh, Aditi +6 · 3 voices · 119 citations
Computer Science · #Artificial Intelligence (cs.AI) #Computation and Language (cs.CL) #FOS: Computer and information sciences #Information Retrieval (cs.IR) #Multi-Agent Systems and Negotiation #Reinforcement Learning in Robotics #Speech and dialogue systems
paper · pdf · doi:10.48550/arxiv.2501.09136
openalex publication_date 2025/01/15 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Abstract
Large Language Models (LLMs) have advanced artificial intelligence by enabling human-like text generation and natural language understanding. However, their reliance on static training data limits their ability to respond to dynamic, real-time queries, resulting in outdated or inaccurate outputs. Retrieval-Augmented Generation (RAG) has emerged as a solution, enhancing LLMs by integrating real-time data retrieval to provide contextually relevant and up-to-date responses. Despite its promise, traditional RAG systems are constrained by static workflows and lack the adaptability required for multi-step reasoning and complex task management. Agentic Retrieval-Augmented Generation (Agentic RAG) transcends these limitations by embedding autonomous AI agents into the RAG pipeline. These agents leverage agentic design patterns reflection, planning, tool use, and multi-agent collaboration to dynamically manage retrieval strategies, iteratively refine contextual understanding, and adapt workflows through operational structures ranging from sequential steps to adaptive collaboration. This integration enables Agentic RAG systems to deliver flexibility, scalability, and context-awareness across diverse applications. This paper presents an analytical survey of Agentic RAG systems. It traces the evolution of RAG paradigms, introduces a principled taxonomy of Agentic RAG architectures based on agent cardinality, control structure, autonomy, and knowledge representation, and provides a comparative analysis of design trade-offs across existing frameworks. The survey examines applications in healthcare, finance, education, and enterprise document processing, and distills practical lessons for system designers and practitioners. Finally, it identifies key open research challenges related to evaluation, coordination, memory management, efficiency, and governance, outlining directions for future research.
Cited by
- ACM: Agentic Context Management for Long Horizon Tasks
- A New Role for Relevance: Guiding Corpus Interaction in Agentic Search
- Robust Interpretation of Historical Documents in Knowledge Graphs Through Query Inference and Execution
- Co-Evolving Graph and Text Memory for Training-Free Multi-Hop Question Answering
- EviBack: Search-Agent Reinforcement Learning via Evidence-Constrained Teacher Backoff
- PANOPTICON: A PII-Based Assemblage of Naturalistic Output Tokens for Investigating Privacy Leakage Within LLM Context Window
- From Naive RAG to Deep Agentic Retrieval: An Evolving Context Engineering Pipeline for Regulatory Compliance
- Agentic Graph Retrieval-Augmented Generation for Auditable Commercial Registry Analysis
- X-GridAgent: An LLM-Powered Agentic AI System for Assisting Power Grid Analysis
- Leveraging Spreading Activation for Improved Document Retrieval in Knowledge-Graph-Based RAG Systems
- VLegal-Bench: Cognitively Grounded Benchmark for Vietnamese Legal Reasoning of Large Language Models
- AgroAskAI: A Multi-Agentic AI Framework for Supporting Smallholder Farmers' Enquiries Globally
- VERAFI: Verified Agentic Financial Intelligence through Neurosymbolic Policy Generation
- Rethinking Retrieval: From Traditional Retrieval Augmented Generation to Agentic and Non-Vector Reasoning Systems in the Financial Domain for Large Language Models
- MuISQA: Multi-Intent Retrieval-Augmented Generation for Scientific Question Answering
- Grounded by Experience: Generative Healthcare Prediction Augmented with Hierarchical Agentic Retrieval
- LiCoMemory: Lightweight and Cognitive Agentic Memory for Efficient Long-Term Reasoning
- MARC: Multimodal and Multi-Task Agentic Retrieval-Augmented Generation for Cold-Start Recommender System
- Q-RAG: Long Context Multi-step Retrieval via Value-based Embedder Training
- Rethinking Retrieval-Augmented Generation for Medicine: A Large-Scale, Systematic Expert Evaluation and Practical Insights
- Reinforcement Learning Improves Traversal of Parametric Knowledge in LLMs
- MicroRemed: Benchmarking LLMs in Microservices Remediation
- A Hierarchical Multi-Agent System for Autonomous Discovery in Geoscientific Data Archives
- CompAgent: An Agentic Framework for Visual Compliance Verification
- Interact-RAG: Reason and Interact with the Corpus, Beyond Black-Box Retrieval
- Retrieval Augmented Generation (RAG) for Fintech: Agentic Design and Evaluation
- Is GraphRAG Needed? From Basic RAG to Graph-/Agentic Solutions with Context Optimization
- Optimizing Retrieval for RAG via Reinforced Contrastive Learning
- Metadata-Driven Retrieval-Augmented Generation for Financial Question Answering
- Hierarchical Sequence Iteration for Heterogeneous Question Answering
- Enterprise Deep Research: Steerable Multi-Agent Deep Research for Enterprise Analytics
- Empowering Real-World: A Survey on the Technology, Practice, and Evaluation of LLM-driven Industry Agents
- A Comprehensive Survey on Reinforcement Learning-based Agentic Search: Foundations, Roles, Optimizations, Evaluations, and Applications
- Scaling Beyond Context: A Survey of Multimodal Retrieval-Augmented Generation for Document Understanding
- MoM: Mixtures of Scenario-Aware Document Memories for Retrieval-Augmented Generation Systems
- Disaster Management in the Era of Agentic AI Systems: A Vision for Collective Human-Machine Intelligence for Augmented Resilience
- Beyond Correctness: Rewarding Faithful Reasoning in Retrieval-Augmented Generation
- RAGCap-Bench: Benchmarking Capabilities of LLMs in Agentic Retrieval Augmented Generation Systems
- Deep Research Brings Deeper Harm
- AccurateRAG: A Framework for Building Accurate Retrieval-Augmented Question-Answering Applications
- Agentic Systems in Radiology: Design, Applications, Evaluation, and Challenges
- When Retrieval Succeeds and Fails: Rethinking Retrieval-Augmented Generation for LLMs
- Exploiting Web Search Tools of AI Agents for Data Exfiltration
- SEER: Sustainability Enhanced Engineering of Software Requirements
- HiPRAG: Hierarchical Process Rewards for Efficient Agentic Retrieval Augmented Generation
- MAPRO: Recasting Multi-Agent Prompt Optimization as Maximum a Posteriori Inference
- AgentRouter: A Knowledge-Graph-Guided LLM Router for Collaborative Multi-Agent Question Answering
- ModernBERT + ColBERT: Enhancing biomedical RAG through an advanced re-ranking retriever
- Demystifying deep search: a holistic evaluation with hint-free multi-hop questions and factorised metrics
- FreeRet: MLLMs as Training-Free Retrievers
- SafeSearch: Automated Red-Teaming for the Safety of LLM-Based Search Agents
- The Context Access Divide: Interaction-Level Architecture as a Complementary Dimension of Agentic Inequality
- Dr-DCI: Scaling Direct Corpus Interaction via Dynamic Workspace Expansion
- Retrieval from Within: An Intrinsic Capability of Attention-Based Models
- AtomRAG: atomic query decomposition for multi-hop retrieval-augmented generation
- Large language model-driven agents in nursing practice: A scoping review
- LLM-based Agents Suffer from Hallucinations: A Survey of Taxonomy, Methods, and Directions
- RALLM-POI: Retrieval-Augmented LLM for Zero-shot Next POI Recommendation with Geographical Reranking
- EHR-MCP: Real-world Evaluation of Clinical Information Retrieval by Large Language Models via Model Context Protocol
- Less Is More: Elevating RAG via Performance-Driven Context Compression
- CC-GSEO-Bench: A Content-Centric Benchmark for Measuring Source Influence in Generative Search Engines
- MultiFluxAI Enhancing Platform Engineering with Advanced Agent-Orchestrated Retrieval Systems
- End-to-End Agentic RAG System Training for Traceable Diagnostic Reasoning
- Can Compact Language Models Search Like Agents? Distillation-Guided Policy Optimization for Preserving Agentic RAG Capabilities
- Context-Adaptive Synthesis and Compression for Enhanced Retrieval-Augmented Generation in Complex Domains
- Test-time Corpus Feedback: From Retrieval to RAG
- Atom-Searcher: Enhancing Agentic Deep Research via Fine-Grained Atomic Thought Reward
- Structuring the Unstructured: A Systematic Review of Text-to-Structure Generation for Agentic AI with a Universal Evaluation Framework
- Retrieval-Augmented Generation in Industry: An Interview Study on Use Cases, Requirements, Challenges, and Evaluation
- Classification is a RAG problem: A case study on hate speech detection
- WARP -- Web-Augmented Real-time Program Repairer: A Real-Time Compilation Error Resolution using LLMs and Web-Augmented Synthesis
- RAVine: Reality-Aligned Evaluation for Agentic Search
- Patho-AgenticRAG: Towards Multimodal Agentic Retrieval-Augmented Generation for Pathology VLMs via Reinforcement Learning
- A Survey on Agent Workflow -- Status and Future
- Open-Source Agentic Hybrid RAG Framework for Scientific Literature Review
- MemTool: Optimizing Short-Term Memory Management for Dynamic Tool Calling in LLM Agent Multi-Turn Conversations
- Advancing Shared and Multi-Agent Autonomy in Underwater Missions: Integrating Knowledge Graphs and Retrieval-Augmented Generation
- A Systematic Review of Key Retrieval-Augmented Generation (RAG) Systems: Progress, Gaps, and Future Directions
- Agentic Satellite-Augmented Low-Altitude Economy and Terrestrial Networks: A Survey on Generative Approaches
- Confident RAG: Enhancing the Performance of LLMs for Mathematics Question Answering through Multi-Embedding and Confidence Scoring
- Agentic RAG with Knowledge Graphs for Complex Multi-Hop Reasoning in Real-World Applications
- GitChameleon 2.0: Evaluating AI Code Generation Against Python Library Version Incompatibilities
- LLM-Based Config Synthesis requires Disambiguation
- HKGAI-V1: Towards Regional Sovereign Large Language Model for Hong Kong
- Am I on the Right Track? What Can Predicted Query Performance Tell Us about the Search Behaviour of Agentic RAG
- Improving Korean-English Cross-Lingual Retrieval: A Data-Centric Study of Language Composition and Model Merging
- BioMARS: A Multi-Agent Robotic System for Autonomous Biological Experiments
- CAL-RAG: Retrieval-Augmented Multi-Agent Generation for Content-Aware Layout Design
- ZKPROV: A Zero-Knowledge Approach to Dataset Provenance for Large Language Models
- Engineering RAG Systems for Real-World Applications: Design, Development, and Evaluation
- Fine-Tuning and Prompt Engineering of LLMs, for the Creation of Multi-Agent AI for Addressing Sustainable Protein Production Challenges
- From Web Search towards Agentic Deep Research: Incentivizing Search with Reasoning Agents
- Advanced Applications of Generative AI in Actuarial Science: Case Studies Beyond ChatGPT
- Deep Research Agents: A Systematic Examination And Roadmap
- From Multimodal Perception to Strategic Reasoning: A Survey on AI-Generated Game Commentary
- RAG+: Enhancing Retrieval-Augmented Generation with Application-Aware Reasoning
- CIIR@LiveRAG 2025: Optimizing Multi-Agent Retrieval Augmented Generation through Self-Training
- AnnaAgent: Dynamic Evolution Agent System with Multi-Session Memory for Realistic Seeker Simulation
- Inter-Passage Verification for Multi-evidence Multi-answer QA
- From Large AI Models to Agentic AI: A Tutorial on Future Intelligent Communications
- LLM Web Dynamics: Tracing Model Collapse in a Network of LLMs
- Get Experience from Practice: LLM Agents with Record & Replay
- Cosmos: A CXL-Based Full In-Memory System for Approximate Nearest Neighbor Search
- InfoDeepSeek: Benchmarking Agentic Information Seeking for Retrieval-Augmented Generation
- CRAKEN: Cybersecurity LLM Agent with Knowledge-Based Execution
- StepSearch: Igniting LLMs Search Ability via Step-Wise Proximal Policy Optimization
- An Empirical Study on Reinforcement Learning for Reasoning-Search Interleaved LLM Agents
- From Cloud to Crowd: Democratizing LLM Service with Decentralized Edge Collaboration for RAG
- Process vs. Outcome Reward: Which is Better for Agentic RAG Reinforcement Learning
- Demystifying and Enhancing the Efficiency of Large Language Model Based Search Agents
- AI Agents vs. Agentic AI: A Conceptual Taxonomy, Applications and Challenges
- Trustless Autonomy: Understanding Motivations, Benefits, and Governance Dilemmas in Self-Sovereign Decentralized AI Agents
- Internet of Agents: Fundamentals, Applications, and Challenges
- ScaleMCP: Dynamic and Auto-Synchronizing Model Context Protocol Tools for LLM Agents
- Chatbots Output Meaningful (but Problematic) Language
- From LLM Reasoning to Autonomous AI Agents: A Comprehensive Review
- Privacy-Preserving Federated Embedding Learning for Localized Retrieval-Augmented Generation
- A RAG-Based Multi-Agent LLM System for Natural Hazard Resilience and Adaptation
- ARCeR: an Agentic RAG for the Automated Definition of Cyber Ranges
- A Survey of Personalization: From RAG to Agent
- Bridging Industrial Expertise and XR with LLM-Powered Conversational Agents
Discussions
- Sorry, serious question: is this paper real? arxiv.org/abs/2501.09136 It's got 24 cites, but does the tech. even exist? How can vaporware be surveyed? The only citation in its bibliography with "agent [bsky, 7 points, 3 comments]
- 🧠 Avevo già fatto delle sperimentazioni unendo il concetto di #RAG a un sistema multi-agent, e questo paper ne definisce proprio il paradigma, con il termine "Agentic RAG".
🔗 Il paper: arxiv.org/a [bsky, 1 points, 1 comments]
- Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG arxiv.org/abs/2501.09136 [bsky, 0 points, 0 comments]
Related