Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG (arXiv, Jan 2025)

• Paper: Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG
• Authors: Aditi Singh, Abul Ehtesham, Saket Kumar, Tala Talaei Khoei, Athanasios V. Vasilakos
• Type: Survey (RAG section of the community list)

One-line summary

A systematic survey of Agentic RAG: how planning, iterative retrieval, tool invocation, and reflection turn static retrieval-augmented generation into dynamic, agent-driven systems for LLMs.

Background and motivation

Traditional search pipelines separate retrieval, ranking, and generation, which struggles to meet LLM-era demands for natural-language interaction, multi-hop reasoning, and up-to-date knowledge. This survey addresses the core systems question: how to redistribute responsibilities across retrieval, ranking, generation, and tool calling—where the new variables are inference budget and action space (whether to retrieve, how many times, and which tools to call).

Core contributions

• A unified perspective that brings scattered related work into a comparable framework.
• Clear decomposition of method components: representation learning, retrievers, re-rankers, planners, generators, and feedback mechanisms.
• Reproducible benchmarks, datasets, and taxonomy tables that lower the entry barrier for follow-up research.
• Discussion of interfaces with LLM tool calling, reinforcement learning, and multi-agent collaboration, including paths from research prototypes to industrial systems.
• Explicit open problems: evaluation credibility, latency and cost, hallucination and safety, cross-lingual and multimodal extension.

Taxonomy

| Dimension | Subtype | Representative ideas | Strengths | Limitations | |---|---|---|---|---| | Modeling paradigm | Discriminative / generative retrieval | Dual-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 objective | Relevance / diversity / freshness | Multi-objective LTR, RLHF, online learning | Business-aligned | Scarce annotations | | Evaluation | Offline / Online / Human | nDCG, MRR, LLM-as-judge, A/B tests | Comparable | Diverges from real satisfaction |

Four research lines compared

• Dense retrieval: high recall, low latency; suited to first-stage retrieval.
• Late interaction (e.g., ColBERT): higher accuracy but larger indexes.
• Generative IR: directly "generates" documents via tokens or docids, simplifying the cascade.
• Agentic search: models search as sequential decision-making, supporting multi-hop queries and self-reflection.

Timeline of the field

• 2019–2021: BERT re-ranking and DPR establish neural retrieval.
• 2022–2023: RAG and FreshLLMs drive retrieval-generation fusion.
• 2024: conversational/agentic search and Gen-RecSys explode.
• 2025–2026: RL-trained search agents, Deep Research, and GraphRAG become growth frontiers.

Evaluation paradigms

Typical coverage includes datasets (MS MARCO, BEIR, Natural Questions, domain corpora), metrics (nDCG@10, MRR, Recall@k, task success rate, latency, token cost), baselines (BM25, dense retrieval, cross-encoder re-ranking, retrieval-free LLMs, commercial search APIs), and ablations on retrieval steps, re-ranking depth, and training data scale. Refer to the original PDF tables for exact numbers.

Key takeaways for search / rec / personalization

1. Architecture: cascade retrieve-rerank-generate remains mainstream, but the agentic paradigm makes retrieval count and strategy themselves learnable. 2. Data: high-quality instruction data and click/session logs matter equally; synthetic data must guard against leakage and distribution shift. 3. Evaluation: the offline-online gap is widening; LLM-as-judge needs cross-validation against human evaluation. 4. Product: latency, cost, explainability, and safety are hard industrial constraints—do not optimize academic benchmarks alone.

Open problems and future directions

Authors and the community point to: lack of unified benchmarks, non-reproducible private data, LLM evaluation bias, and safety/cost constraints of agentic systems. Future work includes finer-grained process supervision, joint retrieval-reasoning training, enterprise metadata governance, and multilingual/multimodal consistency.

Engineering checklist

| Check | Question | Suggestion | |---|---|---| | Data | PII in training/index? Version management? | Partitioned indexes, sanitization, rollback-able embedding versions | | Latency | p99 budget? How many retrieval steps? | Cascade + early stopping, hot-query caching, async re-ranking | | Quality | Do offline gains transfer to online satisfaction? | Interleaving experiments, human audits, citation checks | | Safety | Does open retrieval introduce poisoning/bias? | Source whitelists, adversarial detection, output filtering | | Cost | Token and GPU cost per query? | Small-model routing, distillation, hybrid sparse+dense |

Related entries

• A Survey of Graph Retrieval-Augmented Generation for Customized LLMs (arXiv:2501.13958)
• A Survey on Retrieval-Augmented Text Generation for LLMs (arXiv:2404.10981)
• RAFT: Adapting Language Models to Domain-Specific RAG (Jul 2024)
• RAG vs. GraphRAG: A Systematic Evaluation and Key Insights (arXiv:2502.11371)
• Algolia: Knowledge graphs and ontologies for keyword search
• AutoKnow: self-driving knowledge collection for product taxonomy

Glossary

| Term | Meaning | |---|---| | IR | Information Retrieval | | RAG | Retrieval-Augmented Generation | | LTR | Learning to Rank | | nDCG | Normalized Discounted Cumulative Gain | | Agentic Search | Modeling search as sequential decision-making and tool invocation | | Gen-IR | Generative Information Retrieval |