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.
@article{arxiv.2602.03652,
title = {RAGTurk: Best Practices for Retrieval Augmented Generation in Turkish},
author = {Süha Kağan Köse and Mehmet Can Baytekin and Burak Aktaş and Bilge Kaan Görür and Evren Ayberk Munis and Deniz Yılmaz and Muhammed Yusuf Kartal and Çağrı Toraman},
journal= {arXiv preprint arXiv:2602.03652},
year = {2026}
}