DynaRAG: Bridging Static and Dynamic Knowledge in Retrieval-Augmented Generation
This addresses the need for reliable, real-world question-answering systems by bridging static and dynamic knowledge, though it is incremental as it builds on existing RAG and API-calling methods.
The paper tackles the problem of handling both static and time-sensitive information in retrieval-augmented generation by introducing DynaRAG, which selectively uses external APIs when retrieved documents are insufficient, resulting in improved accuracy on dynamic questions and reduced hallucinations as demonstrated on the CRAG benchmark.
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.