Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG (arXiv 2501.09136, Jan 2025)
Meta information
| Field | Content | |-------|---------| | Title | Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG | | Authors | Aditi Singh, Abul Ehtesham, Saket Kumar, Tala Talaei Khoei, Athanasios V. Vasilakos | | Published | January 2025 | | Source | https://arxiv.org/abs/2501.09136 | | Type | Survey | | Section | RAG |
One-sentence summary
This survey systematically reviews Agentic RAG, reframing retrieval-augmented generation as an agentic, iterative, planning-driven process in which LLM agents decide when, how, and how often to retrieve, reason over sources, and verify outputs.
Background and motivation
In large-scale search, recommendation, and personalization systems, agentic search faces long-standing challenges around efficiency, scalability, and intent understanding. Traditional pipeline approaches treat retrieval, ranking, and generation as disjoint stages, which struggles to meet LLM-era demands for natural-language interaction, multi-hop reasoning, and real-time knowledge. This survey situates itself at this intersection, covering open-domain information access, enterprise knowledge retrieval, conversational search, semantic understanding in recommendation, and end-to-end architectures that couple external knowledge sources with generative models.
Core contributions
- Provides a unified perspective that organizes scattered related work into a comparable framework.
- Decomposes method components (representation learning, retriever, reranker, planner, generator, feedback mechanisms) to ease engineering adoption.
- Offers reproducible benchmarks, datasets, and taxonomy tables, lowering entry barriers for researchers.
- Discusses interfaces with LLM tool calling, reinforcement learning, and multi-agent collaboration, pointing out migration paths from research prototypes to industrial systems.
- Identifies open problems: evaluation trustworthiness, latency and cost, hallucination and safety, and multilingual/multimodal extension.
- Dense retrieval: high recall, low latency, suited to first-stage retrieval;
- Late interaction (e.g., ColBERT): higher precision but larger indexes;
- Generative IR: directly "generates" documents via tokens or docids, simplifying cascades;
- Agentic search: models search as sequential decision-making, supporting multi-hop reasoning and self-reflection.
- 2019–2021: BERT reranking and DPR lay the foundations of neural retrieval.
- 2022–2023: RAG and FreshLLM drive retrieval-generation fusion.
- 2024: Conversational/agentic search and Gen-RecSys surge.
- 2025–2026: RL-trained search agents, Deep Research, and GraphRAG become new growth fronts.
- Datasets: MS MARCO, BEIR, Natural Questions, domain corpora, public recommendation sets.
- Metrics: nDCG@10, MRR, Recall@k, Hit@k, human preference, task success rate, latency, token cost.
- Baselines: BM25, dense retrieval, cross-encoder reranking, retrieval-free LLMs, commercial search APIs.
- Ablations: contribution of retrieval steps, reranking depth, and training data scale.
- A Survey of Graph Retrieval-Augmented Generation for Customized Large Language Models (arXiv 2501.13958)
- A Survey on Retrieval-Augmented Text Generation for Large Language Models (arXiv 2404.10981)
- RAFT: Adapting Language Model to Domain Specific RAG (Jul 2024)
- RAG vs. GraphRAG: A Systematic Evaluation and Key Insights (arXiv 2502.11371)
- Original paper: Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG. https://arxiv.org/abs/2501.09136
Taxonomy and research threads
| Dimension | Subclass | Representative ideas | Strengths | Limitations | |-----------|----------|----------------------|-----------|-------------| | Modeling paradigm | Discriminative / generative retrieval | Bi-encoder, cross-encoder, DSI, GPT indexing | Mature, scalable | Semantic drift, update cost | | LLM integration | RAG / Agent / Tool-use | Retrieval augmentation, search agents, API calls | Flexible, interpretable | Latency, error propagation | | Optimization goal | Relevance / diversity / freshness | Multi-objective LTR, RLHF, online learning | Business-aligned | Annotation scarcity | | Evaluation | Offline / online / human | nDCG, MRR, LLM-as-judge, A/B testing | Comparable | Deviates from real satisfaction |
Surveys in this space typically contrast four main research threads:
Timeline of research evolution
Evaluation paradigms
Typical benchmarks and metrics covered:
Quantitative results should be verified against the original PDF tables.
Key takeaways for search / recommendation / personalization
1. Architecture: Cascade retrieval + rerank + generate remains mainstream, but the agentic paradigm makes "retrieval count and strategy" itself learnable. 2. Data: High-quality instruction data and click/session logs are both critical; synthetic data must guard against leakage and distribution shift. 3. Evaluation: The gap between offline metrics and online satisfaction is widening; LLM-as-judge needs cross-validation with human evaluation. 4. Product: Latency, cost, explainability, and safety are hard industrial constraints—do not optimize academic benchmarks alone.
Open problems and future directions
Open issues include insufficient unified benchmarks, non-reproducible private data, LLM evaluation bias, and the safety/cost constraints of agentic systems. Future work: finer-grained process supervision, joint retrieval-reasoning training, enterprise metadata governance, and multimodal/cross-lingual consistency. Limitations may include GPU-budget-limited experiments, benchmark-user distribution mismatch, English-centric data limiting cross-lingual generalization, and safety risks of open-web agentic systems.
Engineering checklist
| Item | Question | Suggestion | |------|----------|------------| | Data | Does training/indexing contain PII? How are versions managed? | Partitioned indexes, anonymization, rollback-capable embedding versions | | Latency | What is the p99 budget? How many retrieval steps? | Cascade + early stopping, hot-query caching, async reranking | | Quality | Do offline gains translate to online CTR/satisfaction? | Interleaving experiments, human audits, citation verification | | Safety | Does open retrieval introduce poisoning/bias? | Source whitelisting, adversarial detection, output filtering | | Cost | Per-query token and GPU cost? | Small-model routing, distillation, hybrid sparse+dense retrieval |
Glossary
| Term | Meaning | |------|---------| | IR | Information Retrieval | | RAG | Retrieval-Augmented Generation | | LTR | Learning to Rank | | nDCG | Normalized Discounted Cumulative Gain | | Agentic Search | Treating search as sequential decision-making and tool calling by agents | | Gen-IR | Generative Information Retrieval |