Manual context attachment produces a combinatorial collapse in AI task-success probability as personal corpus size and multi-document conjunctivity grow, while dynamic retrieval is insulated from that collapse, creating an inequality dimension complementary to agentic access.
hub Canonical reference
Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG
Canonical reference. 91% of citing Pith papers cite this work as background.
69
Pith papers citing it
Background
91% of classified citations
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their reliance on static training data limits their ability to respond to dynamic, real-time queries, resulting in outdated or inaccurate outputs.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.
hub tools
citation-role summary
background 11
citation-polarity summary
roles
background 11representative citing papers
ChartWalker provides a hierarchical knowledge graph construction method and structure-aware sampling to generate cross-chart RAG benchmarks, releasing ChartWalker-Bench that exposes performance gaps across RAG paradigms.
LLM outputs are meaningful according to standard theories of human language, without requiring anthropomorphic assumptions about the models.
CuSearch reallocates rollout budget in RLVR toward deeper-search trajectories as a proxy for retrieval supervision density, yielding up to 11.8 exact-match gains over uniform GRPO sampling on ZeroSearch.
LatentRAG performs agentic RAG by generating latent tokens for thoughts and subqueries in one forward pass, matching explicit methods' accuracy on seven benchmarks while reducing latency by ~90%.
SCOUT achieves state-of-the-art long-text understanding with up to 8x lower token use by actively foraging for sparse query-relevant information and updating a compact provenance-grounded epistemic state.
RAG-Reflect achieves F1=0.78 on valid comment-edit prediction using retrieval-augmented reasoning and self-reflection, outperforming baselines and approaching fine-tuned models without retraining.
A-MAR decomposes art queries into reasoning plans to condition retrieval, leading to improved explanation quality and multi-step reasoning on art benchmarks compared to baselines.
Corpus2Skill converts document corpora into navigable hierarchical skill directories for LLM agents, improving QA and RAG quality on single-domain enterprise data but not on open-domain or tabular corpora.
E2E-REME outperforms nine LLMs in accuracy and efficiency for end-to-end microservice remediation by using experience-simulation reinforcement fine-tuning on a new benchmark called MicroRemed.
DotRAG reformulates graph retrieval as query-guided path reasoning with Division of Thought, reporting SOTA results on MetaQA and UltraDomain for multi-hop tasks.
Q-RAG trains embedders via RL for multi-step retrieval and reports state-of-the-art results on BabiLong and RULER benchmarks for contexts up to 10M tokens.
HiPRAG adds hierarchical process rewards to RL training for agentic RAG, reducing over-search to 2.3% and achieving 65.4-67.2% accuracy on seven QA benchmarks across 3B and 7B models.
KidnapRAG is a sequential black-box poisoning attack on Agentic RAG systems using Bait, Chain-Link, and Mal-Ins documents to redirect retrieval and reasoning, outperforming prior baselines.
MedEvoEval is an executable longitudinal evaluation framework that converts medical cases into action-gated simulated episodes to track how doctor agents evolve decision-making, resource use, and experience across multiple encounters.
Evaluates 9 RAG scenarios across variants, proposes context engineering reducing token usage 19-53%, and identifies a retrieval-generation gap where more retrieval does not improve generation proportionally.
Empirical study finds isolation drives gains for weak models in multi-agent RAG while scoring matters for strong ones, enabling MADARA for cost-efficient adaptive assessment.
GDP-RAG targets only information deltas in multi-hop RAG through preliminary grounding, gap-conditioned prompts, and skeletal trajectories, reaching 60.63% accuracy at 0.51 cost-of-pass on HotpotQA, 2WikiMultiHopQA, and MuSiQue.
SHACR is a graph-augmented framework that grounds LLMs in a formal knowledge graph to unify logical, semantic, and physical conflict detection in IoT automation, raising F1 from 0.59 to 0.95 on a 203-rule testbed.
EvoEmbedding generates evolvable embeddings via a latent memory updated during sequential processing, outperforming larger models on long-context retrieval and generalizing to 10x longer contexts in downstream tasks.
SHIFT mitigates language bias in MLIR by subtracting estimated relative language vectors from document embeddings during indexing using parallel translation pairs.
Introduces V-RAGBench benchmark and CARVE method that selects per-chunk retrieval configurations via parallel retrievers and adaptive reranking, outperforming eight VideoRAG baselines.
Agentic hybrid RAG with a new muon collider benchmark outperforms baselines in retrieval effectiveness, answer quality, evidence coverage, and factual grounding.
MARDoc introduces a three-agent framework (Explorer, Refiner, Reflector) with dynamically updated structured memory to improve multi-hop reasoning in multimodal long-document QA, outperforming baselines on MMLongBench-Doc and DocBench.
citing papers explorer
-
The Context Access Divide: Interaction-Level Architecture as a Complementary Dimension of Agentic Inequality
Manual context attachment produces a combinatorial collapse in AI task-success probability as personal corpus size and multi-document conjunctivity grow, while dynamic retrieval is insulated from that collapse, creating an inequality dimension complementary to agentic access.
-
ChartWalker: Benchmarking the Cross-Chart RAG Task with Hierarchical Knowledge Graphs
ChartWalker provides a hierarchical knowledge graph construction method and structure-aware sampling to generate cross-chart RAG benchmarks, releasing ChartWalker-Bench that exposes performance gaps across RAG paradigms.
-
Chatbots Output Meaningful (but Problematic) Language
LLM outputs are meaningful according to standard theories of human language, without requiring anthropomorphic assumptions about the models.
-
CuSearch: Curriculum Rollout Sampling via Search Depth for Agentic RAG
CuSearch reallocates rollout budget in RLVR toward deeper-search trajectories as a proxy for retrieval supervision density, yielding up to 11.8 exact-match gains over uniform GRPO sampling on ZeroSearch.
-
LatentRAG: Latent Reasoning and Retrieval for Efficient Agentic RAG
LatentRAG performs agentic RAG by generating latent tokens for thoughts and subqueries in one forward pass, matching explicit methods' accuracy on seven benchmarks while reducing latency by ~90%.
-
SCOUT: Active Information Foraging for Long-Text Understanding with Decoupled Epistemic States
SCOUT achieves state-of-the-art long-text understanding with up to 8x lower token use by actively foraging for sparse query-relevant information and updating a compact provenance-grounded epistemic state.
-
RAG-Reflect: Agentic Retrieval-Augmented Generation with Reflections for Comment-Driven Code Maintenance on Stack Overflow
RAG-Reflect achieves F1=0.78 on valid comment-edit prediction using retrieval-augmented reasoning and self-reflection, outperforming baselines and approaching fine-tuned models without retraining.
-
A-MAR: Agent-based Multimodal Art Retrieval for Fine-Grained Artwork Understanding
A-MAR decomposes art queries into reasoning plans to condition retrieval, leading to improved explanation quality and multi-step reasoning on art benchmarks compared to baselines.
-
Don't Retrieve, Navigate: Distilling Enterprise Knowledge into Navigable Agent Skills for QA and RAG
Corpus2Skill converts document corpora into navigable hierarchical skill directories for LLM agents, improving QA and RAG quality on single-domain enterprise data but not on open-domain or tabular corpora.
-
E2E-REME: Towards End-to-End Microservices Auto-Remediation via Experience-Simulation Reinforcement Fine-Tuning
E2E-REME outperforms nine LLMs in accuracy and efficiency for end-to-end microservice remediation by using experience-simulation reinforcement fine-tuning on a new benchmark called MicroRemed.
-
DOTRAG: Retrieval-Time Reasoning Along Paths
DotRAG reformulates graph retrieval as query-guided path reasoning with Division of Thought, reporting SOTA results on MetaQA and UltraDomain for multi-hop tasks.
-
Q-RAG: Long Context Multi-step Retrieval via Value-based Embedder Training
Q-RAG trains embedders via RL for multi-step retrieval and reports state-of-the-art results on BabiLong and RULER benchmarks for contexts up to 10M tokens.
-
HiPRAG: Hierarchical Process Rewards for Efficient Agentic Retrieval Augmented Generation
HiPRAG adds hierarchical process rewards to RL training for agentic RAG, reducing over-search to 2.3% and achieving 65.4-67.2% accuracy on seven QA benchmarks across 3B and 7B models.
-
KidnapRAG: A Black-Box Attack for Hijacking Reasoning in Agentic Retrieval-Augmented Generation Systems
KidnapRAG is a sequential black-box poisoning attack on Agentic RAG systems using Bait, Chain-Link, and Mal-Ins documents to redirect retrieval and reasoning, outperforming prior baselines.
-
MedEvoEval: Evaluating Continual Evolution of Doctor Agents through Simulated Clinical Episodes
MedEvoEval is an executable longitudinal evaluation framework that converts medical cases into action-gated simulated episodes to track how doctor agents evolve decision-making, resource use, and experience across multiple encounters.
-
Is GraphRAG Needed? From Basic RAG to Graph-/Agentic Solutions with Context Optimization
Evaluates 9 RAG scenarios across variants, proposes context engineering reducing token usage 19-53%, and identifies a retrieval-generation gap where more retrieval does not improve generation proportionally.
-
To Isolate or to Score? Model-Adaptive Assessment for Cost-Efficient Multi-Agent RAG
Empirical study finds isolation drives gains for weak models in multi-agent RAG while scoring matters for strong ones, enabling MADARA for cost-efficient adaptive assessment.
-
Only Ask What You Don't Know: Grounded Delta Planning for Efficient Multi-step RAG
GDP-RAG targets only information deltas in multi-hop RAG through preliminary grounding, gap-conditioned prompts, and skeletal trajectories, reaching 60.63% accuracy at 0.51 cost-of-pass on HotpotQA, 2WikiMultiHopQA, and MuSiQue.
-
SHACR: A Graph-Augmented Semi-Autonomous Framework for Multi-Class Conflict Resolution in Smart Home IoT Automation
SHACR is a graph-augmented framework that grounds LLMs in a formal knowledge graph to unify logical, semantic, and physical conflict detection in IoT automation, raising F1 from 0.59 to 0.95 on a 203-rule testbed.
-
EvoEmbedding: Evolvable Representations for Long-Context Retrieval and Agentic Memory
EvoEmbedding generates evolvable embeddings via a latent memory updated during sequential processing, outperforming larger models on long-context retrieval and generalizing to 10x longer contexts in downstream tasks.
-
SHIFT: Semantic Harmonization via Index-side Feature Transformation for Multilingual Information Retrieval
SHIFT mitigates language bias in MLIR by subtracting estimated relative language vectors from document embeddings during indexing using parallel translation pairs.
-
Rethinking RAG in Long Videos: What to Retrieve and How to Use It?
Introduces V-RAGBench benchmark and CARVE method that selects per-chunk retrieval configurations via parallel retrievers and adaptive reranking, outperforming eight VideoRAG baselines.
-
Agentic Hybrid RAG for Evidence-Grounded Muon Collider Analysis
Agentic hybrid RAG with a new muon collider benchmark outperforms baselines in retrieval effectiveness, answer quality, evidence coverage, and factual grounding.
-
MARDoc: A Memory-Aware Refinement Agent Framework for Multimodal Long Document QA
MARDoc introduces a three-agent framework (Explorer, Refiner, Reflector) with dynamically updated structured memory to improve multi-hop reasoning in multimodal long-document QA, outperforming baselines on MMLongBench-Doc and DocBench.
-
HMARS: A Hierarchical Multi-Agent Memory System for Long-Context Reasoning
HMARS introduces a hierarchical multi-agent memory system that outperforms standard retrieval and other baselines on long-document and multi-turn reasoning tasks through improved evidence coverage.
-
Query Symbolically or Retrieve Semantically? A Dataset and Method for Semi-Structured Question Answering
DualGraph combines semantic textual KGs with symbolic KGs for semi-structured QA and introduces the SpecsQA benchmark, outperforming baselines on both open and specification questions.
-
EfficientGraph-RAG: Structured Retrieval-State Management for Cross-Task Retrieval-Augmented Generation
EfficientGraph-RAG structures retrieval state with TAM, MARS and SMP, ranking first on averaged LongBench answer-quality metrics while cutting token use 3.51x on HotpotQA.
-
GraphMind: From Operational Traces to Self-Evolving Workflow Automation
GraphMind builds and evolves action-centric workflow graphs from traces, navigates them via multi-agent LLM reasoning, and adapts via ATR, outperforming baselines on 93 incidents with 8x less context and 26% lower hallucination in production deployment.
-
Retrieval from Within: An Intrinsic Capability of Attention-Based Models
Attention-based models can retrieve evidence intrinsically by using decoder attention to score and reuse their own pre-encoded chunks, outperforming separate retrieval pipelines on QA benchmarks.
-
An Agentic Approach to Metadata Reasoning
Metadata Reasoner uses agentic LLM reasoning on metadata to select sufficient and minimal data sources, achieving 83.16% F1 on KramaBench and 85.5% F1 on noisy synthetic benchmarks while avoiding low-quality tables 99% of the time.
-
Agentic Graph Retrieval-Augmented Generation for Auditable Commercial Registry Analysis
An agentic GraphRAG system using intent routing, bounded reflection, and state-machine output control answers Swiss commercial-registry questions with 0.83 factual correctness versus 0.14 for a vector-RAG baseline.
-
ADAM: A Systematic Data Extraction Attack on Agent Memory via Adaptive Querying
ADAM extracts data from LLM agent memory with up to 100% attack success rate by estimating data distribution and selecting queries via entropy guidance.
-
Evo-Memory: Benchmarking LLM Agent Test-time Learning with Self-Evolving Memory
Evo-Memory is a new streaming benchmark and evaluation framework for self-evolving memory in LLM agents, unifying over ten memory modules and introducing the ReMem pipeline for continual improvement on multi-turn and reasoning datasets.
-
Think Big, Search Small: Where Capacity Matters in Hierarchical Search Agents?
Scaling the delegation backbone in hierarchical search agents improves EM by ~11 points while scaling the executor moves EM by only ~2.6 points, and a 1.7B SFT executor matches a frontier sub-agent at 37% fewer tokens.
-
A Methodology for Investigating AI Patterns Prevalence in Software Repositories
A methodology that mines 14 AI pattern classes from 44 sources and applies active learning to estimate prevalence of the most common class in 100 GitHub AI repositories, reporting 56% accuracy and 55% recall on an 8-way task.
-
KbSD: Knowledge Boundary aware Self-Distillation for Behavioral Calibration in Agentic Search
KbSD uses a same-size hint-augmented teacher and quadrant-adaptive KL objectives to deliver dense supervision for calibrated behavior across knowledge states in agentic search.
-
CAMI: Cost-Aware Agent-Guided Multi-Indexing for Semantic Retrieval
CAMI frames multi-index construction for semantic retrieval as a budgeted multi-objective portfolio problem and uses agent-guided search plus confidence-aware pruning to find high-recall configurations with reduced evaluation cost.
-
DynaTree: Dynamic Agentic Retrieval Tree for Time-Sensitive News Retrieval
DynaTree separates offline agentic tree construction from online subtree selection to deliver better recall, ranking, and production survival rates than standard or prior agentic RAG for news retrieval.
-
VikingMem: A Memory Base Management System for Stateful LLM-based Applications
VikingMem implements the Memory Base paradigm via event-centric extraction and entity updates on VikingDB with temporal compression, claiming up to 30% better retrieval effectiveness on long-term memory benchmarks.
-
AIBuildAI-2: A Knowledge-Enhanced Agent for Automatically Building AI Models
AIBuildAI-2 introduces a knowledge-enhanced agent with a hierarchical evolving external knowledge base that dynamically loads relevant AI development expertise, achieving first place on MLE-Bench at 70.7% medal rate.
-
SpecHop: Continuous Speculation for Accelerating Multi-Hop Retrieval Agents
SpecHop accelerates multi-hop LLM tool use via continuous multi-threaded speculation with asynchronous verification, approaching oracle latency gains and reducing latency up to 40% on retrieval tasks.
-
AMATA: Adaptive Multi-Agent Trajectory Alignment for Knowledge-Intensive Question Answering
AMATA is an adaptive multi-agent trajectory alignment system that improves factual consistency in knowledge-intensive QA via intra-trajectory preference learning and inter-agent dependency optimization.
-
Personalized Deep Research: A User-Centric Framework, Dataset, and Hybrid Evaluation for Knowledge Discovery
PDR is a user-context-aware framework for LLM research agents that improves report relevance over static baselines, supported by a new dataset and hybrid evaluation.
-
AgenticRAG: Agentic Retrieval for Enterprise Knowledge Bases
AgenticRAG equips an LLM with iterative retrieval and navigation tools, delivering 49.6% recall@1 on BRIGHT, 0.96 factuality on WixQA, and 92% correctness on FinanceBench.
-
Agentic Retrieval-Augmented Generation for Financial Document Question Answering
An agentic RAG system with a finance-tuned retriever, Program-of-Thought code execution, and adaptive routing reports 5.6–9.3 point gains on FinQA, ConvFinQA, and TAT-QA while cutting API cost ~41%.
-
When Agents Handle Secrets: A Survey of Confidential Computing for Agentic AI
A survey providing a taxonomy of TEE platforms, an agent-centric threat model, and open challenges for applying confidential computing to secure agentic AI systems.
-
SiriusHelper: An LLM Agent-Based Operations Assistant for Big Data Platforms
SiriusHelper deploys an LLM agent with intent routing, DeepSearch multi-hop retrieval, and automated SOP distillation to outperform alternatives and reduce ticket volume by 20.8% on Tencent's big data platform.
-
Can QPP Choose the Right Query Variant? Evaluating Query Variant Selection for RAG Pipelines
QPP methods can select query variants that boost end-to-end RAG quality over the original query, though retrieval-optimized variants often fail to produce the best generated answers, revealing a utility gap.
-
Mind DeepResearch Technical Report
MindDR combines a Planning Agent, DeepSearch Agent, and Report Agent with SFT cold-start, Search-RL, Report-RL, and preference alignment to reach competitive scores on research benchmarks using 30B-scale models.
-
ALDEN: Boosting Private Data Extraction from Retrieval-Augmented Generation Systems via Active Learning and Distribution Estimation
ALDEN boosts private data extraction rates from RAG systems by combining active learning for query diversification with dynamic estimation of the underlying knowledge-base topic distribution.