RAGTurk: Best Practices for Retrieval Augmented Generation in Turkish
This addresses the problem of English-centric RAG design for Turkish NLP practitioners, though it is incremental as it adapts existing methods to a new language.
The authors tackled the lack of retrieval-augmented generation (RAG) guidance for morphologically rich languages by creating a Turkish RAG dataset and benchmarking pipeline stages, finding that complex methods like HyDE achieved 85% accuracy (vs. 78.70% baseline) while Pareto-optimal configurations offered comparable performance with lower cost.
Retrieval-Augmented Generation (RAG) enhances LLM factuality, yet design guidance remains English-centric, limiting insights for morphologically rich languages like Turkish. We address this by constructing a comprehensive Turkish RAG dataset derived from Turkish Wikipedia and CulturaX, comprising question-answer pairs and relevant passage chunks. We benchmark seven stages of the RAG pipeline, from query transformation and reranking to answer refinement, without task-specific fine-tuning. Our results show that complex methods like HyDE maximize accuracy (85%) that is considerably higher than the baseline (78.70%). Also a Pareto-optimal configuration using Cross-encoder Reranking and Context Augmentation achieves comparable performance (84.60%) with much lower cost. We further demonstrate that over-stacking generative modules can degrade performance by distorting morphological cues, whereas simple query clarification with robust reranking offers an effective solution.