agentic RAG
In plain wordsRetrieval in which the AI runs its own research: it decides what to search for, looks at the results, searches again if needed, and only then answers.
A form of retrieval-augmented generationretrieval-augmented generation, n.: Giving an AI an open-book exam: before answering, it looks up relevant passages in a set of documents and bases its reply on them, often citing where each point came from. in which retrieval is directed by an agentAI agent, n.: An AI system that does not just answer but acts: given a goal, it decides on steps, uses tools such as a browser or code editor, checks the results and keeps going until the job is done. rather than a fixed retrieve-then-generate pipeline. The model decides whether retrieval is needed, formulates and reformulates queries (query rewritingquery rewriting, n.: Rephrasing a user's question before searching — fixing typos, adding missing details, splitting it into simpler questions — so the search engine finds better results.), chooses among tools and sources (vector index, keyword search, web search, SQL, APIs), reads results, judges whether they suffice, follows leads across several hops, and stops when it can answer, often with citationscitation, n.: A pointer from something an AI says to the source that supports it — a link, document or quoted passage — so people can check the claim for themselves.. Implementations range from a single model in an agent loopagent loop, n.: The repeating cycle at the heart of an AI agent: the model decides what to do next, software carries it out, the result is shown to the model, and the cycle repeats until the task is finished. with search tools, as in ReActReAct, prop. n.: A method in which an AI alternates between thinking out loud about what to do next and taking an action, such as a web search, then uses what it observed to plan its next step., to routers and multi-agent set-ups with planners, retrievers and verifiers; models can also be trained with reinforcement learningreinforcement learning, n.: A way of teaching a computer program by trial and error: it tries actions, earns points for good results and penalties for bad ones, and gradually learns which choices pay off over time. to search well. Agentic RAG handles multi-step and ambiguous questions better than one-shot retrieval, at the cost of more latency, more tokens and more opportunities for error; deep researchdeep research, n.: An AI feature that spends minutes rather than seconds on a question — planning, running dozens of web searches, reading many sources — and then writes a long report with citations. systems are a long-horizon version.
Agentic RAG let the assistant notice that its first search missed the 2025 filings and run a narrower second query.
Origin
Interleaving reasoning with retrieval was explored in 2022–2023 (Self-Ask, ReAct, Demonstrate-Search-Predict, Self-RAG), and Search-R1 (2025) trained models to search with reinforcement learning. The label 'agentic RAG' spread in 2024 through framework and vector-database tutorials; a January 2025 survey by Aditi Singh and colleagues helped settle its definition.
Related
- retrieval-augmented generationretrieval-augmented generation, n.: Giving an AI an open-book exam: before answering, it looks up relevant passages in a set of documents and bases its reply on them, often citing where each point came from.
- AI agentAI agent, n.: An AI system that does not just answer but acts: given a goal, it decides on steps, uses tools such as a browser or code editor, checks the results and keeps going until the job is done.
- tool usetool use, n.: An AI model's ability to reach beyond its own knowledge by operating outside tools — searching the web, running code, using a calculator or an app — and folding the results into its work.
- query rewritingquery rewriting, n.: Rephrasing a user's question before searching — fixing typos, adding missing details, splitting it into simpler questions — so the search engine finds better results.
- deep researchdeep research, n.: An AI feature that spends minutes rather than seconds on a question — planning, running dozens of web searches, reading many sources — and then writes a long report with citations.
- just-in-time retrievaljust-in-time retrieval, n.: Letting an AI agent fetch information only at the moment it needs it — opening a file, running a search — instead of stuffing everything into its prompt in advance.
- ReActReAct, prop. n.: A method in which an AI alternates between thinking out loud about what to do next and taking an action, such as a web search, then uses what it observed to plan its next step.
- retrieval-augmented generationretrieval-augmented generation, n.: Giving an AI an open-book exam: before answering, it looks up relevant passages in a set of documents and bases its reply on them, often citing where each point came from.
Sources
- Singh, A., Ehtesham, A., Kumar, S. & Talaei Khoei, T. (2025). Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG. arXiv:2501.09136. arxiv.org/abs/2501.09136
- Jin, B., Zeng, H., Yue, Z., Yoon, J., Arik, S., Wang, D., Zamani, H. & Han, J. (2025). Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning. arXiv:2503.09516. arxiv.org/abs/2503.09516
- Khattab, O., Santhanam, K., Li, X. L., Hall, D., Liang, P., Potts, C. & Zaharia, M. (2022). Demonstrate-Search-Predict: Composing retrieval and language models for knowledge-intensive NLP. arXiv:2212.14024. arxiv.org/abs/2212.14024
Mentioned in
deep researchdeep research, n.: An AI feature that spends minutes rather than seconds on a question — planning, running dozens of web searches, reading many sources — and then writes a long report with citations., retrieval-augmented generationretrieval-augmented generation, n.: Giving an AI an open-book exam: before answering, it looks up relevant passages in a set of documents and bases its reply on them, often citing where each point came from., citationcitation, n.: A pointer from something an AI says to the source that supports it — a link, document or quoted passage — so people can check the claim for themselves., GraphRAGGraphRAG, prop. n.: A way of letting AI answer questions about a big pile of documents by first turning them into a map of people, things and their connections, then using that map to find and summarize answers., query rewritingquery rewriting, n.: Rephrasing a user's question before searching — fixing typos, adding missing details, splitting it into simpler questions — so the search engine finds better results., just-in-time retrievaljust-in-time retrieval, n.: Letting an AI agent fetch information only at the moment it needs it — opening a file, running a search — instead of stuffing everything into its prompt in advance.