Adaptive-RAG: Learning to Adapt Retrieval-Augmented Large Language Models through Question Complexity

By Soyeong Jeong, Jinheon Baek, Sukmin Cho, Sung Ju Hwang, Jong C. Park. North American Chapter of the Association for Computational Linguistics, 2024

562 citations recorded by the source corpus.

Retrieval-Augmented Large Language Models (LLMs), which incorporate the non-parametric knowledge from external knowledge bases into LLMs, have emerged as a promising approach to enhancing response accuracy in several tasks, such as Question-Answering (QA). However, even though there are various approaches dealing with queries of different complexities, they either handle simple queries with unnecessary computational overhead or fail to adequately address complex multi-step queries; yet, not all user requests fall into only one of the simple or complex categories. In this work, we propose a novel adaptive QA framework that can dynamically select the most suitable strategy for (retrieval-augmented) LLMs from the simplest to the most sophisticated ones based on the query complexity. Also, this selection process is operationalized with a classifier, which is a smaller LM trained to predict the complexity level of incoming queries with automatically collected labels, obtained from actual predicted outcomes of models and inherent inductive biases in datasets. This approach offers a balanced strategy, seamlessly adapting between the iterative and single-step retrieval-augmented LLMs, as well as the no-retrieval methods, in response to a range of query complexities. We validate our model on a set of open-domain QA datasets, covering multiple query complexities, and show that ours enhances the overall efficiency and accuracy of QA systems, compared to relevant baselines including the adaptive retrieval approaches. Code is available at: https://github.com/starsuzi/Adaptive-RAG.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/e5e8c6ac537e0f5b5db14170bc232d6f9e641bbc

Citations in the corpus

• Retrieval-Augmented Generation for AI-Generated Content: A Survey (2024, unknown): “AdaptiveRAG [217] dynamically decides whether to retrieve based on the query complexity by a classifier, which is a smaller LM. b) Iterative RAG : Iterative RAG progressively refines results by repeatedly cycling through retrieval and generation phases, rather than a single round.”
• Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG (2025, unknown): “Adaptive Retrieval-Augmented Generation (Adaptive RAG) [35] enhances the flexibility and efficiency of large language models (LLMs) by dynamically adjusting query handling strategies based on the complexity of the incoming query.”
• Min-Cost Flow Routing for Evidence Assembly in Long Multimodal Documents (2026, unknown): “Adaptive methods adjust retrieval depth or query decomposition to query complexity (Jeong et al., 2024; Yu et al., 2024; Nguyen et al., 2025), while dual-process designs combine fast and reflective stages (Wang et al., 2024b; Cheng et al., 2025; Hu et al., 2026).”
• RASER: Recoverability-Aware Selective Escalation Router for Multi-Hop Question Answering (2026, unknown): “This framing is related to prior adaptive retrieval and routing work (Jeong et al., 2024; Guo et al., 2026; Liu et al., 2026), but differs in two important respects.”
• Malgta: large language model-based guided malware tactical analysis (2025, unknown): “Soyeong Jeong et al. [26] proposed an adaptive retrieval framework that categorizes queries into direct, simple, and complex questions, determining the number of RAG accesses based on the query type.”
• ArchRAG: Attributed Community-based Hierarchical Retrieval-Augmented Generation (2025, unknown): “RAG has been proven to excel in many tasks, including open-ended question answering [28, 57], programming context [9, 10], SQL rewrite [38, 59], and data cleaning [46, 47, 51].”