Toward Agentic RAG for Ukrainian
Incremental exploration of agentic RAG for a low-resource language (Ukrainian), highlighting retrieval as a bottleneck.
The paper investigates Agentic RAG for Ukrainian, combining two-stage retrieval with a lightweight agentic layer, finding that retrieval quality is the primary bottleneck and agentic retry improves accuracy but overall performance is constrained by document identification.
We present an initial investigation into Agentic Retrieval-Augmented Generation (RAG) for Ukrainian, conducted within the UNLP 2026 Shared Task on Multi-Domain Document Understanding. Our system combines two-stage retrieval (BGE-M3 with BGE reranking) with a lightweight agentic layer performing query rephrasing and answer-retry loops on top of Qwen2.5-3B-Instruct. Our analysis reveals that retrieval quality is the primary bottleneck: agentic retry mechanisms improve answer accuracy but the overall score remains constrained by document and page identification. We discuss practical limitations of offline agentic pipelines and outline directions for combining stronger retrieval with more advanced agentic reasoning for Ukrainian.