DynaRAG: Bridging Static and Dynamic Knowledge in Retrieval-Augmented Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liang, Penghao, Yuan, Mengwei, Liu, Jianan, Yang, Jing, Li, Xianyou, Yan, Weiran, Wu, Yichao
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914408212463616
author Liang, Penghao
Yuan, Mengwei
Liu, Jianan
Yang, Jing
Li, Xianyou
Yan, Weiran
Wu, Yichao
author_facet Liang, Penghao
Yuan, Mengwei
Liu, Jianan
Yang, Jing
Li, Xianyou
Yan, Weiran
Wu, Yichao
contents We present DynaRAG, a retrieval-augmented generation (RAG) framework designed to handle both static and time-sensitive information needs through dynamic knowledge integration. Unlike traditional RAG pipelines that rely solely on static corpora, DynaRAG selectively invokes external APIs when retrieved documents are insufficient for answering a query. The system employs an LLM-based reranker to assess document relevance, a sufficiency classifier to determine when fallback is necessary, and Gorilla v2 -- a state-of-the-art API calling model -- for accurate tool invocation. We further enhance robustness by incorporating schema filtering via FAISS to guide API selection. Evaluations on the CRAG benchmark demonstrate that DynaRAG significantly improves accuracy on dynamic questions, while also reducing hallucinations. Our results highlight the importance of dynamic-aware routing and selective tool use in building reliable, real-world question-answering systems.
format Preprint
id arxiv_https___arxiv_org_abs_2603_18012
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle DynaRAG: Bridging Static and Dynamic Knowledge in Retrieval-Augmented Generation
Liang, Penghao
Yuan, Mengwei
Liu, Jianan
Yang, Jing
Li, Xianyou
Yan, Weiran
Wu, Yichao
Computation and Language
Artificial Intelligence
Information Retrieval
We present DynaRAG, a retrieval-augmented generation (RAG) framework designed to handle both static and time-sensitive information needs through dynamic knowledge integration. Unlike traditional RAG pipelines that rely solely on static corpora, DynaRAG selectively invokes external APIs when retrieved documents are insufficient for answering a query. The system employs an LLM-based reranker to assess document relevance, a sufficiency classifier to determine when fallback is necessary, and Gorilla v2 -- a state-of-the-art API calling model -- for accurate tool invocation. We further enhance robustness by incorporating schema filtering via FAISS to guide API selection. Evaluations on the CRAG benchmark demonstrate that DynaRAG significantly improves accuracy on dynamic questions, while also reducing hallucinations. Our results highlight the importance of dynamic-aware routing and selective tool use in building reliable, real-world question-answering systems.
title DynaRAG: Bridging Static and Dynamic Knowledge in Retrieval-Augmented Generation
topic Computation and Language
Artificial Intelligence
Information Retrieval
url https://arxiv.org/abs/2603.18012